Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Untitled

Download as pdf or txt
Download as pdf or txt
You are on page 1of 353

╇ i

SPECULATION
Q
ii
SPECULATION
Q
A HISTORY OF THE FINE LINE
BETWEEN GAMBLING
AND INVESTING

S TUART  B ANNER

1
iv

1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.

Published in the United States of America by Oxford University Press


198 Madison Avenue, New York, NY 10016, United States of America.

© Oxford University Press 2017

All rights reserved. No part of this publication may be reproduced, stored in


a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.

You must not circulate this work in any other form


and you must impose this same condition on any acquirer.

Library of Congress Cataloging-​in-​Publication Data


Names: Banner, Stuart, 1963– author.
Title: Speculation : a history of the fine line between gambling and investing / Stuart Banner.
Description: New York, NY : Oxford University Press, 2017. | Description based on print version
record and CIP data provided by publisher; resource not viewed.
Identifiers: LCCN 2016017786 (print) | LCCN 2016016710 (ebook) | ISBN 9780190623050
(E-book) | ISBN 9780190623067 (E-book) | ISBN 9780190623043 (hardcover : alk. paper)
Subjects: LCSH: Investments—Law and legislation—United States—History. |
Capital market—Law and legislation—United States—History. | Speculation—Law and
legislation—United States—History. | Speculation—United States—History.
Classification: LCC KF1070 (print) | LCC KF1070 .B365 2017 (ebook) |
DDC 346.73/0922—dc23 LC record available at https://lccn.loc.gov/2016017786

1 3 5 7 9 8 6 4 2
Printed by Sheridan Books, Inc., United States of America
C ON T E N T S

q
Acknowledgmentsâ•… vii
Abbreviationsâ•… ix

Introductionâ•… 1
1.╇ The Land of Speculation╅ 6
2.╇ Betting on Prices╅ 56
3.╇ The Anti-╉Option Eraâ•… 105
4.╇ Selling Blue Skyâ•… 138
5.╇ Aftershocks of the Crash╅ 164
6.╇ Land and Onions╅ 218
7.╇ Inside Information╅ 241
8.╇ Speculation or Investment?╅ 279
9.╇ Deregulation and Crisis?╅ 307

Indexâ•… 331

v
vi
AC K NOW L E D G M E N T S

q
For advice on drafts of chapters, thanks to Iman Anabtawi, Steve
Bainbridge, Steve Bank, James Park, and participants in workshops at
the University of Minnesota, UNLV, and UCLA. For help with the re-
search, thanks to the librarians at the UCLA School of Law and the ar-
chivists at the Franklin D. Roosevelt Presidential Library, the Harvard
Law Library, the Herbert Hoover Presidential Library, the John
F. Kennedy Presidential Library, the Library of Congress, the Morgan
Library, the New  York Stock Exchange, the University of Illinois at
Chicago, the University of Virginia, the Wisconsin Historical Society,
and Yale University.

vii
viii
A BBR E V I AT IONS

q
AAB Adolf A. Berle papers, Franklin D. Roosevelt Presidential
Library, Hyde Park, New York
AS Alexander Sachs papers, Franklin D. Roosevelt Presidential
Library, Hyde Park, New York
CBT Chicago Board of Trade records, Special Collections,
University of Illinois at Chicago
CG Carter Glass papers, Albert and Shirley Small Special
Collections Library, University of Virginia
FDR Franklin D. Roosevelt papers as president, Franklin
D. Roosevelt Presidential Library, Hyde Park, New York
FF Felix Frankfurter papers, Manuscript Division, Library of
Congress, Washington, DC
GEA George E. Akerson papers, Herbert Hoover Presidential
Library, West Branch, Iowa
HHL Hoover Presidential files, Herbert Hoover Presidential
Library, West Branch, Iowa
HPPF Hoover Post-​Presidential files, Herbert Hoover Presidential
Library, West Branch, Iowa
JML-​HLS James McCauley Landis papers, Harvard Law School
Library, Cambridge, Massachusetts
JML-​LC James McCauley Landis papers, Manuscript Division,
Library of Congress, Washington, DC

ix
x

x Abbreviations

JNF Jerome New Frank papers, Sterling Memorial Library, Yale


University, New Haven, Connecticut
JPK Joseph P. Kennedy personal papers, John F. Kennedy
Presidential Library, Boston, Massachusetts
JPL Joseph P. Lash papers, Franklin D. Roosevelt Presidential
Library, Hyde Park, New York
JPM J. P. Morgan Jr. papers, Morgan Library, New York
NYSE New York Stock Exchange Archives, New York
WDS Wisconsin Department of Securities: Investigation and
Enforcement files, 1923–​62, Wisconsin Historical Society,
Madison, Wisconsin
WFV William F. Vilas papers, Wisconsin Historical Society, Madison,
Wisconsin
WOD William O. Douglas papers, Manuscript Division, Library of
Congress, Washington, DC
  xi

SPECULATION
Q
xii
Introduction

“When as a young and unknown man I started to be successful I was


referred to as a gambler,” the banker Ernest Cassel once said. “My op-
erations increased in scope. Then I was a speculator. The sphere of my
activities continued to expand and presently I was known as a banker.
Actually I had been doing the same thing all the time.”1
Cassel was retelling an old joke, but behind it lurked a dilemma that has
troubled the legal system for a long time, a problem that is the subject of
this book. For centuries there has been a consensus that investment is very
useful and ought to be encouraged. There has also been a near consen-
sus, one that has weakened recently but is still substantial, that gambling
is dangerous and ought to be at least discouraged and maybe even pro-
hibited. But what about speculation? Speculation lies somewhere between
investment and gambling. It has attributes of both. Should speculation
be legal? Should it be illegal? Should some kinds be legal and other kinds
illegal? Which kinds? Throughout American history (and the history of
other places too, but this book is just about the United States), specula-
tion has presented a puzzle to the legal system. To figure out how to treat
speculation, we have always needed to distinguish between two kinds of
risky commerce: a good kind the law should promote and a bad kind the
law should deter. We have always needed to draw a line between invest-
ment and gambling. This book is a history of our efforts to draw this line.

1 The quotation was attributed to Cassel by Bernard Baruch. Carter Field,


Bernard Baruch: Park Bench Statesman (New York: McGraw-​Hill, 1944), 76–​77.

1
2

2 Speculation

The issue is particularly salient today, when many argue that the
recent financial crisis was caused by overspeculation, and that the way
to prevent a recurrence is to prohibit certain risky transactions. Others
respond by citing the benefits of robust markets and the dangers of
overregulation. This book shows that this debate has been a perennial
feature of our history, and that many of the same arguments have been
made, on both sides, after every financial downturn since the 1790s.
Ambivalence about speculation can be found in the slipperiness of
the word itself, which is notoriously hard to pin down.2 Americans have
offered competing definitions for two centuries, without coming any
closer to agreement. Sometimes “speculation” is a pejorative term, im-
plying that a speculator is someone engaging in unsavory conduct, as
distinguished from an investor or a businessperson. Sometimes it is used
in a morally neutral sense. And sometimes “speculation” is a term used
with approval, often to distinguish the speculator from the gambler.
The difficulty in distinguishing the good risky transactions from the
bad, and thus the difficulty in figuring out how to regulate speculators,
has been caused by genuine substantive disagreement about the pros
and cons of particular sorts of transactions. Should people be allowed
to risk their fortunes and the welfare of their families by buying assets
solely for the purpose of selling them later at a higher price? Should
they be allowed to sell assets they don’t yet own, in the hopes of prof-
iting from a decline in the assets’ value? Should they be allowed to
buy and sell complex financial derivatives? Questions like these have
constantly been asked throughout American history. The conflicting
answers have been driven in part by differences in opinion about the
economic consequences of particular transactions and in part by differ-
ences in moral views as to the propriety of particular forms of conduct.

2 Reuven Brenner with Gabrielle A. Brenner, Gambling and Speculation: A Theory,


a History, and a Future of Some Human Decisions (Cambridge:  Cambridge
University Press, 1990), 90–​112; Christine Hurt, “Regulating Public Morals
and Private Markets:  Online Securities Trading, Internet Gambling, and the
Speculation Paradox,” Boston University Law Review 86 (2006): 371–​441; Thomas
Lee Hazen, “Disparate Regulatory Schemes for Parallel Activities:  Securities
Regulation, Derivatives Regulation, Gambling, and Insurance,” Annual Review
of Banking & Financial Law 24 (2005):  375–​441; Shaheen Borna and James
Lowry, “Gambling and Speculation,” Journal of Business Ethics 6 (1987): 219–​24.
  3

Introduction 3

In economic terms, speculators have always presented the legal


system with a dilemma, because the very things that make speculation
useful can also make it dangerous. Sometimes speculators are like insur-
ance companies: They make money by taking on risks others wish to
shed. Sometimes speculators make markets more liquid, which makes
markets more useful for people wishing to buy or sell. Sometimes
speculators make markets more stable by selling when prices are high
and buying when they are low. These benefits of speculation counsel
in favor of not regulating speculators too strictly. On the other hand,
speculators sometimes bankrupt themselves, and sometimes their fi-
nancial distress hurts others, including their families and their credi-
tors. Speculation can even imperil the entire financial system. These
dangers counsel in favor of strict regulation. This is the dilemma. The
more we give speculators free rein, the greater the risk that speculators
will bring harm to others. But the more we constrain speculators, the
smaller the benefits of speculation.
For two centuries, Americans have tried to come up with forms of
regulation that will give us the good aspects of speculation without
the bad. We have tried to prohibit certain transactions while allowing
others. We have tried to keep particular speculators on a tight leash
while giving more freedom to others. But no solution has ever been
uncontroversial or permanent. The economic dilemma has never gone
away, and probably never will, because it is a real tradeoff between posi-
tive and negative consequences of the same activity.
Speculation has presented the legal system with a moral dilemma
as well, one that is also likely to be with us forever. Speculation has
long been decried as immoral—​a method of profiting by impoverish-
ing others, a device for making money without performing any useful
service, and a scheme rigged by crafty insiders to exploit the ordinary
person. Speculation has been defended against such attacks for just as
long. This moral debate is connected to the economic debate, because
some of the moral critique of speculation has always rested on the belief
that speculators are not providing value to society in the same way that
producers are. But the debates are different and always have been. One
can disapprove of speculators on moral grounds even if one accepts that
speculators are sometimes useful, and one can refrain from moral disap-
proval even if one thinks that speculators help no one but themselves.
4

4 Speculation

The moral dilemma, like the economic dilemma, is how to prohibit


the bad without also prohibiting the good. The most ardent moral crit-
ics of speculation have never wanted to prohibit investment. The trou-
ble is in distinguishing the two, because otherwise anything we do to
reduce the quantity of one will reduce the quantity of the other. There
is again a real tradeoff between the positive and negative attributes of a
single activity.
Of course, attitudes toward speculation have been influenced as
much by short-​term contemporary events as by long-​term basic be-
liefs. Regulation has followed a familiar pattern: New restrictions are
virtually always imposed after market downturns, while existing restric-
tions tend to be relaxed, at least in practice, while the market is on the
rise.3 When people are making money, few elected officials are eager
to stop the party, and when people who have lost money are looking
around for someone to blame, speculators are always a tempting target.
Debates about speculation tend to flare up at extremes, during booms
and busts, when the benefits and perils of speculation are most visible.
But the attitudes those debates elicit are never absent, and in some ways
they have been remarkably consistent over two centuries.
Conventional thought on these questions has changed considerably
over the past two centuries, in some respects. In the late eighteenth
century, courts were reluctant to enforce commercial contracts that
seemed too risky, but they nevertheless enforced bets on horse races.
A  half century later, the courts had switched positions on both. In
other respects, however, conventional thought has scarcely changed at
all. Consider these reactions to the financial crisis: The crisis was caused
by overspeculation; the victims included small investors lured by un-
scrupulous speculators with promises of high returns; and the federal
government should have intervened before prices rose too high. These
are familiar sentiments, but these particular samples are reactions to the
financial crisis of 1792, from Alexander Hamilton, Thomas Jefferson,

3 Erik F. Gerding, Law, Bubbles, and Financial Regulation (London: Routledge,


2014); Nolan McCarty, Keith T.  Poole, and Howard Rosenthal, Political
Bubbles:  Financial Crises and the Failure of American Democracy (Princeton,
N.J.: Princeton University Press, 2013).
  5

Introduction 5

and a correspondent to John Adams, respectively.4 Similar views have


been expressed after every market downturn up to the present.
Debates over speculation have always been closely related to many
other issues, but because this book maintains a tight focus on the dif-
ficulty of drawing a line between good risks and bad, I want to be as
clear as possible as to what the book is not about. It is not a history of
all commercial regulation or all regulation of financial markets. A great
deal of commercial and financial regulation concerns (and has always
concerned) other matters, like preventing fraud, but this book has little
to say about them. This is not a book that adds to the already-​enormous
literature about what causes financial crises or how to prevent them.
Worries about financial crises have played a big role in forming at-
titudes toward speculation, but this book is about the attitudes, not
the crises themselves. Nor is this book a history of speculation or of
speculators. There are already many books about celebrated speculators
and speculative episodes, and I am interested in them here only to the
extent that they had some bearing on how the legal system has tried to
distinguish investing from gambling.
So what is the difference between investing and gambling? Where
exactly is the line that separates good speculation from bad speculation?
The point of this book is that these are questions over which people
have always disagreed, and are likely always to disagree, because at all
times people have held different views about the moral and economic
consequences of risky commerce. If you have an opinion on these ques-
tions, you’re likely to find that many people expressed your opinion
long before you were born, and that many of their contemporaries
thought your opinion was nonsense. A good place to start is with one
prominent critic of speculators, Abraham Lincoln.

4 Alexander Hamilton to William Seton, 4 Apr. 1792, Harold C.  Syrett et  al.,
eds., The Papers of Alexander Hamilton (New York: Columbia University Press,
1961–​87), 11:225–​26; Thomas Jefferson to David Humphreys, 9 Apr. 1792, Julian
P. Boyd et al., eds., The Papers of Thomas Jefferson (Princeton, N.J.: Princeton
University Press, 1950–​), 23:386–​87; Nathaniel Hazard to John Adams, 16 Apr.
1792, Founders Online, National Archives (founders.archives.gov/​documents/​
Adams/​99-​02-​02-​1337).
1
Q
The Land of Speculation

Abraham Lincoln blamed the speculators.


In 1862, under the fiscal strain of the Civil War, the government had
gone off the gold standard. The “greenback,” the Union paper currency,
had previously been fixed at one paper dollar per gold dollar, but once
the greenback floated against gold, its value plummeted. By the summer
of 1864, a dollar of gold was trading for nearly $2.50 in greenbacks. The
Union currency had lost more than half its value. Because the govern-
ment paid for the war in greenbacks, a weaker currency meant a weaker
military. If speculators were intentionally driving down the greenback,
that would be an act tantamount to sabotage.1
Andrew Curtin, the governor of Pennsylvania, happened to be vis-
iting the White House on one of the greenback’s better days, when
it gained a few percent against gold. When Curtin mentioned this
news, Lincoln’s face knotted up in anger. “Curtin,” he asked, “what do
you think of those fellows in Wall Street, who are gambling in gold at
such a time as this?” “They are a set of sharks,” Curtin replied. Lincoln
pounded his fist on the table. “For my part,” he exclaimed, “I wish
every one of them had his devilish head shot off!”2

1 Richard Franklin Bensel, Yankee Leviathan: The Origins of Central State


Authority in America, 1859–​ 1877 (Cambridge: Cambridge University Press,
1990), 152; Gregor W. Smith and R. Todd Smith, “Greenback-​Gold Returns
and Expectations of Resumption, 1862–1879,” Journal of Economic History 57
(1997): 698; New York Times, 23 June 1864, 3.
2 Francis Bicknell Carpenter, Six Months at the White House with Abraham Lincoln
(New York: Hurd and Houghton, 1866), 84.

6
╇ 7

The Land of Speculation 7

Lincoln was hardly alone in blaming speculators for the depreciation


of the greenback. Salmon Chase, the secretary of the Treasury, sternly
informed Congress that the dollar’s decline against gold was attribut-
able to “the efforts of speculators.” Senator John Sherman, who had
succeeded to Chase’s seat in the Senate and who would himself become
Treasury secretary a decade later, shared Lincoln’s and Chase’s certainty.
“We are confident,” he declared, that “speculative operations in gold
at this time have an injurious effect on the public credit.” Sherman ac-
cordingly introduced a bill to prohibit speculative transactions in gold.3
Lincoln, Chase, and Sherman were reacting to an immediate crisis,
but behind their harsh judgment was a long and contradiction-╉filled
history of thought regarding speculators and speculation. Americans
were inveterate speculators—╉on that point there was little disagree-
ment. But the consequences of all this speculation, and the moral char-
acter of the speculators, were matters of considerable debate.

The Greatest Sharpers in€the Universe


Foreign visitors to the United States often remarked on Americans’
passion for speculation. “An American merchant is an enthusiast who
seems to delight in enterprise in proportion as it is connected with
danger,” declared the German journalist Francis Grund. “He ventures
his fortune with the same heroism with which the sailor risks his life;
and is as ready to embark on a new speculation after the failure of a
favorite project, as the mariner is to navigate a new ship, after his own
has become a wreck.” In Chicago, the English writer Harriet Martineau
reported that “the streets were crowded with land speculators, hur-
rying from one sale to another,” to the point where “it seemed as if
some prevalent mania infected the whole people.” The French econo-
mist Michel Chevalier found the same atmosphere in Pennsylvania.
“Every body is speculating, and every thing has become an object of
speculation,” he observed. “The most daring enterprises find encour-
agement; all projects find subscribers.” For an American, insisted the

3 Congressional Globe, 15 Apr. 1864, 1640.


8

8 Speculation

English missionary Isaac Fidler, trading “upon speculation and credit”


was simply “the custom of his country.”4
Observers offered a variety of theories to explain why Americans
were such ardent speculators. Tocqueville chalked it up to democracy.
“Those who live amid democratic instability constantly have the image
of chance before their eyes, and in the end they love all undertakings in
which chance plays a role,” he reasoned. “They are therefore all brought
into commerce, not only because of the gain it promises them, but
for love of the emotions that it gives them.” Chevalier cited puritani-
cal social norms, which left speculation the only available way to have
fun. “Public opinion and the pulpit forbid sensual gratifications, wine,
women, and the display of a princely luxury; cards and dice are equally
prohibited,” he mused. “The American therefore, has recourse to busi-
ness for the strong emotions which he requires to make him feel life.
He launches with delight into the ever-​moving sea of speculation.” The
English naval officer Frederick Marryat, perhaps more plausibly, em-
phasized the fast-​changing economic conditions in the United States.
“If an American has money sufficient to build a two-​story house, he
will raise it up to four stories on speculation,” Marryat remarked. “They
speculate on the future, but the future with them is not distant as it is
with us.” But whatever the reason, foreign travelers had little doubt of
Americans’ propensity to speculate. “Were I to characterise the United
States,” one English tourist concluded, “it should be by the appellation
of the land of speculation.”5

4 Francis J. Grund, The Americans, in Their Moral, Social, and Political Relations
(Boston: Marsh, Capen and Lyon, 1837), 240–​41; Harriet Martineau, Society in
America (London: Saunders and Otley, 1837), 1:350; Michel Chevalier, Society,
Manners and Politics in the United States (Boston: Weeks, Jordan and Co., 1839),
305; Isaac Fidler, Observations on Professions, Literature, Manners, and Emigration,
in the United States and Canada (New York: J. & J. Harper, 1833), 76.
5 Alexis de Tocqueville, Democracy in America (1835–​ 40), ed. Harvey C.
Mansfield and Delba Winthrop (Chicago: University of Chicago Press, 2000),
528; Chevalier, Society, Manners and Politics, 309; Frederick Marryat, Diary in
America (1839), ed. Jules Zanger (Bloomington: Indiana University Press, 1960),
137; William Priest, Travels in the United States of America (London: J. Johnson,
1802), 132.
  9

The Land of Speculation 9

When Americans noticed the same trait in themselves, they tended


to deplore it. “We know that there is a wild and daring spirit of mer-
cantile speculation among us, and that there is such a thing as tempt-
ing this spirit, to a most disastrous extent,” lamented the New Yorker
Charles Glidden Haines. “The spirit of speculation beats high; extrava-
gant and disastrous contracts are entered into, and golden dreams of
wealth too often mislead, even the most cautious.” A  contemporary
agreed that “speculation may perhaps be considered one of the foibles,
to use no harsher name, of the American character.” Another lamented
“those peculiar characteristics of the American people which lead them
rather to shine as jobbers and gamblers than to excel as merchants.” As
more and more businesses took the corporate form to sell shares to in-
vestors, some worried about all the new corporations for precisely this
reason. “The people of these United States, are a race of men having
few, if any equals in the world,” declared one skeptic. “But they have
their foibles. They cannot harden their hearts against the seductions
of speculation.” As another put it, “there is not a nation on earth so
vigilant to watch and so eager to catch at a prospect of speculation, as
the citizens of the U. States.”6
Were Americans more prone to speculation than Europeans? The
economist John McVickar thought so. “We observe in our country,”
he reasoned, “the greater frequency of individual failures and of general
commercial revolutions, than is exhibited in most of the older countries
of Europe, where something like the slow pulse of age seems to render
comparatively innoxious those seeds of disease, which in our own
warmer temperament, run out into unsound and feverish speculation.”
The Bankers’ Magazine thought so too. “The English might teach us
Americans a wholesome lesson,” it reflected. “It seldom happens that an
English merchant, tradesman, or banker abandons his rightful trade or
calling to engage in any thing like doubtful speculation.” A missionary
journal agreed that “in no other country has this rage for speculation

6 Charles G.  Haines, Substance of Mr. Haines’ Remarks, Made at the City Hotel
(New York: William A. Mercein, 1823), 9; Auctions Inconsistent with Regular Trade
(New York: Van Winkle, Wiley & Co., 1817), 4; “Merchants and Speculators,”
The Round Table 9 (1869):  294; Letter on the Use and Abuse of Incorporations
(New York: G. & C. Carvill, 1827), 23; New-​Bedford Mercury, 15 Apr. 1825, 3.
10

10 Speculation

been more fully shown than in the United States.” But this was hardly
a universal view. James Kent, one of the leading judges of the early
republic, recalled that there was no shortage of speculation in Europe
either. “The dreams and the madness of speculation,” he concluded, “is
a disease which has prevailed on each side of the Atlantic.”7
The primary commodity in which early Americans speculated was
land. There was plenty of it; it came in all kinds, from small parcels of
developed land in eastern cities to vast and scarcely populated western
territories; and its prices fluctuated, sometimes quite sharply. As Charles
Francis Adams complained, “the great object of domestic speculation in
America is undoubtedly land.” When lots were sold in the new capital
city of Washington, according to Justice Hugh Henry Brackenridge of
the Pennsylvania Supreme Court, most of the initial purchases “were
on the speculation of an under sale to others, before the money was
paid” by the ostensible buyer. Up in Maine, reported the promoter
Moses Greenleaf, many of the landholders were nonresident specula-
tors, who “purchased with a view of retaining the land for a time, in
expectation of immense profits from the sale, when the country about
them should become improved, and peopled by the exertions of others,
without any trouble of their own.” Affluent and prominent Americans
pooled their capital to form land companies, some of the largest busi-
ness enterprises of the era, to buy enormous swaths of land in the West
in the hope that future settlement would make the land more valuable.
Land was simply a smart investment, advised the French lawyer J. P.
Brissot de Warville, who took a close look at the land market while
touring the new nation in 1788. Rather than putting money in a bank
to provide for one’s children, Brissot suggested, “nothing appears to me
better to answer this wise precaution, than to place such money on the
cultivated soil of the United States.”8

7 John McVickar, Hints on Banking (New York: Vanderpool & Cole, 1827), 4;


“The Causes of Commercial Crisis,” Bankers’ Magazine 5 (1850):  1; “On the
Fluctuations of Property, and the Increase of Speculation,” The Panoplist, and
Missionary Herald 16 (1820): 454; Jackson v. Port, 17 Johns. 479 (N.Y. Sup. 1820).
8 Malcolm J. Rohrbough, The Land Office Business: The Settlement and Administra­tion
of American Public Lands, 1789–​1837 (New York: Oxford University Press, 1968);
Robert P. Swierenga, Pioneers and Profits: Land Speculation on the Iowa Frontier
(Ames: Iowa State University Press, 1968); Elizabeth Blackmar, Manhattan
  11

The Land of Speculation 11

Contemporaries had little doubt that the United States was very dif-
ferent from Europe in this regard. “In Europe the value of real estate is
in general comparatively stationary,” explained the chemist Robert Hare,
“but here it is always an article of speculation; and as a large portion,
while unproductive, is still held with a view to its future value, the es-
timate put upon that value is liable to great changes.” In Britain, noted
Massachusetts Chief Justice Isaac Parker, no one owned sections of forest
far away from their homes. “In this country, on the contrary,” he ob-
served, “there are many large tracts of uncultivated territory owned by
individuals who have no intention of reducing them to a state of im-
provement, but consider them rather as subjects of speculation and sale.”
It was because of “the unrestrained spirit of speculation,” agreed William
Tilghman, the chief justice of Pennsylvania’s Supreme Court, that “we see
freehold estates pass in rapid circulation from owner to owner.”9
Financial assets soon became a second big area of speculation, first
with the issuance of state and national public debt securities to finance
the American Revolution, then with the sale of US debt securities in
the early 1790s, and finally with the steady expansion, all through the
early republic, of the market for shares in banks and other business cor-
porations. “Stockjobbing drowns every other subject,” James Madison
complained in 1791, shortly after shares of the Bank of the United States
were first sold to the public. In New  York, he reported, “the Coffee
House is in an eternal buzz with the gamblers.” Henry Lee traveled from
Philadelphia to Alexandria, Virginia, around the same time and discov-
ered that “my whole rout[e]‌presented to me one continued scene of
stock gambling; agriculture commerce & even the fair sex relinquished,

for Rent, 1785–​1850 (Ithaca, N.Y.: Cornell University Press, 1989), 183–​212; Charles
Francis Adams, Reflections Upon the Present State of the Currency in the United
States (Boston: Ezra Lincoln, 1837), 21; Stoddart v. Smith, 5 Binn. 355 (Pa. 1812);
Moses Greenleaf, A Statistical View of the District of Maine (Boston: Cummings
and Hilliard, 1816), 92; Shaw Livermore, Early American Land Companies: Their
Influence on Corporate Development (New York: Commonwealth Fund, 1939);
J. P. Brissot de Warville, New Travels in the United States of America (Boston:
Joseph Bumstead, 1797), 41.
9 Robert Hare, An Effort to Refute the Opinion, That No Addition Is Made to the
Capital of a Community by Banking (s.l.: s.n., 1834); Conner v. Shepherd, 15 Mass.
164 (1818); Jack v. Shoemaker, 3 Binn. 280 (Pa. 1810).
12

12 Speculation

to make way for unremitted exertion in this favourite pursuit.” As shares


of government debt and shares of business corporations became more
widely traded, flurries of financial speculation would become regular
features of American commercial life. By 1822, Chief Justice Tilghman
could already deplore how quickly his fellow citizens threw their money
into all these new corporations. “There has prevailed among us, to an
unfortunate degree, a pestilent spirit of speculation, which has induced
some, without means of payment, to subscribe to projects of all kinds,
with a hope to selling out to advantage, as soon as the stock has risen,”
Tilghman lamented. “These speculative subscriptions have many bad
consequences.”10
Tilghman was hardly alone in worrying that Americans were spec-
ulating too much. As early as 1779, in the midst of the American
Revolution, George Washington was already concerned. “Virtue and
patriotism are almost extinct!” he declared. “Stockjobbing, speculating,
engrossing, seem to be the great business of the day.” (A decade earlier,
Washington himself had been a major speculator in western land.) This
sort of criticism never let up. “One most pernicious consequence of the
unsettled state of commerce, is the prevalence of a spirit of speculation,”
insisted a committee of New York merchants in 1817. They deplored
how speculation “corrupts the morals of the community, tarnishes its
reputation, deranges its commerce, and more or less directly injures
every portion of the state.” “As fast as one bubble of speculation breaks
another is blown up,” one editorialist fretted. “We are in the high road
of becoming the greatest sharpers in the universe.”11

10 E. James Ferguson, The Power of the Purse: A History of American Public


Finance, 1776–​1790 (Chapel Hill: University of North Carolina Press, 1961);
Ronald E. Seavoy, The Origins of the American Business Corporation, 1784–​
1855 (Westport, Conn.: Greenwood Press, 1982); James Madison to Thomas
Jefferson, 10 July 1791, in William T. Hutchinson et al., eds., The Papers of
James Madison (Chicago: University of Chicago Press, 1962–​), 14:43; Henry
Lee to James Madison, 24 Aug. 1791, in ibid., 14:73; Bernia Turnpike Road v.
Henderson, 8 Serg. & Rawle 219 (Pa. 1822).
11 George Washington to Henry Laurens, 5 Nov. 1779, in Jared Sparks, ed.,
The Writings of George Washington (Boston:  Russell, Odiorne, and Metcalf,
1834–​37), 6:397; Columbian [New  York], 11 Jan. 1817, 2; General Advertiser
[Philadelphia], 20 Mar. 1792, 2.
  13

The Land of Speculation 13

Such expressions of concern often came near the top of a cycle, as a


rise in prices drew newcomers to the market. For example, during the
bubble of 1791–​92, the first stock market boom and crash in American
history, many observers were nervous even while things seemed to be
going well. “They say the evil will cure itself,” Thomas Jefferson brooded
as the market kept heading up and up. “I wish it may. But I have rarely
seen a gamester cured even by the disasters of his vocation.” In 1835,
the rising prices of land and corporate shares created what the politi-
cal economist Thomas Dew called “a speculating mania,” one he ex-
pected would cause ruin when prices came back down. “A gambling
spirit is apt to prove epidemic,” Niles’ Weekly Register worried. “Verily,
the people are mad!”12
And of course the bottom of a cycle drew even more criticism of
speculation, from writers eager to explain what had gone wrong. After
the crash of 1792, one gathering offered as a toast: “The Guillotine
to all Tyrants, Plunderers, and funding Speculators.” A committee of
the Pennsylvania legislature blamed the panic of 1819 on “the inclina-
tion of a large portion of the community, created by past prosper-
ity, to live by speculation and not by labor.” Looking back at the
recession of the late 1820s, the cartographer John Melish cited as a
cause the fact that “speculation has been too often substituted for
industry.” The accountant Benjamin Foster offered the same explana-
tion for the recession of the late 1830s. He recalled how in 1836 “the
spirit of speculation and adventure pervaded the entire community.
The gambling propensities of human nature were constantly solic-
ited into action; and crowds of individuals of every description,—​
the credulous and the suspicious—​the crafty and the bold—​the raw
and the experienced—​the intelligent and the ignorant—​politicians,
lawyers, physicians and divines, hastened to venture some portion
of their property in schemes of which scarcely any thing was known
except the name.” Sure enough, disaster ensued:  “Suddenly, how-
ever, the bubble burst—​unlimited confidence gave place to universal

12 Thomas Jefferson to Gouverneur Morris, 30 Aug. 1791, in Julian P. Boyd et al.,


eds., The Papers of Thomas Jefferson (Princeton, N.J.:  Princeton University
Press, 1950–​), 22:105; Thomas R. Dew, “On Price,” Farmer’s Register 3 (1835): 65;
Niles’ Weekly Register 48 (1835): 167.
14

14 Speculation

distrust—╉a reaction took place, and thousands found themselves in-


volved in bankruptcy, misery and ruin.”13 No one seems to have wor-
ried that Americans were too timid with their money. The concern
was all in the opposite direction.

Speculation’s Numerous Intrigues


What exactly was wrong with speculation? Americans in the early re-
public addressed this question at great length. Most of their efforts, it
seems fair to say, were instinctive expressions of suspicion rather than
well-╉thought-╉out theories. Yet by the middle of the nineteenth century,
so much had been written on the topic that we can piece together a
few basic principles underlying what seems to have been a widespread
antipathy toward speculation.
Perhaps the most common critique of speculation was that it
was a form of gambling. At times speculation was merely analo-
gized to gambling—╉critics charged that speculation “introduced a
spirit of gaming,” produced “all the pernicious effects of gaming,”
or was conducted “on common gambling principles.” But others in-
sisted that speculation was gambling, dressed up with a fancy name.
“Gambling in business,” insisted the printer James Ronaldson, “in
our modern commercial technicality, is called ‘speculation.’â•›” The
journalist William Leggett likewise condemned “this pervading
spirit of speculation (or spirit of gambling, as it might with more
propriety be called, for it is gambling, and gambling of the most
desperate kind).” Henry Ward Beecher warned young men that “a
Speculator on the exchange, and a Gambler at his table, follow one
vocation, only with different instruments. One employs cards or
dice, the other property.” On Wall Street, agreed the satirist George
Foster, “gambling is carried on as the chief business there; but it is

13 William Cobbett, A Bone to Gnaw, for the Democrats (Philadelphia: Thomas


Bradford, 1795), 44; Journal of the Thirtieth House of Representatives of the
Commonwealth of Pennsylvania (Harrisburg, Penn.: James Peacock, 1819–╉20),
422; John Melish, The Traveller’s Manual; and Description of the United States
(New York: A. T. Goodrich, 1831), 127; B. F. Foster, The Merchant’s Manual
(Boston: Perkins & Marvin, 1838), 32–╉33.
  15

The Land of Speculation 15

upon so gigantic, so systematic a scale, that it reaches the dignity of


history.”14
Speculation, on this view, was simply a euphemism employed to
make gambling sound more respectable. “The words speculate and spec-
ulation have been substituted for gamble and gambling,” the English
writer William Cobbett argued in a book republished in New  York.
“The hatefulness of the pursuit is thus taken away.” Indeed, until the
late eighteenth century, speculation meant philosophical contempla-
tion. The word’s newer commercial meaning was only a few decades old
in the early nineteenth century. Using it to refer to risky transactions
thus carried a whiff of legitimization that it no longer does today. As
the actor Leman Thomas Rede pointed out, “this species of gambling
is called speculation, a philosophical word, dragged forcibly into the
service of chicane and avarice.”15
Critics who equated speculation with gambling were drawing upon a
long tradition of antigambling thought in both Britain and the United
States. Gambling “is an offence of the most alarming nature,” lectured
the English judge William Blackstone in his ubiquitous Commentaries,
“tending by necessary consequence to promote public idleness, theft,
and debauchery.” And that was just among the poor. “Among persons
of a superior rank,” Blackstone continued, “it hath frequently been at-
tended with the sudden ruin and desolation of antient and opulent
families, an abandoned prostitution of every principle of honour and
virtue, and too often hath ended in self-​murder.” The young lawyer (and

14 Freeman’s Journal [Philadelphia], 27 July 1791, 3; Facts Important to Be Known


by the Manufacturers, Mechanics, and All Other Classes of the Community
(New York: s.n., 1831), 37; William M. Gouge, A Short History of Paper Money
and Banking in the United States (Philadelphia: T. W. Ustick, 1833), 75; Laban
Heath, On Paper Money (Philadelphia: s.n., 1825), 13; James Ronaldson, Banks
and a Paper Currency (Philadelphia: s.n., 1832), 4; Theodore Sedgwick Jr., ed.,
A Collection of the Political Writings of William Leggett (New York: Taylor and
Dodd, 1840), 2:87; Henry Ward Beecher, Lectures to Young Men, on Various
Important Subjects (New  York:  M.  H. Newman & Co., 1849), 76; George
G. Foster, New York in Slices (New York: W. F. Burgess, 1849), 16.
15 William Cobbett, Advice to Young Men (New  York:  John Doyle, 1831), 17;
Leman Thomas Rede, The Art of Money Getting (Boston: Richardson, Lord &
Holbrook, 1832), 36.
16

16 Speculation

future lord chancellor) Thomas Erskine agreed that gambling “is from
the very beginning a sordid, ungenerous, dishonest passion.” American
ministers had long decried gambling as an especially grievous sin. “It
is not easy to conceive any vice more hateful to God than Gambling,”
screamed the itinerant minister Mason Locke Weems in all upper-​case
letters, “because none can be conceived more diametrically op-
posite to the very end of our creation!” Secular critics emphasized
that gambling led inexorably to idleness and financial ruin, and often
to madness and even suicide. Calling speculation a kind of gambling
was a shorthand way of summoning up all these dire consequences.
“Speculation and Gambling,” mused one journalist—​“on many oc-
casions, these words are synonymous—​and on some, there is another
which has nearly the same meaning—​to wit, roguery. Sometimes, per-
haps also, insanity may convey nearly a similar idea.”16
Like gambling, speculation was widely viewed as immoral and un-
christian. This was a common theme in the religious press. “Speculation
is contrary to the spirit of Christianity and a great hindrance to the
spread of the gospel throughout the world,” declared the Religious
Intelligencer. The Episcopal Recorder maintained that “it is our duty not
to sanction the pursuit of wealth through the medium of speculation,”

16 William Blackstone, Commentaries on the Laws of England (Oxford: Clarendon


Press, 1765–​69), 4:171; Thomas Erskine, Reflections on Gaming, Annuities, and
Usurious Contracts (London: T. Davies, 1776), 6; William Stith, The Sinfulness
and Pernicious Nature of Gaming (Williamsburg, Va.: William Hunter, 1752);
A Letter to a Gentleman on the Sin and Danger of Playing at Cards and Other
Games (Boston: D. Fowle and Z. Fowle, 1755); Eli Hyde, A Sermon; in Which
the Doctrine of the Lot, Is Stated, and Applied to Lotteries, Gambling, and Card-​
Playing, for Amusement (Oxford, N.Y.: John B. Johnson, 1812); Mason Locke
Weems, God’s Revenge Against Gambling (Augusta, Ga.: Hobby & Bunce,
1810), 12; Jacob Rush, Charges, and Extracts of Charges, on Moral and Religious
Subjects (New York: Jonathan Weeden, 1804), 120–​21; Samuel Miller, The Guilt,
Folly, and Sources of Suicide (New York: T. and J. Swords, 1805), 52–​53; Report of
a Committee Appointed to Investigate the Evils of Lotteries (Philadelphia: Daniel
B. Shrieves, 1831), 11–​12; George William Gordon, A Lecture Before the Boston
Young Men’s Society (Boston: Temperance Press, Ford & Damrell, 1833), 12; Job
R. Tyson, A Brief Survey of the Great Extent and Evil Tendencies of the Lottery
System (Philadelphia: William Brown, 1833), 27; “Speculation and Gambling,”
Niles’ Weekly Register, 3 July 1819, 16.
  17

The Land of Speculation 17

while Zion’s Herald warned that “a successful speculation is oftener an


evil than a blessing; … it is a moral evil, because it violates the law of
nature.” A Quaker newspaper explained that “speculation in dealing,
however common and popular,” was nevertheless “wrong in principle,”
because “it is unjust for one man to seek to gain at the expense of an-
other’s loss.” But assertions of the immorality of speculation pervaded
the secular press as well. One newspaper called it a form of covetous-
ness and idolatry. Another published a poem called To Land Speculators,
which began:

Of you, oh, ye land-​sellers, will heaven demand


An account of your buying and selling of land:
By you is the arm of the murderer nerved;
By you are the public with robbers well served:
By you is the razor of suicide guided;
By you is the christian religion derided.

One farmers’ magazine adapted a quotation from the Book of Matthew.


“When the demon of speculation has seized upon a man,” the maga-
zine pronounced, “he walketh through dry places, seeking rest but fin-
deth none.”17
Sometimes speculation was deplored because of its effects on the spec-
ulators themselves. Unless it was stopped, worried a New Jersey news-
paper as the 1791 bubble began to inflate, “the rage for speculation …
cannot fail of ruining many innocent and credulous persons, who are
not proof against the temptations which are held out, nor capable of
discerning the snares that are laid for them.” Such amateur venturers
were not the only victims, because their wives and children might be
dragged down with them. “How many families enjoying a competence
in quietude and peace,” wondered one observer during the recession of
the late 1820s, “have by such an unnatural state of things, been plunged

17 “Money Speculations Among Christians,” Religious Intelligencer 21 (1836): 385;


“Warning Against the Spirit of Pecuniary Speculation,” Episcopal Recorder 14
(1836): 153; “Speculation,” Zion’s Herald 8 (1837): 112; “On Speculation,” Friends’
Intelligencer 15 (1858): 7; New York Observer, 17 July 1841, 1; Young America, 10
May 1845, 4; “The Spirit of Speculation,” Maine Farmer 33 (1865): 2.
18

18 Speculation

into distress and ruin[?]‌” Learned treatises and popular advice manu-
als alike advised the uninitiated to avoid speculating, lest they plunge
themselves and their families into poverty. The French Duke de la
Rochefoucault, traveling through the United States in the 1790s, wor-
ried that middling farmers were losing their farms—​ironically, after an
especially plentiful harvest—​because they had been speculating in crop
prices, which fell due to the increase in supply. Speculation was dan-
gerous even for those who could afford to lose the most, warned John
Bannister Gibson, Pennsylvania’s long-​serving chief justice. “We know
from experience,” he cautioned, “that when a spirit of speculation, or
desire of inordinate gain, infects the rich, it terminates in scenes of ruin
and devastation, as wide spread and deplorable in their consequences
of misery and want, in the domestic relations of life, as if it had been
confined to those who had comparatively little to lose.”18
Other critics focused not on speculation’s economic effects but on its
moral and psychological implications. The problem was not that specu-
lation made people poorer, but that it made them selfish and imprudent.
By enabling one person to gain at the expense of another, one correspon-
dent charged, speculation “tends to dissolve that natural tie, that binds
man to man.” As they grew accustomed to cheating, another alleged,
“speculators have been generally strangers to mercantile habits, to mer-
cantile accuracy and rectitude.” The stereotype of the speculator was one
who thought only of his own profit and would not hesitate to take ad-
vantage of others. As the poet Philip Freneau depicted him in the 1790s,

With soothing words the widow’s mite he gain’d,


With piercing glance watch’d misery’s dark abode,
Filch’d paper scraps while yet a scrap remain’d,
Bought where he must, and cheated where he cou’d.

18 New Jersey Journal, 10 Aug. 1791, 3; A Peep Into the Banks (New York: Vanderpool
& Cole, 1828), 29; Francis Wayland, The Elements of Political Economy
(New  York:  Leavitt, Lord & Co., 1837), 379; Hints to Young Tradesmen
(Boston: Perkins & Marvin, 1838), 51; “Commercial Chronicle and Review,”
Merchants’ Magazine 28 (1853): 466–​67; Duke de la Rochefoucault Liancourt,
Travels Through the United States of North America, 2nd ed. (London: R. Phillips,
1800), 1:242–​43; Seidenbender v. Charles’s Administrators, 4 Serg. & Rawle 151
(Pa. 1818).
  19

The Land of Speculation 19

The printer Asa Greene summed up this line of thought in his account
of New York’s commercial life, a book he subtitled A Taste of the Dangers
of Wall Street. “The more a man engages in speculation,” Greene argued,
“the less tender his conscience grows on the subject of doing to others
as he would have them do to him. So debasing are the arts and shifts
which are too apt to prevail in a mercantile community.”19
Speculation did not just make people meaner, on this view; it also
made them turn to lives of luxury and vice. The young Clement Clarke
Moore, who as an older man would become better known for the poem
“A Visit from St. Nicholas,” condemned “a spirit of speculation which
tends, not to advance the permanent good of the society at large, but
to the introduction of that luxury and dissipation which pervades our
sea-​ports.” The scientist William Maclure agreed that speculation had
the effect of “increasing luxury, dissipation and vice, by introducing an
artificial anticipation of riches.”20
Meanwhile, speculation made people unhappy and unpleasant.
Speculative emotions in business “are transfused from the counting-​
house to the fire-​side,” one critic charged. “The visionary profits of one
day stimulate extravagance, and the positive losses of another engender
spleen, irritation, restlessness, a spirit of gambling and domestic inqui-
etude.” What is the “difference between speculation and honesty?” an-
other asked. “Honesty is modest, humble, distant and reserved,” while
“speculation is forward, froward, bullying, hectoring and insinuating.”
The reason, according to one rural newspaper, was that “a Speculator
appears surrounded by wants, when in fact he wants Nothing. He too
frequently grasps at what he cannot get.”21

19 Dunlap’s American Daily Advertiser [Philadelphia], 28 Feb. 1793, 2; John


M’Donald, The Danger of America Delineated (Cooperstown, N.Y.:  Elihu
Phinney, 1799), 17; Philip Freneau, “On Pest-​eli-​hali, the Travelling Speculator,”
in Poems (Monmouth, N.J.: Philip Freneau, 1795), 429; Asa Greene, The Perils
of Pearl Street (New York: Betts & Anstice, 1834), 172.
20 Clement Clarke Moore, An Inquiry Into the Effects of Our Foreign Carrying
Trade (New  York:  E.  Sargeant, 1806), 41; William Maclure, Opinions on
Various Subjects (New Harmony, Ind.: School Press, 1831), 1:329.
21 Ferris Pell, Letter to Albert Gallatin, Esq. on the Doctrine of Gold and Silver
(New York: Gould and Van Pelt, 1815), 34; James Jackson, The Letters of Sicilius,
to the Citizens of the State of Georgia (Augusta, Ga.: s.n., 1795), 54; Kennebec
Gazette, 14 Nov. 1800, 1.
20

20 Speculation

Worse, speculators were susceptible to illnesses brought on by the


pressure of their vocation. “The gambling spirit as constantly haunts
the Exchange and Corn-​market as the play-​table,” one doctor cau-
tioned, “and, by perplexing and distracting the mind, soon saps the
basis of health, and anticipates old age.” Benjamin Rush, the most
prominent physician of the early republic, even cited speculation as a
leading cause of mental illness. “In commercial countries, where large
fortunes are suddenly acquired and lost, madness is a common disease,”
he noted. “It is most prevalent at those times when speculation is sub-
stituted to regular commerce.” This was particularly noticeable in the
United States, he continued. “The funding system, and speculations in
bank scrip, and new lands, have been fruitful sources of madness in our
country,” Rush observed. “Sixteen persons perished from suicide in the
city of New York, in the year 1804, in most of whom it was supposed
to be the effect of madness, from the different and contrary events of
speculation.”22
Speculation was also widely criticized, not for its effect on the specu-
lators, but for its effect on the economy as a whole. One common line
of attack was that speculation was a nonproductive activity that sucked
time and money away from productive labor. “Agriculture and manu-
factures, I consider as the sources and foundation of national wealth,”
declared Tench Coxe, who would soon become assistant secretary of
the Treasury under Alexander Hamilton. Speculation, by contrast, was
in Coxe’s view “of no benefit to the nation, as the stock of national
wealth receives no augmentation therefrom.” (Coxe did not let this
dim view of speculation prevent him from becoming one of the lead-
ing land speculators in the country, or from using information gained
from his Treasury position to give investment tips to his friends.) Every
hour spent speculating was an hour taken away from a more useful
activity. “In my opinion nothing scarcely can be worse,” explained the
Philadelphia merchant Pelatiah Webster. Speculation just tended “to
draw people from the honest and painful method of earning fortunes,

22 “Physical Effects of Mercantile Speculation,” Episcopal Recorder 26 (1848): 88;


Benjamin Rush, Medical Inquiries and Observations, Upon the Diseases of the
Mind (Philadelphia: Kimber & Richardson, 1812), 66.
  21

The Land of Speculation 21

and to encourage them to pursue chimerical ways and means of ob-


taining wealth by sleight of hand.” As one poet complained during the
1791–​92 bubble,

Gambling, stock-​jobbing, form most powerful leagues


With Speculation’s numerous intrigues;
Intrigues that add not to the nation’s store,
But for the rich defraud the helpless poor.

Or as Thomas Jefferson complained right around the same time, “ships


are lying idle at the wharfs, buildings are stopped, capitals withdrawn
from commerce, manufactures, arts and agriculture, to be employed in
gaming, and the tide of public prosperity almost unparalleled in any
country, is arrested in its course, and suppressed by the rage of getting
rich in a day.”23
In later decades the point would be repeated by some of the early
American political economists, who likewise distinguished between
productive labor and nonproductive speculation. “Increasing the for-
tunes of individuals does not necessarily increase public wealth,” Daniel
Raymond cautioned, “unless the gain is caused by an increase of in-
dustry; for what one gains by speculation, another must lose.” John
McVickar agreed that speculation could “have no influence on national
prosperity,” because “as the wealth of the country merely changes hands,
the profits are but the criterion of a rising market, since such accumu-
lation would have taken place had the commodity continued in the
hands of its original holders.”24 Early advocates for labor made the same

23 Tench Coxe, Observations on the Agriculture, Manufactures and Commerce of


the United States (New York: Francis Childs and John Swaine, 1789), 85; Jacob
E. Cooke, Tench Coxe and the Early Republic (Chapel Hill: University of North
Carolina Press, 1978), 79–​82, 311–​33, 176–​77; Pelatiah Webster, A Seventh Essay
on Free Trade and Finance (Philadelphia: Eleazer Oswald, 1785), 23; The Glass;
or, Speculation: A Poem (New York: s.n., 1791), 4; Thomas Jefferson to Edward
Rutledge, 25 Aug. 1791, in Papers of Thomas Jefferson, 22:74 (spelling corrected).
24 Daniel Raymond, Thoughts on Political Economy (Baltimore: Fielding Lucas
Jr., 1820), 306; Paul K. Conkin, Prophets of Prosperity: America’s First Political
Economists (Bloomington: Indiana University Press, 1980), 88–​93; John Ramsay
McCulloch, Outlines of Political Economy, ed. John McVickar (New York:
Wilder & Campbell, 1825), 91 (the quoted material is from McVickar’s editorial
22

22 Speculation

argument:  real wealth came from production, not from speculation.


“What are speculators?” asked a farmers’ magazine, which immediately
supplied an answer: speculators were a class that “are a burthen upon
society—​that live off of it like leeches.”25 Speculation thus harmed the
economy, by distracting people from doing more valuable things.
Speculation in land was especially vulnerable to this critique, because
land speculators were not merely accused of wasting their own time and
money; they were also open to the charge of wasting the productivity of
the land itself, and in a particularly conspicuous way, when absentee-​
owned land lingered unexploited. Land speculators “traverse this fine
country like a pestilent blight,” Morris Birkbeck reported from Illinois.
“When they see the promise of a thriving settlement, from a cluster of
entries being made in any neighbourhood, they purchase large tracts
of the best land, and lock it up in real mortmain, for it is death to all
improvement.” A Boston newspaper agreed that “the Land speculations
in the United States are become a Pandora’s box of evils to the people”
and urged that public land should never be sold to speculators but only
to actual settlers. When land was owned by speculators rather than
farmers, concluded William Tilghman, “agriculture suffers—​either the
lands remain a desert, or they are occupied by poor intruders, who
knowing the instability of their title, are afraid to attempt any valuable
improvement.”26 Speculation, on this view, hindered the productivity
of American land as much as American labor.
Many critics also cited a second reason speculation harmed the
economy:  it gave rise to a boom-​and-​bust cycle that destroyed even
those who had not engaged in speculation themselves. “Of the panics
and destruction of credit attendant on the re-​action of all extensive

notes). On the range of early American economists’ views of speculation, see


Brian Harding, “Transatlantic Views of Speculation and Value, 1820–​1860,”
Historical Research 66 (1993): 209–​21.
25 Stephen Simpson, The Working Man’s Manual (Philadelphia:  Thomas
L. Bonsal, 1831), 74; “Labor vs. Speculation,” Maine Farmer and Journal of the
Useful Arts 4 (1837): 394; “Speculators,” Prairie Farmer 6 (1860): 147.
26 Morris Birkbeck, Letters from Illinois (Philadelphia: M. Carey and Son, 1818),
31; Boston Gazette, 21 Mar. 1796, 2; William Tilghman, An Address Delivered
Before the Philadelphia Society for Promoting Agriculture (Philadelphia: William
Fry, 1820), 7.
  23

The Land of Speculation 23

speculations,” observed the lawyer William Beach Lawrence, “no one


conversant with the transactions of our commercial emporium can be
ignorant.” This sort of criticism understandably tended to swell at the
bottom of each cycle. As one correspondent put it, just after the crash
of 1792, “the mad moment of Stock Jobbing—​Land Jobbing—​and
wild Speculation is past—​leaving us to rue the fatal effects.” Timothy
Dwight, the former president of Yale, was in nearby Middletown,
Connecticut, not long after the Panic of 1819. The commerce of the
town “has obviously declined within a few years past,” he noted. “The
first cause, both in time and efficiency, of this evil was, if I mistake not,
what has been proverbially called in this country Speculation.” In 1837,
as banks were failing throughout the country, the New York socialite
Philip Hone took real estate prices as a measure of his city’s distress.
Unimproved lots which had cost $480 just a year earlier were now sell-
ing for $50. “The immense fortunes which we heard so much about in
the days of speculation have melted away like the snows before an April
sun,” Hone wrote in his diary. “No man can calculate to escape ruin but
he who owes no money.”27
Most such critics simply assumed a causal connection between spec-
ulation and economic cycles, but some tried to identify the mechanism
linking the two. The economist Mathew Carey argued that speculators
were the only beneficiaries of “those sudden vibrations of bank accom-
modation whereby money is rendered superabundant at one time, and
immoderately scarce at another,” an oscillation “pernicious to morals,
industry, trade, and commerce.” During the recession of 1834, a com-
mittee of the House of Representatives laid out the same mechanism
in more detail. The committee identified the cause of recent economic
conditions as the overwillingness of banks to lend to speculators during
good times. “When business has been so successful for a time as to oc-
casion an eager desire to speculate in merchandise and real estate,” the
committee reasoned, “extensive discounts and loans will be demanded,

27 William Beach Lawrence, Two Lectures on Political Economy (New York: G. &
C. & H. Carvill, 1832), 63–​64; Federal Gazette [Philadelphia], 25 Apr. 1792, 3;
Timothy Dwight, Travels; in New England and New York (New Haven, Conn.:
Timothy Dwight, 1821), 1:218; Bayard Tuckerman, ed., The Diary of Philip
Hone 1828–1851 (New York: Dodd, Mead and Co., 1889), 1:251.
24

24 Speculation

which the banks cheerfully furnish to any extent, as it can be done


without cost, and to their great profit.” When things turned sour,
and borrowers could not repay these loans, the result was “heavy and
destructive losses, and much real suffering,” to an extent “impeding,
temporarily, the ordinary and salutary course of trade and industry.”
On this view, speculation was not a cause of the business cycle but an
intensifier of it, a factor making the highs higher and the lows lower
than they otherwise would be. But this distinction was not particularly
relevant to most critics, because either way, speculation was at the root
of the nation’s troubles. “To what shall we ascribe this suspension of
business which has thrown so many of the industrious out of employ-
ment and rendered it impossible for the poor mechanic to collect from
his rich employer the wages of his labor?” asked one populist pamphlet.
“I answer that the evil of the times may be comprehended in three
words—​overtrading—​overbanking and speculation.”28
Speculators were also widely accused of manipulating prices, to
benefit themselves at the expense of the broader economy. They were
often charged with making staple goods more expensive for consumers.
“We every day hear loud complaints against the avarice of speculators,”
reported a Boston newspaper, which explained that speculators had
driven the price of corn so high “that it is scarcely possible for a poor
man to procure a bushel of indian corn to make jonny-​cakes for his
half-​starved children.” A New York paper reported that within the past
two years, without any scarcity, “the mania of speculation” had caused
the price of salt to rise from 42 cents to $1.75 per pound, and molasses
from 45 cents to $1.50 per gallon. The Maine Farmer agreed that “the
present extra high price of breadstuffs in this country is kept up by the
mean tricks of the speculators.” “Speculators are controlling the prices
of sugar,” another critic charged a few years later. “They buy up large
quantities and withdraw most of it from the market, and then raise a
hue and cry of scarcity.” When food prices rose during the War of 1812,
many placed the responsibility on speculators, who were alleged to have

28 Mathew Carey, Essays on Banking (Philadelphia: Mathew Carey, 1816), 89–​


90; Report on Gold and Silver Coins, by a Select Committee of the House of
Representatives of the United States (Washington, D.C.: Gales & Seaton, 1834),
32; The Times: or, The Pressure and Its Causes Examined (Boston: s.n., 1837), 6.
  25

The Land of Speculation 25

created artificial shortages by buying up staples and keeping them off


the market. The citizens of Wilmington, Delaware, held a public meet-
ing to consider “the propriety of some measures with regard to the state
of speculation and speculators,” including whether to “withdraw all in-
tercourse commercial and social, with those who are combining to raise
the price of the necessaries of life.”29
At other times, however, speculators were accused of reducing the
prices received by farmers and ranchers. “The farmers of this state are
annually, in the sale of their wheat crops, ‘fleeced’ to the tune of sev-
eral hundred thousand dollars, which go into the pockets of chuck-
ling, greedy speculators,” despaired one correspondent to the Michigan
Farmer. Another alleged a conspiracy among “speculators to keep the
prices of wool down, until most of it has been purchased of those who
produce it.”30
Whether speculators were raising or lowering prices, critics were cer-
tain that a small wealthy cabal was exploiting the poor and middling
classes. “It is speculation, in its endless diversities, which simultane-
ously augments the capital of the rich, the acute, the prosperous, and
depresses the condition of the simple, the unfortunate, and the labour-
ing poor,” charged the Southern Quarterly Review. And when ordinary
citizens tried to get in on the game by speculating for themselves,
another magazine alleged, they would lose their all to “a minority of
wealthier speculators” who could “absolutely control the course of the
market for their own benefit.”31
Finally, speculation was often criticized for causing political cor-
ruption. The major speculative assets in the early republic—​land and
public debt securities—​were both initially brought onto the market
through the decisions of government officials, and even business

29 New-​England Palladium, 30 Nov. 1813, 1; The Columbian, 17 Dec. 1813, 2;


“Tricks of Breadstuff Speculators,” Maine Farmer 15 (1847):  2; “The Spirit
of Speculation—​How to Check It,” The Plough, the Loom and the Anvil 8
(1856): 519; New Hampshire Patriot, 28 Dec. 1813, 2; Petersburg Daily Courier, 3
Oct. 1814, 2; Enquirer [Richmond], 28 Dec. 1813, 3.
30 “Farmers vs. Speculators,” Michigan Farmer 6 (1848): 106; “The Wool Market,”
Michigan Farmer 11 (1853): 82.
31 “Speculation and Trade,” Southern Quarterly Review 30 (1856):  1–​2; “The
Declension of Speculation,” Merchants’ Magazine 62 (1870): 103.
26

26 Speculation

corporations were created one by one, through acts of the legislature.


There was money to be made, by speculators and by public officers
themselves, with foreknowledge of government decisions. In an era
before government officials were prohibited from using that knowledge
for their own gain, they were some of the leading investors in the public
debt and the public lands, and they were known to share investment
tips with their associates. “Our new Government is a Government of
Stock-​Jobbing and Favouritism,” the elderly George Mason grumbled.
William Findley, the Pennsylvania representative who was one of the
leading opponents of Hamilton’s financial program, agreed that the
national debt “promoted a depravity of morals and a great decline of
republican virtue.” The point was made again and again: speculation
in public assets like debt securities and land created an unholy alliance
of legislators and gamblers powerful enough to transfer wealth “from
those who earned it, to those who pilfered it,” and “to concentrate it
in a few hands.” The result, as one newspaper editor put it, was “the
subversion of our republican institutions.”32
Speculation corrupted the nation’s political culture in a second way,
critics charged—​by creating the conditions for European-​style inequal-
ity. “In our rising manufacturing villages,” a New Hampshire news-
paper cautioned, “this speculating mania rages to a great extent, and
is laying a foundation for that poverty, dependence and wretchedness
which characterizes the population of similar places in Europe. The few
are becoming immensely rich, the middling interest poor, and the poor
abject.” Unless speculation could be checked, “what the people affect so
much to deprecate, a monied aristocracy, is in a great measure created
by their own folly.”33

32 Public Speculation Unfolded (New York: David Denniston, 1800), 3; George


Mason to James Monroe, 9 Feb. 1792, in Robert A. Rutland, ed., The Papers
of George Mason, 1725–1792 (Chapel Hill: University of North Carolina Press,
1970), 3:1256; William Findley, A Review of the Revenue System (Philadelphia:
T. Dobson, 1794), 52; John Taylor, An Enquiry Into the Principles and Tendency
of Certain Public Measures (Philadelphia: Thomas Dobson, 1794), 42; James
Cheetham, Remarks on the “Merchants” Bank (New York: Southwick &
Hardcastle, 1804), 19.
33 Farmer’s Cabinet, 8 May 1835, 3.
╇ 27

The Land of Speculation 27

There were thus plenty of reasons in the early republic to be wary of


speculation. Like other forms of gambling, it bankrupted the specula-
tors themselves, it distracted people from engaging in more produc-
tive pursuits, and it made its practitioners selfish and sometimes even
suicidal. And speculation posed other dangers that most kinds of gam-
bling did not: it regularly plunged the country into economic crises,
and it corrupted the new nation’s political culture. By the middle of the
nineteenth century, there was a considerable body of antispeculation
thought. Most of it was neither systematic nor comprehensive, when
taken one book or one article at a time. But when put together, it con-
stituted a reservoir of ideas that could be drawn upon to criticize any
particular speculator or instance of speculation. If the United States was
a land of speculation, it was also a land of critics of speculation.

The Madness of€a Few for€the Gain of€the€World


But there was another side to the story. While Americans were elaborat-
ing reasons to dislike speculation, they were also building a set of jus-
tifications of the practice. This way of thinking was no more organized
or complete than its antithesis, but it too could serve as a grab bag of
debating points to rebut criticism of speculative transactions. By the
middle of the nineteenth century, there was a body of thought defend-
ing speculation.
It began with the uncontroversial premise that the investment of
capital in American enterprises was beneficial for the country. “It is a
long time, since the reaping of interest from capital, has been set free
from an erroneous prejudice,” recalled one commentator. “Labor, with-
out the assistance of capital, is very inefficient in the production of
value,” agreed the political economist Willard Phillips. Investors “hold
in their hands the essential means, not only of future production, but
of the very existence of the community.” Speaking at the opening of
the Auburn and Owasco Canal, the young politician William Seward
gloried in the willingness of Auburnites to invest in exciting new ven-
tures. If a traveler came to Auburn, Seward remarked, “we may call
his attention to our market to which are appropriated 20,000 dol-
lars, to this canal constituting an investment of 100,000 dollars, and
to the Auburn and Syracuse Rail Road with its capital of 400,000
28

28 Speculation

dollars, all of which funds, with inconsiderable exceptions, have been


subscribed by our own citizens, since the first of January last, and
ask what other population of five thousand have manifested within
the year a more efficient spirit of enterprise?” Even when investment
came from abroad, another future governor of New York suggested,
Americans were the beneficiaries. “The interest of this capital goes to
the foreign capital,” John Dix conceded, “but the profits of the in-
dustry, which it puts in motion, comprehending the subsistence and
compensation of the individuals employed, is obviously a gain to the
country, in which the investment is made. The industry of New-​York
has been in some degree stimulated by contributions of this nature.”34
There was little doubt that investment was something to be welcomed
and even encouraged.
Investment was necessary, but there was also no doubt that it was
risky. “When a man loans his property to another, there is always a
risk of his never being repaid,” lectured Francis Wayland, the presi-
dent of Brown University. “Now, the greater this risk, the greater will
be the interest which the capitalist may justly demand.” Few ways
of employing capital were riskier than putting it into a new business
venture. Looking at all the new manufacturing companies that had
been started in Massachusetts over the past twenty years, one lawyer
marveled at how many investors “were making this hazardous experi-
ment, which if successful would prove a great public benefit.” These
new enterprises could scarcely exist unless people with money were
willing to put it at risk. And what was speculation but putting money
at risk? “Let us say what we may on the subject,” one writer pointed
out, “the establishment of any new branch of trade, or the setting
on of a manufactory, partakes in no small degree of the character
of a speculation; indeed commerce itself can be deemed little else.”
The same argument was made more enthusiastically by Ralph Waldo

34 An Essay Upon the Principles of Political Economy (New York: Theodore Foster,


1837), 31; Willard Phillips, A Manual of Political Economy (Boston: Hilliard,
Gray, Little, and Wilkins, 1828), 63; Address Delivered by William H. Seward, at
the Commencement of the Auburn and Owasco Canal (Auburn, N.Y.: H. Ivison
& Co., 1835), 6; John A. Dix, Sketch of the Resources of the City of New-​York
(New York: G. & C. Carvill, 1827), 41.
  29

The Land of Speculation 29

Emerson. “How did our factories get built?” Emerson asked. “How
did North America get netted with iron rails, except by the impor-
tunity of these orators who dragged all the prudent men in? Is party
the madness of many for the gain of a few? This speculative genius is
the madness of a few for the gain of the world. The projectors are
sacrificed, but the public is the gainer.”35
To encourage investment (which was universally acknowledged to be
good) while discouraging speculation (which was often said to be bad),
one would need to distinguish between the two. Contemporaries sug-
gested several ways to draw the line separating investment and industry
from speculation and gambling. One possibility was to draw a line be-
tween investors who had a long-​term interest in the success of a venture
and those who had no such interest but were merely hoping to sell after
a short-​term rise in the value of an asset. Alexander Hamilton seems to
have been groping toward this distinction between two types of holders
of the national debt in its early bubbly days. “’Tis time there should be
a line of separation,” he insisted, “between respectable stockholders and
dealers in the funds, and mere unprincipled Gamblers.”36 Hamilton’s
meaning was not entirely clear, but his separation of “stockholders”
from “Gamblers” suggests a difference between those who invested in
the federal government out of patriotic motives, or at least an interest
in the long-​run welfare of the United States, and those who invested in
the debt just to make money in the short run.
Another distinction that could be drawn was between people who
bought an asset because they actually wanted to use the asset itself
and those who bought the asset only because they hoped to be able
to sell it later for a higher price. Congressman Gulian Verplanck
drew this distinction in his treatise on the law of contracts. In his

35 Francis Wayland, The Elements of Political Economy (New York: Leavitt, Lord &


Co., 1837), 357; “Manufacturing Corporations,” American Jurist 2 (1829): 105;
“The Joint-​Stock Companies,” Museum of Foreign Literature, Science, and Art
7 (1825): 154; Ralph Waldo Emerson, “Wealth” (published 1860, but written
1851–​52), in The Complete Essays and Other Writings of Ralph Waldo Emerson,
ed. Brooks Atkinson (New York: Modern Library, 1940), 698.
36 Alexander Hamilton to Philip Livingston, 2 Apr. 1792, in Harold C. Syrett,
ed., The Papers of Alexander Hamilton (New York: Columbia University Press,
1961–​81), 11:218–​19.
30

30 Speculation

view, speculation was “where purchases are made of lands peculiarly


situated, or goods are bought, not with reference to present use or
profit, but with relation to a future, and anticipated state of the
market.”37 Someone who bought land and actually settled on it
would not be a speculator, even if he hoped the land’s value would
rise; a speculator was someone who bought land solely in the hope
of a rise in value.
A different strategy was to distinguish between different kinds of
transactions, the necessary and the unnecessary. Usually, reasoned the
former broker William Armstrong, “the business of buying and selling
stocks is legitimate and necessary.” That was because “in the ordinary
course of events, causes too numerous to enumerate have an unavoid-
able tendency to make them change hands more or less, as different
circumstances which influence men in making or withdrawing invest-
ments arise.” But he felt very differently about “those transactions called
time operations,” or what today would be called futures, agreements to
buy stock at a specific price at a specific time in the future. “They are
essentially the same as so many bets upon what the future will bring
forth,” Armstrong argued, “for hardly one share out of fifty which is
sold on time is actually transferred, the contract being ‘settled’ by one
party paying to the other the ‘difference’ between the price specified at
the time of the sale and the market price at the time the stock is to be
delivered.”38
Yet another way to draw the line was to consider the degree of
risk: speculation was fine up to a point, but not when it became too
hazardous. “In all trade there must be speculation to a certain extent;
it is the very essence of commerce,” acknowledged one writer, “but
reckless gambling, in matters of trade, is as injurious, as in horse-​
racing, the hazard-​table, or cards.” Crèvecoeur’s American Farmer
praised the whalers of Nantucket for observing this distinction. “Many
voyages do not repay the original cost of fitting out,” he noted, but
the whalers “bear such misfortunes like true merchants, and, as they

37 “The Downfall of Speculation,” New-​Yorker 8 (1839): 57; Gulian C. Verplanck,


An Essay on the Doctrine of Contracts (New York: G. & C. Carvill, 1825), 109.
38 William Armstrong, Stocks and Stock-​Jobbing in Wall Street (New York: New York
Publishing Co., 1848), 11–​12.
  31

The Land of Speculation 31

never venture their all like gamesters, they try their fortunes again;
the latter hope to win by chance alone, the former by industry, well-​
judged speculation, and some hazard.” The same line was drawn by
the commercial moralist John Frost, who sought to confine the term
speculation to “incurring extensive hazards of any kind in the hope of
extensive emolument; in short, to whatever is foreign to the proper
business of the individual.”39 On this view, there was nothing wrong
with the inherent risks of a conventional business venture. It was only
when the risk was larger than normal that it should be condemned as
overly speculative.
A final possibility was to distinguish between positive-​sum and zero-​
sum transactions, and to condemn only the latter. “There are two ways
of acquiring wealth, not only essentially different, but as opposite to
each other as east is opposite to west,” the Mechanics’ Magazine lec-
tured its younger readers. “One of these ways may be properly denomi-
nated enterprise, the other speculation. The first of these ways creates the
wealth it accumulates by bringing into existence the articles of which it
is composed, or by increasing the value of articles which existed before;
the other draws the wealth, generally, by some kind of deception or
delusive practices, out of the possession of its right owner, without in-
creasing its value, or adding anything to the public stock.” The maga-
zine had no doubt which was better. “The first of these modes is honest
and highly laudable,” readers were advised, but the second is “wholly
dishonest, and subversive to the peace and happiness of mankind, be-
cause in the same ratio that it makes one richer it makes others poorer.”
The Princeton Review drew the same distinction, between “normal
trade,” in which the trader was compensated for performing some
useful function like transporting a commodity from the producer to
the consumer, and speculation, in which the trader performed no such
function but merely hoped to profit from a change in the price. Even
where they were equally risky, the risks of normal trade were inherent

39 “ ‘Nothing Venture, Nothing Have,’ the Maxim of the Speculator and


the Merchant,” Merchants’ Magazine 30 (1854):  517; J.  Hector St. John de
Crèvecoeur, Letters from an American Farmer (1782), ed. Susan Manning
(Oxford: Oxford University Press, 1997), 111; John Frost, The Young Merchant
(Boston: George W. Light, 1840), 157.
32

32 Speculation

in a commercial society, but speculation was “little else than simple


unmitigated gambling.”40
But many concluded that such definitional efforts were fruit-
less: there was just no way to distinguish useful investment from harm-
ful speculation. “A merchant is in some sense a gambler,” acknowledged
a correspondent in Maine. “If his chance be unfortunate he will specu-
late again, hoping that his ‘luck will turn’!” As the English economist
Thomas Corbet put it, “trade and speculation seem to be, to a certain
extent or in a certain manner, unavoidably connected.” Risk was simply
an inherent part of commercial life. “It is as difficult to tell at what
point safe and laudable enterprise leaves off, and speculation begins, as
it would be to mark off the precise line where delicate India ink shading
begins to lighten,” a Massachusetts writer observed.41 All investment
was a kind of speculation.
And who exactly were these oft-​condemned speculators? “Gentlemen
talk of these persons as if they were the veriest brokers and stockjob-
bers in the world,” Representative Charles Ingersoll of Pennsylvania
noted during the debates over the Bank of the United States. “But
no misconception could be more unfounded.” They were mostly or-
dinary people who had prudently invested their savings in the hope
of a modest return. If one visited the bank on the day dividends were
paid, Ingersoll argued, one would “find the widow and the orphan, the
aged and the infirm, as well as the wealthy and the competent, wait-
ing for their shares; some of them for small sums.” There was no way
to condemn speculators without also condemning the honest mer-
chant and the frugal investor. “If, Sir, we forget the horrors that Envy
and Ignorance have added to the sound of Speculation,” thundered
one letter writer to a Philadelphia newspaper, “we shall find child-
ish Philosophers, and beardless Moralists, railing at Contracts, and

40 “Hints to Young Men, or Enterprise and Speculation Contrasted,” Mechanics’


Magazine 5 (1835):  5; “Ethics and Economics of Commercial Speculation,”
Biblical Repertory and Princeton Review 41 (1869): 237; for the same distinc-
tion, see “Trade and Speculation,” Independent, 10 Feb. 1870, 8.
41 Eastern Argus [Portland, Me.], 5 Dec. 1815, 1; Thomas Corbet, An Inquiry Into
the Causes and Modes of the Wealth of Individuals (London: Smith, Elder & Co.,
1841), 159; The Pressure and Its Causes (Boston: Otis, Broaders & Co., 1837), 26.
  33

The Land of Speculation 33

execrating Commerce.”42 Speculation simply could not be separated


from ordinary business.
John Sergeant, the Philadelphia congressman and diplomat, spoke at
length on this theme in 1819, when the House was yet again debating
the bank issue. “Speculation and speculators, sir, are terms of very vague
import, and of very extensive application,” he pointed out. “There are
speculators of many kinds—​there are speculators in lands—​there are
speculators in merchandise—​there are speculators in manufactures—​
there are speculators in stocks; the variety is infinite, and in no country
upon earth greater than in this. Every thing about us invites to specula-
tion.” How could one praise some people who took commercial risks
and condemn others? “I should like, then, to know,” Sergeant contin-
ued, “in what the discrimination consists, which makes one kind of
speculation offensive, and another innocent, if both are permitted by
law, and neither unfairly or fraudulently conducted. What is the differ-
ence between speculating in land, and speculating in merchandise, or
the stocks?”43
Indeed, mused one anonymous writer, if one took the broad view,
life was full of speculation, because life was full of risk. “The wife spec-
ulates with her husband when she constrains him by persuasion and
tenderness, to convey her all his estate,” he reasoned, “and when the
husband in like circumstances, gains a like object, he speculates upon his
wife.” And that was not all. “The girl who, from 16 to 25, devotes herself
to the acquisition of a good husband with a fortune, and finally obtains
him, speculates well.” “Some seem to suppose that it is peculiar to Wall-​
street,” agreed the Cincinnati Merchants’ Magazine, but the magazine
insisted “that those seeking the gain of money are not the only specula-
tors: but that men, women, and children, are all endeavoring to acquire
something of which they are not now possessed—​in fact, that we are all
speculators.” When life itself was a gamble, how could one distinguish
the good risks from the bad? “As well might you attempt to stop the

42 Legislative and Documentary History of the Bank of the United States (Washington,
D.C.: Gales and Seaton, 1832), 519; General Advertiser [Philadelphia], 23 Mar.
1792, 2.
43 Selected Speeches of John Sergeant, of Pennsylvania (Philadelphia: E. L. Carey &
A. Hart, 1832), 164.
34

34 Speculation

flow of the sea, or direct the course of the whirlwind,” despaired the
playwright and journalist Mordecai Noah, “as to say who shall, or who
shall not speculate.”44
The difficulty of distinguishing gambling from commerce was not
necessarily a reason to refrain from regulating speculation. “If there
be a want of precision in defining the thing condemned, it arises from
the very nature of the case,” one editorialist reasoned. “It is one which
admits of no precise and infallible rules. There are instances where the
evil is so palpable, that every eye can see it; but between this point
and the point of departure from legitimate business, there is an infi-
nite variety of abuses.” “Let us not,” another agreed, “confound reckless
speculation and commercial pursuits in the same category.” While there
might be a blurry line between them, “they are in truth ‘wide as the
poles asunder.’ ” If “it would perhaps be difficult, in some cases,” to dis-
tinguish legitimate from illegitimate speculation, a third insisted, “there
are nevertheless certain broad distinctions between the two classes of
operations.”45
But others concluded that even if some speculation should, in prin-
ciple, be prohibited, it would be impossible for the law to ban the bad
without also sweeping away the good. “Speculation in paper hath been
a kind of gambling,” one observer lamented, but there was nothing
the law could do about it. “It is not possible,” he explained, for the
government to inspect “all the private bargains of jockies, speculators
and dunces—​No rule of right can be ascertained” for separating the
permissible from the impermissible. And even if it could be done, he
continued, the cost would far exceed the benefits: “The expence of at-
tempting it would bring a new debt on the country, ten fold greater
than all the savings which can be made.” Speculators would have to be
left to their own devices. “One quarter of the bargains made are detri-
mental to one, and some of them to both parties,” he concluded. “They

44 Weekly Museum, 19 May 1792, 1; “Speculation,” Merchants’ Magazine and


Commercial Review 24 (1851):  781; M.  M. Noah, “Speculation,” New-​York
Mirror 17 (1839): 186.
45 National Era, 29 July 1847, 2; “Speculation and Commerce,” Flag of Our
Union 12 (1857): 133; “Speculation—​Legitimate and Illegitimate,” Merchants’
Magazine 58 (1868): 297.
  35

The Land of Speculation 35

were made in folly, but for national policy particularly to inspect them
would be greater folly, and a most unmanageable and endless business.”
People entered into all kinds of transactions in all kinds of commodi-
ties at all kinds of prices, another agreed. How could one ever figure
out which transactions were too speculative? “Even at this time there
is an extensive speculation all over the country in the article of bread
stuffs, and advantage is taken of the necessities of consumers to demand
exorbitant prices,” he pointed out. “Does the law afford any relief? And
what relief could it afford? Would you annul contracts between land-
lord and tenant, because the rents were excessive?”46 If the law could
not draw a line between the speculation that should be allowed and the
speculation that should be prohibited, the wisest course was to refrain
from limiting speculation at all.
Some took the argument a step further. Not only was speculation
inseparable from ordinary commercial life, they contended, but specu-
lation was also a positive good in its own right. One early proponent of
this view was Alexander Hamilton, who had to justify speculation while
responding to criticism of his financial program, a key part of which
was the issuance of debt securities that were the primary speculative
assets in the bubble and crash of 1791–​92. Just after the crash, Thomas
Jefferson wrote a long letter to George Washington summing up his
complaints about Hamilton’s policies. One of Jefferson’s critiques was
his characteristically negative view of the buying and selling of securi-
ties. “All the capital employed in paper speculation is barren and useless,
producing, like that on a gaming table, no accession to itself,” Jefferson
argued. The capital spent on speculation “is withdrawn from commerce
and agriculture where it would have produced addition to the common
mass,” he continued. Instead, “it nourishes in our citizens habits of vice
and idleness instead of industry and morality.” Washington promptly
wrote to Hamilton to hear his response. (Although Washington did not
specify whose critique he was quoting—​he referred only to his efforts
“to learn from sensible & moderate men … the sentiments which are

46 Connecticut Courant, 28 Jan. 1790, 4; A Familiar View, of the Operation and


Tendency of Usury Laws (New York: John Gray, 1837), 53.
36

36 Speculation

entertained of public measures”—​Hamilton might well have guessed


the source of the criticism.)47
Hamilton responded with one of the earliest sustained defenses of
speculation in the United States. The buying and selling of paper did
not remove capital from the productive economy, he argued. Rather,
it added capital to the economy. “It is true that the Capital, that is the
specie, which is employed in paper speculation, while so employed, is
barren and useless,” Hamilton began. “But the paper itself constitutes
a new Capital, which being saleable and transferrable at any moment,
enables the proprietor to undertake any piece of business as well as
an equal sum in Coin. And as the amount of the Debt circulated is
much greater than the amount of the specie which circulates it, the new
Capital put in motion by it considerably exceeds the old one which is
suspended.” Paper securities made investment much easier, Hamilton
added, because they allowed people to convert assets that were not easily
invested into assets that were. “Every thing that has value is Capital—​an
acre of ground a horse a cow or a public or a private obligation; which
may with different degrees of convenience be applied to industrious
enterprise,” Hamilton advised Washington. “That which, like public
Stock, can at any instant be turned into money is of equal utility with
money as Capital.” If Jefferson’s zero-​sum view of the economy were
correct, Hamilton continued, money would be scarcest in the societies
with the most securities trading. In fact, the opposite was true. “Let it
be examined whether at those places where there is most debt afloat and
most money employed in its circulation, there is not at the same time a
greater plenty of money for every other purpose,” Hamilton noted. “It
will be found that there is.” He conceded that the limited number of
people who spent significant portions of their time engaged in specula-
tion might be diverting their hours from more productive pursuits, but
that loss was swamped, he argued, by the gain to the economy from the
capital they injected into it. “Jobbing in the funds has some bad effects
among those engaged in it,” he concluded. “But if the proposition be

47 Thomas Jefferson to George Washington, 23 May 1792, in Papers of Thomas


Jefferson, 23:537; George Washington to Alexander Hamilton, 29 July 1792, in
Papers of Alexander Hamilton, 12:129–​31.
  37

The Land of Speculation 37

true, that Stock operates as Capital, the effect upon the Citizens at large
is different. It promotes among them industry by furnishing a larger
field of employment.”48
Hamilton was defending a broad financial program, of which pro-
viding the material for speculation was just a part, but others made
the same point while offering more specific defenses of speculation in
its own right. Speculation did not reduce the amount of capital in the
productive economy, one correspondent insisted, but rather increased
that amount. “Speculators neither bury nor export their money,” he
pointed out. “On the contrary, they continually circulate it.” To be
sure, speculation was liable to abuse, but so was anything. “Honor and
religion—​everything that is useful, noble or virtuous, may be abused.
Ought they therefore to be condemned or avoided? No, certainly.” And
while speculators might harm themselves, argued the Reverend Cyrus
Mason of New  York University, by spurring economic activity, they
indirectly helped others. Mason was speaking just after the bursting
of the mulberry tree bubble of the 1830s. (Silkworms live in mulberry
trees, and the prospect of a domestic silk industry gave rise to overin-
vestment in mulberry trees.) “When the fever of speculation has gone
by, and the laughers at speculation have had their turn, the mulberry
trees will be found growing in vast numbers all over the face of this
country,” Mason affirmed. “And when there has been time to mature
their growth, the business will be found attractive and profitable.” He
looked back at the Merino sheep bubble of a few years earlier, when the
collapse of wool prices had left farmers throughout the northern states
with a glut of sheep. “The fever had its course, and the laughers their
fun,” Mason recalled, “but the improved breed of sheep remains, the
public revenue from imported fine wool and cloth has been reduced,
and a rich blessing to our country has been the result.”49
A second common defense of speculation was that it provided the
useful service of deepening markets. It was widely recognized in the

48 Alexander Hamilton to George Washington, 18 Aug. 1792, in Papers of


Alexander Hamilton, 12:246–​47.
49 General Advertiser [Philadelphia] 29 Mar. 1792, 2; Cyrus Mason, The Oration
on the Thirteenth Anniversary of the American Institute (New York: Hopkins &
Jennings, 1840), 23.
38

38 Speculation

late eighteenth century that speculators in any kind of asset made non-
speculators more willing to hold that asset, and thus made the asset
more valuable, because the speculators stood ready to buy and sell
at any time. “As to Stock-​jobbers he saw no reason for the censures
thrown upon them,” Elbridge Gerry remarked at the Constitutional
Convention. “They keep up the value of the paper. Without them there
would be no market.” Robert Morris, one of the leading speculators of
the era, agreed that “even if it were possible to prevent Speculation, it is
precisely the Thing which ought not to be prevented; because he who
wants Money to commence, pursue or extend his Business, is more
benefited by selling Stock of any kind (even at a considerable Discount)
than he could be by the Rise of it at a future Period.”50 On this view,
speculators reduced the risk of investment, because an investor could
have some confidence that someone would step up to buy in the event
he needed to sell.
Another justification of speculation became common in the first half
of the nineteenth century. Speculators, some argued, stabilized prices,
by buying when the market was low and selling when it was high.
“When the price has very much declined,” Willard Phillips reasoned,
“those who suppose it to be verging on its lowest point, and about to
seek again its natural level, the mean cost, naturally use their funds and
credit in purchasing, to take advantage of the change.” By boosting
demand, speculators reversed the decline in prices. “Speculations may
have the effect of limiting the vibrations of price within narrower ex-
tremes,” Phillips continued, “for if a great quantity be withdrawn from
a declining market, it may cause a rise, and the same quantity being
thrown onto a rising market may accelerate the change to a decline
of the price.” Speculators thus helped their fellow participants in the
economy. “As far as speculation has these effects it is useful,” Phillips
concluded, “since extremes and irregular fluctuations of price, are fol-
lowed by embarrassments and bankruptcies, by which the community
at large suffers.” By the same mechanism, others suggested, speculators

50 Notes of Debates in the Federal Convention of 1787 Reported by James Madison


(New York: Norton, 1987), 529; Robert Morris to John Hanson, 29 July 1782,
in E. James Ferguson et al., The Papers of Robert Morris, 1781–​1784 (Pittsburgh,
Pa.: University of Pittsburgh Press, 1973–​99), 6:70.
  39

The Land of Speculation 39

smoothed out fluctuations in supply, by bringing commodities to


market when they were scarce and withholding them when they were
plentiful. Without speculators, argued Alonzo Potter of Union College,
the supply of grain and other goods “would be necessarily so irregular as
to occasion profusion and waste at one period, and dearth and famine,
as their consequence, at another.”51
In any event, others contended, speculators were merely responding
to public demand for their services. “When a financial crisis overtakes
the community,” the Atlantic observed, “we hear much and sharp cen-
sure of all speculation. Speculators, one and all, are forthwith consigned
to an abyss of obloquy.” But why were speculators buying and selling
in the first place? Because of “that same public who denounce specu-
lation in one breath, and in the next clamor for goods at low prices,
and force the jobber into large stores and large sales at small profits as
the indispensable condition of his very existence. Those who thus rail
at speculation are generally quite unaware that their own inexorable
demand for goods at low prices is one of the principal efficient causes
of that of which they complain.”52
On this view, speculation was quite different from gambling. Like
any other business, it was simply a matter of giving customers what
they wanted. As one defender of the “fluctuations of mercantile specu-
lation” put it, “trade has all the fascination of gambling, without its
moral guilt.” Speculators did not force anyone to buy or to sell, and
they could not raise or lower prices at will; like anyone else, they had to
accept whatever prices the market produced. “There is a palpable mis-
take in the assumption” that speculators inflated prices, one observer
insisted. “The real cause is an actual scarcity.” And when speculators
profited from scarcity, why should they be attacked? “The individual
who bought one thousand barrels of flour last August at eight or nine
dollars per barrel, and now sells it or refuses to sell it for thirteen, has

51 Willard Phillips, A Manual of Political Economy (Boston: Hilliard, Gray, Little,


and Wilkins, 1828), 50–​51; Henry Vethake, The Principles of Political Economy
(Philadelphia: Nicklin & Johnson, 1838), 233; Alonzo Potter, Political Economy
(New York: Harper & Brothers, 1840), 163.
52 “A Dry-​Goods Jobber in 1861,” Atlantic Monthly 7 (1861): 204–​5.
40

40 Speculation

violated no law of society, trade, or government, and may have acted


uprightly throughout.”53
Defenders of speculation attributed critiques of the practice to the
ignorance of their fellow citizens. “Speculation,” suggested the journal-
ist Richard Hildreth, was simply one “of those verbal reasons by which
practical men account for things they do not understand.” Another
proponent concluded that “when judicious minds condemn, en masse,
speculators and speculation, I imagine that it is always for want of suf-
ficient examination and reflection on the subject.” On the other side,
critics of speculation interpreted justifications of the practice as self-╉
interested sophistry. It was hardly a coincidence that someone like
Robert Morris would defend speculators, an anonymous opponent
sneered. He was “engaged in those deep speculations” himself. By the
middle of the nineteenth century, Americans had engaged in decades of
debate over the merits of speculation. Both sides had ample arguments
to draw upon.54

A Practicable Line of€Discrimination


Of course, this was rarely a debate about speculation in the abstract.
Arguments for and against speculation were almost always put forward
in the context of individual speculative episodes, specific regulatory
proposals, or particular legal questions. Should the legislature prohibit
the most speculative kinds of transactions? Should speculative contracts
be enforceable in court? Questions like these came up regularly in the
early republic. There was always an occasion to think about the pros
and cons of speculation.
The most fundamental of these issues involved the legality of gam-
bling. English law had allowed most forms of gambling, with ad hoc
exceptions. “All wagers are legal,” explained the barrister John Disney

53 Village Register and Norfolk County Advertiser, 27 Apr. 1821, 1; “High Prices—╉
Speculation,” New-╉Yorker 2 (1837): 345.
54 Richard Hildreth, The History of Banks (Boston: Hilliard, Gray & Co., 1837),
77; National Era, 29 July 1847, 4; “Lucius” to Robert Morris, 12 Mar. 1783,
in Papers of Robert Morris, 7:561; Scott A. Sandage, Born Losers: A History of
Failure in America (Cambridge, Mass.: Harvard University Press, 2005).
  41

The Land of Speculation 41

in his treatise on the law of gaming, “excepting those which tend to a


breach of the peace;—​or are contra bonos mores;—​or tend to hurt the
character of, or expose to ridicule, a third person;—​or are against the
rules of sound policy and morality;—​or are contrary to a positive enact-
ment of the law.” The United States inherited this rule at independence.
In several early reported cases, American courts enforced contracts for
wagers on the outcome of horse races. When the federal government
tried to prosecute the proprietor of a gaming house in the District of
Columbia, the court stopped the prosecution. Gambling, the judges
told the prosecutor, was simply not illegal. On the other hand, courts
refused to enforce wagers they deemed contrary to public policy. As
New York’s chief judge James Kent put it, “a wager contract is void,
if it be against the principles of public policy, equally as if it contra-
vened a positive law.” For example, when two residents of Kent County,
Maryland, bet on an election for sheriff, Maryland’s General Court
refused to order the loser to pay the winner. “The election of a sheriff
is of great importance to the community,” the court explained, “and
ought to be free from corrupt and undue influence.” A New York court
took an equally dim view of a suit brought by the winner of a wager on
the state’s 1807 gubernatorial election. On the other hand, the Illinois
Supreme Court was willing to enforce a wager between two Illinois
residents on how many votes William Henry Harrison would receive in
Kentucky in the 1840 presidential election, on the theory that the bet
could not corrupt the election, because Illinoisans would be unable to
influence events in Kentucky. As the New York lawyer George Caines
summed up the early American law of gambling, it was “a system in
which wagers are merely tolerated but not favoured.”55
The law’s toleration for gambling diminished over time. Virtually
from the beginning, states passed statutes banning particular kinds of

55 John Disney, The Law of Gaming, Wagers, Horse-​Racing, and Gaming-​Houses


(London:  J.  Butterworth, 1806), 2; Williams v.  Cabarrus, 1 N.C. 54 (1793);
Barret v. Hampton, 2 Brev. 226 (S.C. 1807); United States v. Willis, 28 F. Cas.
698 (C.C.D.C. 1808); Mount & Wardell v. Waite, 7 Johns. 434 (N.Y. Sup. 1811);
Wroth v. Johnson, 4 H. & McH. 284 (Md. 1799); Bunn v. Riker, 4 Johns. 426
(N.Y. Sup. 1809); Morgan v. Pettit, 4 Ill. 529 (1842); Davy v. Hallett, 3 Cai. R. 16
(N.Y. Sup. 1805).
42

42 Speculation

gambling. For instance, Delaware prohibited wagers on horse racing in


1786, and Virginia banned the card game of faro not long after. Within
a few decades these narrow prohibitions had become so numerous and
broad that they coalesced into a new general rule. Gambling had once
been lawful except where specifically banned, but by the second half of
the nineteenth century it tended to be unlawful except where specifi-
cally allowed (for example, some states had lotteries, and others allowed
horse racing). Gambling hardly disappeared, but without the benefit of
the formal legal system, nineteenth-​century gamblers were increasingly
left to norms of honor and the use of violence for the enforcement of
their contracts.56
The law governing commercial speculation took the opposite
trajectory—​from ambivalence in the late eighteenth century to ap-
proval, with certain exceptions, by the early nineteenth. In 1778, for
example, Nathan Wright agreed to sell 1,800 acres of land straddling
the Maryland–​Delaware border to Isaac Perkins. The contract specified
that Wright would only convey the land once Perkins had made three
annual payments in continental currency. The currency depreciated
sharply over the next two years, so although Perkins diligently made
all three payments, they turned out to be worth much less than Wright
had expected. Wright refused to convey the land to Perkins, because
in his view he had not been fully paid. As he testified, “he is willing to
convey the said land upon receiving the real value thereof, agreeably to
the true meaning of the contract.” Perkins’s suit came before Maryland
Chancellor Alexander Hanson, who explained that the case raised a
difficult question. On the one hand, he acknowledged, any contract
involving payment in the future could be said to involve speculation,
because the medium of payment might fluctuate in value between the
contract date and the payment date. On the other hand, some contracts

56 State v. Blackiston, 2 Del Cas. 229 (1786); Commonwealth v. M’Guire, 3 Va.


119 (1798); The Development of the Law of Gambling: 1776–​1976 (Washington,
D.C.: National Institute of Law Enforcement and Criminal Justice, 1977), 65–​
74, 142–​56, 246–​63; Ann Fabian, Card Sharps, Dream Books, & Bucket Shops:
Gambling in 19th-​Century America (Ithaca, N.Y.: Cornell University Press,
1990); John M. Findlay, People of Chance: Gambling in American Society from
Jamestown to Las Vegas (New York: Oxford University Press, 1986).
  43

The Land of Speculation 43

were simply too speculative to be enforced, and he decided that this was
one of them. Hanson concluded that a “court ought not to give its aid
towards enforcing the specific performance of any speculating engage-
ments.” Rather than ordering Wright to convey the land to Perkins, he
ordered Wright to give Perkins his money back. “A speculating contract
for continental money,” Hanson explained in a case involving a similar
transaction, “is such a contract as this court has again and again said it
will not enforce.”57
Other early American judges, however, took a more benign view of
speculative transactions. In November 1791, John Wilkes agreed to buy
US “stock” (what we would today call bonds) from Robert Gilchreest
ten months later, in September 1792. The stock’s value dropped in the
interim, a period encompassing the crash of 1792, so when Gilchreest
delivered it to Wilkes in September, Wilkes refused to pay. The
Pennsylvania Supreme Court had little trouble finding the contract
enforceable. “The sale of stock is neither unlawful nor immoral,” the
court declared. “It is confessed, that an inordinate spirit of speculation
approaches to gaming, and tends to corrupt the morals of the people.
When the public mind is thus affected, it becomes the legislature to
interpose. But we have no such law at present.” The risk that prices
might rise or fall was inherent in virtually all transactions, the court
noted. “Call the 6 per cent. stock so many bushels of wheat,” the court
asked: “if it had fallen in price on the day of delivery, and the vendor
was then ready and willing, and offered to perform his contract in all
parts, ought not the principal or his surety to make him full compensa-
tion? If the wheat had risen in price, would not the adverse party be
enabled to get like compensation, in case the vendor had receded from
his bargain?”58 If risk was a normal part of commercial life, judges had
no basis for distinguishing lawful transactions from those too specula-
tive to enforce.
This became the prevailing view in the early nineteenth century. In
1803, for example, Virginia’s Supreme Court of Appeals considered the

57 Perkins v. Wright, 3 H. & McH. 324 (Md. Ct. App. 1793); Hopkins v. Stump, 2
H. & J. 301 (Md. Ct. App. 1808) (Hanson decided this case in 1804 and died
in 1806; the Court of Appeals affirmed his decision in 1808).
58 Gilchreest v. Pollock, 2 Yeates 18 (Pa. 1795).
44

44 Speculation

enforceability of what one of its members called “a mere speculation


upon the paper currency of the country.” Willis (the parties are identi-
fied only by their last names) had agreed to pay Griffin £25,000 in paper
money in 1780 and 1781, in exchange for Griffin’s promise to pay Willis
£2,500 in specie ten years later, in 1790. “The contract in this case, was
founded upon speculation on both sides,” one of the judges acknowl-
edged. “Griffin thought, the present use of the money would be advan-
tageous to him; and Willis, that it would be more beneficial to receive
the specie at a distant day.” Yet the judges all agreed that the contract
was nevertheless enforceable. The Maryland Court of Appeals reached
a similar decision in a case involving a complex contract to speculate in
US bonds, an arrangement that required one party to pay the other a
sum that depended on the difference in the bonds’ value on two dates.
The defendant contended that the contract was unenforceable, because
it was merely a wager. The court disagreed. It accepted the plaintiff’s
argument that “almost every contract depends on a contingency; and if
this is considered a wager, it will put an end to all contracts.”59
The Virginia judge Dabney Carr, Thomas Jefferson’s nephew,
summed up this way of thinking in an 1819 case that was just one tiny
part of the litigation arising from the bankruptcy of Robert Morris two
decades before. (“Morris was known to every one, as a man of great
talents, particularly for speculation,” Carr observed. “One of those
Giants in speculation never falls alone; all connected with him feel the
shock:—​and those within his immediate vortex are generally drawn
down along with him.”) The transaction at issue involved the sale of
notes signed by Morris. Both the seller and the buyer learned only after
the purchase that Morris was bankrupt and that the notes were thus
worth very little. “It is said,” as Carr described the buyer’s argument for
why the contract was unenforceable, “that this was a speculating con-
tract, with which this Court ought not to interfere.” But Carr dismissed
this contention. “I can not see how a practicable line of discrimination
can be drawn between this sort of commerce, and that which we see
every day carried on in Bonds, Bills and Notes,” he reasoned. “A man

59 Brachan v. Griffin, 7 Va. 433 (1803); Ridgely v. Riggs, 4 H. & J. 358 (Md. Ct.
App. 1818).
  45

The Land of Speculation 45

wanting to raise money, takes these evidences of debt, as he would any


other species of property, to the market, and sells them for what they
will bring; sometimes more, sometimes less.”60 If the value of any asset
could rise or fall, all commercial transactions were speculative. There
was no principled way for judges to distinguish between the good and
the bad.
In certain contexts, however, legislatures stepped in to ban prospec-
tively categories of transactions the judges would not annul in retro-
spect. The most important of these statutes was New York’s 1792 “Act
to prevent the pernicious Practice of Stock-​Jobbing.” After the crash
of 1792, when concern about speculative stock trading was at its peak,
New York declared void all contracts for the sale of securities that the
ostensible seller did not own at the time of contract. This provision was
copied from a 1734 English statute that had likewise been intended to
curb speculative trading. The idea was to put a stop to a common form
of speculation, in which two parties would agree to sell stock at a par-
ticular price on a particular date in the future. When that date arrived,
the seller would not in fact transfer the stock to the buyer. Rather,
one party would simply pay to the other the difference between the
contract price and the stock’s actual price. New York’s statute remained
in force until 1858. Massachusetts had a similar law from 1836 to 1910.
Neither statute had much effect. Traders could no longer use the courts
to enforce these transactions, but the transactions remained common
nevertheless, enforced by the Stock Exchange and more diffusely by the
loss of reputation that would be suffered by a trader known to breach
his contracts.61
Legislatures sporadically considered, and occasionally enacted,
other limits on speculation. In 1793 and again in 1794, Congress de-
bated whether to tax securities transactions, in part to raise revenue
and in part to dampen speculation. The idea was well received in
Congress, to the dismay of Fisher Ames of Massachusetts. “This dis-
tresses Hamilton exceedingly, and well it may,” Ames remarked. “To

60 Armstrong v. Hickman, 20 Va. 287 (1819).


61 Laws of the State of New-​York (New York: Francis Childs and John Swaine, 1792),
66; Stuart Banner, Anglo-​American Securities Regulation: Cultural and Political
Roots, 1690–1860 (Cambridge: Cambridge University Press, 1998), 171–​75.
46

46 Speculation

begin to tax the public debt, when we are afraid to tax snuff, is a bad
omen.” This proposed tax was approved by a committee of the whole
House, but it was then added to a long list of other taxes that were
rejected as a package. In 1834, a committee of the New York Assembly
condemned “the great and palpable evils attendant upon the exist-
ing system of buying and selling stocks in the city of New-​York,” a
system that in the committee’s view “gives rise to a spirit of gambling
which is carried out to an alarming extent.” Nothing came of this
report. But other states imposed limits on speculative transactions.
Pennsylvania and Maryland voided all futures contracts for the sale of
securities where the delivery date would be more than five days after
the contract date. Several states required shareholders to own stock
for a certain minimum period before selling it, to encourage long-​
term investment while discouraging short-​term speculation. Some
states restricted the amount of shares that could be purchased in a
single day, to make it harder for one person to buy large quantities.
The federal government repeatedly tried to discourage speculation in
public land, through such methods as limiting the amount a single
person could purchase and requiring grantees to reside on the land.62
These were all efforts to ban a narrow range of the most speculative
transactions without tramping too heavily on commercial activity
considered more legitimate.
American law governing risky transactions thus distinguished, at
an early date, between gambling and speculation. Transactions classi-
fied as wagers were presumed unenforceable, with certain exceptions.
Transactions classified as speculation, by contrast, were presumed en-
forceable, again with certain exceptions. As historians have pointed out,
the difference between the two categories was never completely stable,
and it drew in part on a class distinction between the socially accept-
able risks taken by respectable members of society and the less polite

62 Seth Ames, ed., Works of Fisher Ames (Boston: Little, Brown and Co.,
1854), 1:141; Documents of the Assembly of the State of New-​York, Fifty-​
Seventh Session (Albany, N.Y.: E. Croswell, 1834), No. 339; Banner,
Anglo-​American Securities Regulation, 222–​2 6; Paul W. Gates, History of
Public Land Law Development (Washington, D.C.: Government Printing
Office, 1968).
  47

The Land of Speculation 47

risks taken by those on the margins.63 But the difference between gam-
bling and speculation was not entirely a matter of class, and it was not
entirely void of substance. It was based on a perception, apparently
widely shared among the managers of the legal system, that risk was
tolerable as a means to some greater end, but that it was not a worth-
while end in itself. In the transactions proscribed as “wagers”—​bets on
horse races, card games, and the like—​there was no purpose beyond
the participants’ enjoyment of the risk itself. The transactions called
“speculation,” on the other hand, involved some other useful societal
end, even if the risks associated with speculation were indistinguishable
from the risks associated with gambling. Speculation in trade made
society richer, even if it made half the speculators poorer. Speculation
in shares gave rise to valuable new enterprises, even if it beggared some
of the shareholders. The line between speculation and gambling was
often difficult to draw with precision, but this seems to have been the
intuition behind the view that some such line had to be drawn.
Insurance provides the clearest early example. Marine insurance was
already common by the late eighteenth century. It looked a lot like
gambling:  the insured would pay a small sum, and if the ship went
down, he would receive a large payoff from the insurer. As Judge Jacob
Peck of the Tennessee Supreme Court wondered, gaming “depends
upon hazard and chance; so do all insurances depend upon hazard
and chance. What is the difference?” Yet insurance also served a useful
purpose. Insurance was “much for the benefit and extension of trade,”
William Blackstone explained in his commentaries on English law, “by
distributing the loss or gain among a number of adventurers.”64
English law accordingly adopted the concept of the “insurable inter-
est,” which permitted people to buy insurance only if they themselves

63 Karen Halttunen, Confidence Men and Painted Women: A Study of Middle-​


Class Culture in America, 1830–1870 (New Haven, Conn.: Yale University
Press, 1982), 16–​20; Fabian, Card Sharps, 1–​11.
6 4 Hannah Atlee Farber, Underwritten States: Marine Insurance and the Making
of Bodies Politic in America, 1622–1815 (Ph.D. dissertation, University of
California, Berkeley, 2014); Edwin J. Perkins, American Public Finance and
Financial Services 1700–1815 (Columbus: Ohio State University Press, 1994),
282–​92; State v. Smith, 10 Tenn. 272 (1829); William Blackstone, Commentaries
on the Laws of England (Oxford: Clarendon Press, 1765–​69), 2:461.
48

48 Speculation

would bear a loss if the ship failed to arrive. Insurance policies on other
people’s voyages were unenforceable, because they “were very justly
considered as mere gaming or wager-​policies,” declared the barrister
James Park in his treatise on insurance. American law followed suit.
“Wager policies … though dressed up in the garb of insurance, are
in fact nothing more than gambling contracts,” huffed George Caines
of New York. A wager policy was one in which “the party insured has
not any interest in the thing underwritten.” When the question came
before the Massachusetts Supreme Court for the first time in 1806, the
court found it an easy one. Thomas Amory had purchased two different
insurance policies, from two different companies, on the same cargo
aboard a ship that was to sail from the Canary Islands to the United
States. On the way, the ship and its cargo were captured by the British
and taken to Jamaica. Armory recovered on his first policy. When he
tried to recover on the second, the second insurance company defended
on the ground that Armory had already been fully compensated, and
that his second policy was nothing but a wager that the ship would
never reach its destination. The court agreed. “It would seem a dis-
graceful occupation of the courts of any country, to sit in judgment
between two gamblers, in order to decide which was the best calculator
of chances, or which had the most cunning of the two,” sniffed Justice
Isaac Parker. “There could be but one step of degradation below this—​
which is, that the judges should be the stake-​holders of the parties.”
Chief Justice Francis Dana took the same view. “Wager policies are
injurious to the morals of the citizens,” he reasoned, and “tend to en-
courage an extravagant and particularly hazardous species of gaming.”
Insurance “ought to be reserved for the benefit of real commerce.”65
By allowing merchants to insure their own losses but not the losses
of others, judges were able to draw a line between commerce and gam-
bling. “An insurance amongst us, is a contract of indemnity,” reasoned
Judge Jasper Yeates of Pennsylvania. “Its object is, not to make a positive
gain, but to avert a possible loss. … A  policy therefore made without

65 James Allan Park, A System of the Law of Marine Insurances, 2nd American ed.
(Boston: Thomas and Andrews, 1800), 262; George Caines, Enquiry Into the
Law Merchant of the United States (New York: Abraham and Arthur Stansbury,
1802), 1:313; Amory v. Gilman, 2 Mass. 1 (1806).
  49

The Land of Speculation 49

interest, is a wager policy, and has nothing in common with insurance,


but name and form.” James Kent of New York put it rather more grump-
ily. “Wager policies ought not to be encouraged,” he snorted, “and it is
not pleasant that the time of the court should be occupied in discussing
them.”66 Insuring one’s own losses was like ordinary commercial specu-
lation; it was a method of shifting risk from one party to another, for the
benefit of all. One side might be the loser in any given transaction, but
all those transactions would net out positive for the broader economy.
Insuring other people’s losses was understood as gambling. Some would
win and some would lose, but no broader purpose would be achieved.
Insurers’ desire to distinguish wagers from legitimate commerce
was especially strong regarding life insurance, because of the unset-
tling aspect of betting that someone would die, and because of the
distinct possibility that the holder of a policy on someone else’s life
might take steps to make his death come sooner. When life insur-
ance first came on the American market in the early nineteenth cen-
tury, promoters were thus careful to distinguish it from gambling.
“It is very commonly imagined, that Life Insurance is of a nature
purely speculative,” conceded one insurance company—​“that it is
a mere gambling contract between the assurer and the assured.”
But nothing could be further from the truth, declared another. Life
insurance “does not rest upon uncertain theories or vague specula-
tions, as is by many supposed,” the company explained; “nor is it
founded upon calculations whether death will or will not happen;
for that event is certain; but is based upon the law of the aver-
age duration of life, and secures the benefits of a mean duration of
years, by rules and tables approximating so nearly to a certainty,
that they may be depended upon, in all calculations.” Life insur-
ance “has been supposed to have some affinity to gambling,” ac-
knowledged another promoter. “But the analogy is shallow. It may
more justly be considered a method of delivering human life from
the tyranny of chance.” When the president of a Utica bank de-
clared in the pages of the Merchants’ Magazine that “life insurance
possesses many of the elements of gambling,” he drew immediate

66 Pritchet v.  Insurance Company of North America, 3 Yeates 458 (Pa. 1803);
Clendining v. Church, 3 Cai. R. 141 (N.Y. Sup. 1805).
50

50 Speculation

rejoinders from correspondents who pointed out that banking re-


sembled gambling more than life insurance did. Life insurance “is,
or ought to be, if rightly managed, a science,” one insisted. The
“tables of the duration of life, are, to say the least, more reliable
than the chances of Banking in the United States.” The president of
the Mutual Life Insurance Company of New York assured readers
that “the application of the law of average, so far from giving it the
character of a gambling transaction, in reality goes far to equalize
among all connected with it, a participation in all the chances of
life.”67
Like older forms of insurance, life insurance took the form of a
wager, but it was a wager that served useful purposes. “It might be
the means of preventing many families of industrious tradesmen and
others, from becoming chargeable on the public charity and provide
the means of education to their orphan children,” a committee of the
Pennsylvania legislature recognized in 1812, when asked to charter the
state’s first life insurance company. Life insurance “might also assist
the aged infirm not blessed with an abundance of wealth to pass the
downward path of life with ease and comfort,” the committee added.
A Rhode Island newspaper concluded that “life insurance companies
are in fact nothing more, when well conducted, than Savings Banks.
A man pays year after year his premium, and when he dies, whether
it be soon or late, his family has the benefit in such form as not to be
accessible to creditors.” One minister advised that life insurance was “a

67 Geoffrey Clark, Betting on Lives: The Culture of Life Insurance in England,


1695–1795 (Manchester: Manchester University Press, 1999), 40–​59; Sharon
Ann Murphy, Investing in Life: Insurance in Antebellum America (Baltimore:
Johns Hopkins University Press, 2010), 81–​85; Prospectus of the Union
Insurance Company (New York: J. Seymour, 1818), 3; Life Insurance: Its
Principles, Operations and Benefits, as Presented by the Connecticut Mutual
Life Insurance Company (Hartford, Conn.: Case, Tiffany & Burnham,
1846), 3; “The Regulation of Life Insurance,” Merchants’ Magazine and
Commercial Review 27 (1852): 541; A. B. Johnson, “The Relative Merits of
Life Insurance and Savings Banks,” Merchants’ Magazine and Commercial
Review 25 (1851): 670; “Life Insurance,” New York Daily Times, 26 Dec. 1851,
1; Joseph B. Collins, “Life Insurance,” Merchants’ Magazine and Commercial
Review 26 (1852): 196.
  51

The Land of Speculation 51

laudable, honorable and safe way, of preventing a very large amount of


human wretchedness.”68
Courts accordingly used the concept of an insurable interest to distin-
guish genuine life insurance from wagers that others would die sooner
than expected. “In order to prevent individuals from gambling in life-​
insurance,” one writer explained, the law “has declared that A  shall not
insure the life of B, unless he have what is called an insurable interest in
that life; that is, unless A have some pecuniary interest in B’s continuing to
live.” As early as 1815, when the life insurance business was still in its infancy,
the Massachusetts Supreme Court considered the case of Nancy Lord, an
unmarried woman who had insured the life of her brother, a seaman who
died off the coast of Africa. Lord could not recover from the insurance
company unless she had an insurable interest in her brother’s life, the court
reasoned, “for, otherwise, it would be a mere wager-​policy, which we think
would be contrary to the general policy of our laws, and therefore void.”
The court found that Lord did have an insurable interest in her brother,
because she was “a young female without property,” who “had been for
several years, supported and educated at the expense of her brother, who
stood towards her in loco parentis.”69 As Lord stood to suffer economic loss
from her brother’s death, she was understood not as betting that he would
die, but rather as investing in a possible future after his death.
By the early nineteenth century, the law thus reflected the tension in
American thought concerning risk. The law distinguished between gam-
bling, the assumption of risk for its own sake, which was generally (al-
though not always) proscribed, and speculation, the assumption of risk in
a way that served some societal purpose, which was generally (although
not always) permitted. This distinction was never a perfectly clean one.

68 Journal of the Twenty-​Second House of Representatives of the Commonwealth of


Pennsylvania (Lancaster: Benjamin Grimler, 1811 [sic]), 157 (although this pub-
lication bears an 1811 publication date, the committee report quoted is dated 6
Jan. 1812); Rhode Island Republican, 30 Apr. 1839, 3; “Life Insurance,” Bankers’
Magazine and Statistical Register 4 (1849): 371.
69 Rates and Proposals of the New-​ York Life Insurance and Trust Company
(New York: Clayton & Van Norden, 1830), 11; Augustus de Morgan, An Essay
on Probabilities and on Their Application to Life Contingencies and Insurance
Offices (London: Longman, Orme, Brown, Green & Longmans, 1838), 102;
Lord v. Dall, 12 Mass. 115 (1815).
52

52 Speculation

There would always be debate over precisely where to draw the line, and
over whether particular kinds of transactions should be placed on one side
or the other. But there was little controversy over whether some such line
had to be drawn. Speculation was something to be tolerated in general,
but specific instances of speculation might be too harmful to be allowed.

Gamblers in€Gold
The controversy over gold speculation during the Civil War was just one
more installment in this recurring debate, but it was one with an espe-
cially sharp edge. Speculators were blamed, not merely for bringing ruin
on themselves or for harming the wider economy, but for subverting the
war effort. Abraham Lincoln was but one of many observers who held
speculators responsible for the dollar’s decline against gold. “Two-╉thirds
of the public have been until now, and a very large proportion of them
are still thoroughly satisfied that the rise in gold has been solely due to
the efforts of speculators to break down the Government,” reported the
New York Times. One critic charged that fluctuations in gold prices were
caused by “the reckless and gambling maneouvres of speculators, through
fictitious sales, long and short, ‘cornering’ operations, and false rumours
in regard to the war, and everything else that can further their nefarious
schemes, by producing public distrust.” Another New  York paper de-
clared that “partisans of the Rebellion quartered in our city have system-
atically and by concert striven and employed their means to increase the
premium on Gold.” The paper was certain that “they did this on behalf of
their master, Jeff. Davis, and in the conviction that they were aiding the
Rebellion as truly and palpably as though they were wielding muskets in
the front rank of Lee’s army.” A judge in Washington, DC, went so far as
to instruct a jury that the common law prohibited speculation in gold,
which he deemed “contrary to public policy, and calculated to depreciate
the value of the Treasury notes of the United States.”70
In response to this widespread belief that gold speculators were weak-
ening the Union Army, Senator John Sherman of Ohio introduced

70 New York Times, 20 Apr. 1864, 4; “The War and National Wealth,” Princeton Review
36 (1864): 470–╉71; New-╉York Daily Tribune, 15 June 1864, 4; “The Legal Character
of Gold Speculations,” Bankers’ Magazine and Statistical Register 14 (1864): 33.
  53

The Land of Speculation 53

legislation to prohibit speculative transactions in gold. The bill prohib-


ited gold futures—​that is, contracts for the sale of gold to be delivered
at a later date. “The purpose of this bill is simply to prevent people from
selling gold who have none to sell,” Sherman declared on the Senate
floor, “in such a way as to depreciate the price of our paper money.”71
But if many blamed speculators for the fall of the dollar, there seemed
to be just as many who thought it absurd to blame the speculators.
Gold “is merchandise,” the Railway Times observed, “and its price rises
and falls for precisely the same reasons which govern the prices of other
merchandise.” It was “the fluctuating relation of demand and supply”
that pushed gold prices up or down, not the machinations of specula-
tors. “It is so with gold, as with everything else.” A correspondent to
another paper acknowledged that “gamblers in gold have caused it to
rise and fall considerably at times,” but insisted that speculators had no
power to keep gold rising permanently. “There must be a permanent
cause for a permanent effect,” he reasoned. “In the case of gold there is
no mystery about it at all. It rises because its demand is constantly press-
ing on the limits of supply.” The political economist Simon Newcomb
deplored the common tendency to blame price movements on “good
and evil spirits … who are waging a perpetual war, or playing an end-
less game of Stygian ten-​pins, using government stocks for the pins. …
Bull and bear are in fact equally powerless to effect any great or perma-
nent change in the price of so universal and easily transportable an ar-
ticle as gold. At best they can only foresee the change. The stock market
registers the price of gold, but does not fix it. It is governed entirely by
the relation between the supply and demand.”72
This view was well represented in Congress too. “Does any sane man
in the world pretend to believe that the efforts of these people in thus
betting can influence in the slightest the price of gold in the world’s
market?” asked an incredulous Edgar Cowan of Pennsylvania. The price
of gold was determined “not by what the speculators or gamblers of

71 Congressional Globe, 15 Apr. 1864, 1640.


72 “Gold—​Real Causes of Its Fluctuations,” Railway Times, 14 May 1864, 154;
New York Times, 4 July 1864, 2; Simon Newcomb, A Critical Examination of
Our Financial Policy During the Southern Rebellion (New York: D. Appleton
and Co., 1865), 16–​17, 31.
54

54 Speculation

New York may choose to do in the premises, but what the world at large
does,” Cowan continued. “The operation of this bill is not worth any-
thing.” Even some of the supporters of Sherman’s bill admitted that it
was very unlikely to stop the dollar from falling against gold. “Although
we may not believe or feel that a bill of this kind will necessarily produce
the effect,” explained Senator William Fessenden of Maine, “it is never-
theless a duty in the present condition of things in this country to leave
nothing untried.” If Congress could do nothing about the fall of the
dollar, at least it could make a show of trying to do something. Sherman’s
bill was enacted into law in June 1864, shortly after the Confederacy de-
feated the Union at the Battle of Cold Harbor.73
The ban on gold futures did nothing to halt the dollar’s slide. Its only
effect was to shut down the Gold Exchange, which made it more diffi-
cult even for lawful buyers and sellers of gold to find one another. “The
passage of the gold act will long be remembered as the acme of folly at
Washington,” the Bankers’ Magazine lamented. “Thus far every attempt
made by the Secretary of the Treasury and Congress to stop the rise in
gold has only accelerated the downward course of currency,” despaired the
Commercial Advertiser. “The veriest tyro in political science might have
foreseen this.” A hastily assembled group of New York bankers and mer-
chants appointed a committee to travel to Washington to urge Congress
and the Lincoln administration to repeal the gold law. At a meeting with
Treasury Secretary Salmon Chase—​which Chase found “good tempered
on both sides & to me instructive”—​the bankers argued that restricting
the gold trade would only drive the greenback lower against gold. Chase
replied that he “could not see that licence to gambling was essential to
freedom of trade,” and that the bankers’ arguments were in any event
better directed at Congress. Congress proved to be more receptive than
Chase. The prohibition of gold futures, in effect since June 17, was re-
pealed on July 2.74 The speculators were back in business.

7 3 Congressional Globe, 15 Apr. 1864, 1641, 1640; 13 Stat. 132 (1864).


74 Kinahan Cornwallis, The Gold Room and the New  York Stock Exchange and
Clearing House (New York: A. S. Barnes & Co., 1879), 9–​10; “The Financial
Policy of the Government,” Bankers’ Magazine and Statistical Register 14
(1864): 1 (quoting the Commercial Advertiser); New York Times, 23 June 1864,
2; John Niven, ed., The Salmon P.  Chase Papers (Kent, Ohio:  Kent State
University Press, 1983–​88), 1:461–​62; 13 Stat. 344 (1864).
  55

The Land of Speculation 55

In its details, the controversy over gold futures grew out of the unique
crisis of the Civil War, but the controversy shared a structure with many
other debates that took place in the early republic, and many more that
would arise in later years. It required drawing a line between trans-
actions thought to be harmful and those thought to be beneficial, a
line between gambling and legitimate commerce. Everyone agreed that
gambling was bad and that commerce was necessary. The difficulty was
in telling the two apart.
2
Q
Betting on Prices

Ira Foote and Stephen hooker were carpenters from Peru, Illinois.
They both moved to Chicago in 1854, within a few months of each
other. Chicago was booming: the city’s population nearly quadrupled
in the 1850s, so there was plenty of work in the building trades. Foote
was already thirty-eight years old when he arrived in Chicago. He
would remain a carpenter for the rest of his working life. Hooker was
a younger man. When he got to the big city he did not stay a carpen-
ter for long. There was an exciting new kind of business underway in
Chicago, and Hooker got in early. He became a broker on the Chicago
Board of Trade.1
Foote kept his distance from the board of trade for many years. As
the Chicago judge William McAllister would later put it, Foote was
“rather ignorant and inexperienced in matters outside of his own proper
business.” It was not until 1874, when Foote was in his late fifties, that
he was tempted to try his hand at speculating. McAllister said Foote
was suddenly “seized with the desire to make money fast and easy,” but
that seems an uncharitable view.2 After all, fortunes were being made
at the board of trade, including by Hooker, Foote’s former colleague
in carpentry. For two decades Foote had been on the outside looking

1 Information in this and the following several paragraphs is from Transcript of


Record, 73–​97, Pearce v. Rice, 142 U.S. 28 (1891).
2 Tenney v. Foote, 4 Ill App. 594, 597 (1879) (affirming on the basis of McAllister’s
opinion in the trial court).

56
  57

Betting on Prices 57

in. Now he finally ventured inside. Through a broker named George


Adams, Foote purchased some oats.
Oat prices were rising, and within a short time Foote was $3,000
ahead. He asked Adams to sell the oats, but Adams persuaded Foote
not to sell for a while—​“I have got an inside track on the oats busi-
ness,” Adams claimed, with an air of mystery. By the time Adams finally
sold the oats for Foote, most of Foote’s gains had evaporated. To make
matters worse, when Adams sent Foote a check for his much-​reduced
profits, he had deducted fees for storing and handling the oats. Foote
resolved never to have anything to do with the board of trade again.
Soon after, however, he ran into his old friend Stephen Hooker.
Foote recalled:  “Mr. Hooker says, I  am dealing pretty largely on the
board of trade and I will speculate for you. Well, I said, I don’t know.
I thought I wouldn’t do it any more. I had enough with Mr. Adams.”
Hooker would not give in easily. “He said, I can make you some money.
Well, I said, of course I would like to make money. We all want to make
money, but I don’t want any more trading as I traded with Mr. Adams.”
Hooker offered Foote a deal too good to refuse: Hooker would trade
on Foote’s behalf without requiring Foote to put up any margin—​that
is, Foote would not have to deposit any money with Hooker to cover
potential losses. Even better, Hooker offered to let Foote front-​run his
bigger clients. “I have large deals,” Hooker declared, “have some people
in Canada, some in Troy, some in Buffalo, and sometimes they give me
large orders to buy.” Hooker told Foote: “If the market is right I will
buy for you first, then go on these large orders and raise the market, …
and I  can sell them right out for you and make you some money.”
Foote accepted the offer, but he was still upset that Adams had charged
him for handling and storing oats, so he insisted on one condition.
Under no circumstances was Hooker to buy any physical grain for him.
“I told him I wouldn’t have any grain,” Foote recalled. Instead, Foote
instructed Hooker to “speculate in what they call options, from one
month to the other.” The understanding between the two men was that
Foote would neither accept nor deliver any physical commodity. His
profit or loss would be the difference between the value of an option
when he purchased it and its price when he sold it. As Foote explained
it, “I was to pay the difference or they me.”
58

58 Speculation

Foote’s second venture in commodity trading did not go as well as


his first. Within a year he was down $22,000, a colossal sum at a time
when the average annual wage for nonfarm employees was only around
$400. In a normal brokerage relationship, a man of limited means like
Foote could never have racked up such a large debt, because once his
losses exceeded the margin he had deposited with his broker, the broker
would have kept the margin and closed Foote’s account. Special treat-
ment from his old friend Hooker was what put him so far in the hole.
Foote gave up his sideline as a speculator. “I said, I  don’t want any
more; I must quit; I have no money for this thing.” Hooker was left to
pay Foote’s debts and seek reimbursement from Foote, but Foote could
not pay Hooker anything close to $22,000. All he could do was give
Hooker notes (that is, promises to pay), which Hooker passed on to a
variety of other people, doubtless at prices much lower than their face
value to reflect the unlikelihood of payment. The holders of those notes
would hound Foote for the rest of his life.
Foote spent his last years renting a room in a Chicago boarding-
house, unable to work after suffering a stroke, accumulating more
debts, and issuing more notes. Some of the notes were to James Rice,
who paid Foote’s rent and medical bills. One was to the boardinghouse
keeper, Sarah Reed. Foote had promised to marry Reed, but when he
changed his mind after his stroke, Reed insisted on a note as the price
of releasing him from his marriage commitment.3
Most of Foote’s outstanding notes were the result of his misadven-
ture on the board of trade. They gave rise to two lawsuits against Foote
to recover the amounts he had promised to pay. His lawyer raised the
same defense in both: he argued that Foote had no obligation to pay,
because the transactions Hooker had undertaken for him on the board
of trade had been illegal. They had not been genuine purchases or sales,
the lawyer contended. Rather, they had been wagers. Because Foote and
Hooker had been betting on prices, he argued, the notes were nothing
but gambling debts, and gambling debts were unenforceable.
These two cases came before two different Chicago judges within
a span of three years. They posed the identical question: were Foote’s

3 Pearce v. Rice, 142 U.S. 28, 31–​32 (1891); Chicago Daily Tribune, 29 Dec. 1892, 14.
  59

Betting on Prices 59

speculative trades unenforceable wagers? The law applicable to the cases


was the same: it was the law of Illinois, which did not change in any rel-
evant respect between the first case and the second. But the two judges
reached opposite results.
In the first case, Judge William McAllister agreed that Foote’s trans-
actions were unlawful wagers. McAllister emphasized the fact that
Foote and Hooker had explicitly agreed not to buy or sell any actual
commodities; Foote intended neither to deliver nor receive any physi-
cal grain. McAllister reasoned: “It was, therefore, a mere engagement
on the part of Hooker & Co. to gamble for Foote.” He could perceive
no reason “why this species of gambling, though wearing the respect-
able aspect of business, should be looked upon with any less disfavor by
the courts than any other species.” Because Foote had given his notes
as payment for the debts he incurred in these transactions, he had no
obligation to pay them off.4
The second case came before Judge Henry Blodgett, the former pres-
ident of the Chicago and Northwestern Railway and thus a man more
comfortable than McAllister with the ways of commercial life. Foote’s
transactions seemed perfectly ordinary to Blodgett. Hooker “did not
contemplate or intend to make any different transactions for the de-
fendant [Foote] than for his other customers,” Blodgett explained. As
brokers on the board of trade normally did, Hooker “could so manage
the defendant’s deals that he need not take any commodity bought,
but could settle simply the difference between the purchase price and
the market price.” Because these trades were not illegal, Blodgett con-
cluded, Foote could not evade his obligation to pay his debts.5
These particular events were of little significance to anyone besides
Ira Foote and his creditors, but this disagreement between the two

4 Tenney, 597–​601.
5 Jackson v. Foote, 12 F. 37, 39–​41 (N.D. Ill. 1882). Strictly speaking, Blodgett con-
ceded that the trades might have been unenforceable at common law as wagers,
had the suit been brought by Hooker against Foote, but he held that this would
not bar the notes’ enforcement by a bona fide holder, and that the transactions
did not violate the Illinois anti-​option statute, which would have made the
notes unenforceable by a third party. In the other case, McAllister held that the
transactions did violate the Illinois statute. This statute and the common law
will be discussed later.
60

60 Speculation

judges was representative of a broader legal question that became quite


important in the second half of the nineteenth century. The debate
was over the legality of a very common type of transaction, one that
was fundamental to the American economy. On the Chicago Board
of Trade, or the New York Stock Exchange, or any of the many other
commercial exchanges in cities all over the country, most of what was
bought and sold lacked any physical existence. Contracts on the board
of trade were denominated in bushels of grain, but few of the buyers
and sellers ever saw any actual grain. Members of the stock exchange
traded what were nominally shares of the ownership of large corpora-
tions, but in many, perhaps most, of these transactions neither party
ever saw any physical share certificates. Contracts were settled not by
the conveyance of items from one party to another but rather by the
payment of money, representing the difference in the prices of those
items at two different times. If one person expected the price of wheat
or the value of a railroad company to rise, while another thought that
the price of wheat or railroad shares would decline, the exchanges facili-
tated a deal between the two in which neither party needed to possess
any wheat or any railroad shares. Whoever turned out wrong would
simply compensate whoever turned out right.
Was this a smoothly functioning marketplace, in which investments
could be made without the frictions of having to transfer and store the
goods being bought and sold? Or was it a casino, in which the ostensi-
ble buyers and sellers were merely betting on whether prices would rise
or fall? People were clearly speculating on the exchanges, but how were
the winners earning money? Were they being compensated for making
shrewd economic judgments and assuming risks others wished to shed?
Or were they merely getting lucky on wild gambles on events they had
no power to control? These questions preoccupied the American legal
system throughout the second half of the nineteenth century.6 The

6 For a wide range of perspectives on this set of questions, see Urs Stäheli,
Spectacular Speculation: Thrills, the Economy, and Popular Discourse, trans. Eric
Savoth (Stanford, Calif.:  Stanford University Press, 2013), 43–​92; Jonathan
Levy, Freaks of Fortune: The Emerging World of Capitalism and Risk in America
(Cambridge, Mass.:  Harvard University Press, 2012), 231–​63; Roy Kreitner,
Calculating Promises:  The Emergence of Modern American Contract Doctrine
(Stanford, Calif.:  Stanford University Press, 2007), 105–​25; Joshua C.  Tate,
╇ 61

Betting on Prices 61

answers that were worked out would help shape much of the economy
for a long time to come.

The Intent to€Deliver


Under the English common law inherited by the United States, there
was nothing illegal about betting on prices. Stock speculators in London
were trading options and futures as early as the late seventeenth cen-
tury. It was widely recognized that many such sellers, perhaps most, did
not actually own any of the shares they were nominally committing to
sell, and that many such buyers, again perhaps most, had no intention
of receiving any of the shares they were promising to buy. Transactions
were settled, not by the transfer of shares, but by the payment of
money, in the amount of the difference between the contract price and
the market price on the settlement date. By this method, one observer
of the London market pointed out in 1712, “a very beneficial Trade was
daily driven with imaginary Stocks, and many Thousands bought and
sold, to great Advantage, by those who were not worth a Groat.” John
Barnard, the mid-╉eighteenth-╉century member of Parliament who took
the greatest interest in suppressing these transactions, complained that
the market was nothing but “a Lottery, or rather a Gaming-╉House,
publickly set up in the Middle of the City of London.” Traders were wa-
gering on stock prices, but wagers were enforceable under the common
law as long as they weren’t contrary to public policy, and English judges
found that they were not. As one English judge, rebuffing a challenge to
the legality of a difference contract on the price of Spanish bonds, put
it, “we have no right to say that a wager on the price of foreign stock is
void at common law.”7

“Gambling, Commodity Speculation, and the ‘Victorian Compromise,’â•›” Yale


Journal of Law & the Humanities 19 (2007):  97–╉114; Pat O’Malley, “Moral
Uncertainties: Contract Law and Distinctions Between Speculation, Gambling,
and Insurance,” in Risk and Morality, ed. Richard V. Ericson and Aaron Doyle
(Toronto: University of Toronto Press, 2003), 231–╉57.
7 A New Way of Selling Places at Court (London: John Morphew, 1712), 4; The
History and Proceedings of the House of Commons (London: Richard Chandler,
1742–╉44), 7:381; Morgan v. Perrer, 132 Eng. Rep. 486, 490 (C.P. 1837).
62

62 Speculation

English critics of these speculative transactions repeatedly urged


Parliament to enact statutes prohibiting them. One critic was Daniel
Defoe, who proposed that all sales of stock should be illegal unless
the seller immediately transferred the stock to the buyer. The English
stock market crash of 1720—​the famous South Sea Bubble—​brought
forth a wave of similar calls for regulation. The Scottish essayist Thomas
Gordon, for example, proposed “declaring all fictitious Contracts here-
after Illegal and Void, which shall not be immediately comply’d with,
and punctually fulfill’d; and by inflicting a proper Punishment on all
Persons assuming a false Power, and pretending to sell and buy Stock
for themselves, or others, who have neither Money to Purchase, nor
Stock to Deliver.” Parliament eventually responded. Sir John Barnard’s
Act of 1734 banned stock options, prohibited parties from settling stock
contracts by paying price differentials, and voided all contracts for the
sale of stock that the seller did not actually own at the time the con-
tract was entered into. This statute would remain in force until it was
repealed in 1860. On paper, the law made it impossible to wager on
stock prices, except by simply buying stock and holding it in the hope
that the price would rise. In practice, options and futures remained
common, as did the settlement of contracts by the payment of price
differentials. Such contracts were no longer enforceable in court, but
they could still be enforced by exclusion from the informal community
of brokers, and later, with the creation of a formal stock exchange, by
expulsion from the exchange.8
The United States inherited English common law but not most
English statutes, so there was nothing illegal about betting on prices in
the early United States either. As a Philadelphia speculator suggested to
a colleague in Boston in 1789, “a sum of Continental Debt … might be
sold to be delivered in any given time: that is to fix the price at present,
and agree upon a mode of fixing it at the end of the Time,—​and then
without transferring, or delivering any Certificate to pay or receive the

8 Daniel Defoe, The Villainy of Stock-​Jobbers Detected (London: s.n., 1701), 24;
Thomas Gordon, An Essay on the Practice of Stock-​Jobbing (London: J. Peele,
1724), 18; 7 Geo. III c. 8 (1734); Stuart Banner, Anglo-​American Securities
Regulation: Cultural and Political Roots, 1690–1860 (Cambridge: Cambridge
University Press, 1998), 106–​7.
  63

Betting on Prices 63

difference of price. This is the Common practice in England.” It soon


became a common practice in the United States as well, as evidenced by
the published court decisions of the 1790s enforcing such transactions.
A few states responded by enacting statutes voiding certain speculative
contracts for the sale of securities, as explained in the previous chapter.
The most important of these statutes, New York’s ban on the sale of se-
curities the seller did not own at the time of the contract, was repealed
in 1858, after a few decades of exerting little practical effect on the
market. As one New York judge explained in 1844, “it is a well known
fact that shares of stock are constantly sold at the board of brokers,
which shares exist only in the imagination of the nominal buyers and
sellers.” One guide to the New York market noted that “the practice of
betting upon the future rise and fall of prices is too well understood to
need any particular explanation.”9
These statutes were narrowly targeted at the stock market. They did
not proscribe similarly speculative transactions in other commodities,
which soon began to grow common. There was nothing new about
speculating in commodity prices. Merchants who expected prices to
rise had long been able to purchase items with the aim of reselling
them, and those who expected prices to decline could agree to sell items
in the future that one did not yet possess, in the hope of buying them
at lower prices later. As the English judge William Blackstone observed
in his widely used eighteenth-​century treatise, “property may also in
some cases be transferred by sale, though the vendor hath none at all in
the goods.” And there had long been futures markets in other countries.
What was new to the United States in the early nineteenth century
was the development of organized markets for commodity speculation
along the lines of the stock markets that were already familiar. As early
as 1819, the New York commercial press was reporting on the specula-
tive market in tallow futures. (Tallow, or rendered animal fat, was used

9 Joseph Stancliffe Davis, Essays in the Earlier History of American Corporations


(Cambridge, Mass.: Harvard University Press, 1917), 1:196; Groves v. Graves, 1
Va. 1 (1790); Livingston v. Swanwick, 2 U.S. 300 (C.C.D. Pa. 1793); Graham
v.  Bickham, 4 U.S. 149 (Pa. 1796); Dykers v.  Allen, 7 Hill 497, 500 (N.Y.
Sup. Ct. 1844); William Armstrong, Stocks and Stock-​Jobbing in Wall Street
(New York: New-​York Publishing Co., 1848), 11.
64

64 Speculation

to make candles.) Those who expected the price of tallow to rise “pur-
chased on speculation, never intending to take or pay for them [that is,
shipments of tallow], but to receive the difference between the contract
prices and the market rates.” Such merchants were not in the candle
business; they were placing bets on the price of tallow. By the 1840s,
New York and Buffalo had markets for trading grain futures, markets
likewise organized to permit the “purchase” of future rights to receive
agricultural commodities by people with no intention of ever receiving
them, and the “sale” of such rights by people who would never possess
them.10 There was not yet any law prohibiting these transactions. Nor
was there any in England, where the analogous statutes did not restrict
trading in anything other than securities.
Speculating on commodity prices was nevertheless troubling to some
judges, who looked for ways of interpreting the law so as to limit these
transactions. In an English case from 1822, Chief Justice Charles Abbott
allowed a buyer to rescind a contract for the future purchase of wheat
when the seller admitted that he could not show samples of the wheat
to the buyer because he did not yet have any wheat. If such a contract
were enforceable, Abbott reasoned, “a man might bargain to deliver
corn not then in his possession, and rely upon making a future pur-
chase in time to fulfil his undertaking; but that is a mode of dealing not
to be encouraged.” A few years later Abbott refused to enforce a similar
contract for the sale of nutmeg. In February, the seller had agreed to de-
liver nutmeg in May, but he did not acquire any nutmeg until March.
The seller was speculating that nutmeg prices would decline between
February and March. “Such a contract amounts, on the part of the
vendor, to a wager on the price of the commodity,” Abbott lectured,
“and is attended with the most mischievous consequences.” When
the seller’s lawyer informed him that such contracts were common,
and that his decision would “introduce a most material change in the
proceedings of the Royal Exchange,” Abbott retorted that any such

10 William Blackstone, Commentaries on the Laws of England (Oxford: Clarendon


Press, 1765–​69), 2:449; Mark West, “Private Ordering at the World’s First
Futures Exchange,” Michigan Law Review 98 (2000): 2574–​615; Commercial
Advertiser, 4 Mar. 1819, 2; Jeffrey C. Williams, “The Origin of Futures
Markets,” Agricultural History 56 (1982): 306–​16.
  65

Betting on Prices 65

change would be for the better. “If it had been acted upon during the
last twelve months,” he declared, “much of that distress which now
presses upon the community would have been avoided.”11
In the United States, Supreme Court Justice Joseph Story shared
Abbott’s concern about the ability of merchants to wager on prices by
selling things they did not yet own. Story tried to influence the law as
much through his legal treatises as through his judicial opinions. In his
Commentaries on the Law of Agency, first published in 1839, he quoted
Abbott at length in urging the adoption of a similar rule in the United
States. “Such a contract amounts, on the part of the vendor, to a wager
on the price of the commodity, and is attended with such mischie-
vous consequences as to be prohibited by public policy,” Story insisted.
“There can be little doubt, that it is a gaming speculation, which no
court ought to enforce or encourage.”12 Story, like Abbott, proposed to
curb speculation with a simple rule: one could not contract to sell what
one did not own.
The difficulty with the rule Abbott and Story sought to introduce was
that such contracts had long been common. Farmers, for example, sold
crops before the harvest, while manufacturers sold products before they
were made. In such circumstances, the seller could not avoid selling
something he did not yet own. It was hard to argue that such contracts
posed dangers to trade, and even harder to claim that all along they
had been contrary to the common law. In England, Abbott’s holdings
lasted only a bit more than a decade before being overruled. “I cannot
see what principle of law is at all affected by a man’s being allowed to
contract for the sale of goods, of which he has not possession at the
time of the bargain,” declared Judge James Parke. “Such a contract does
not amount to a wager … and even if it were a wager, it is not ille-
gal, because it has no necessary tendency to injure third parties.” Judge
William Henry Maule agreed that Abbott’s rule was “contrary to law,
and most inconvenient in practice: and I have often heard it spoken of
with great suspicion, both by lawyers and mercantile men upon both

11 Lorymer v.  Smith, 107 Eng. Rep.  1, 2 (K.B. 1822); Bryan v.  Lewis, 171 Eng.
Rep. 1058, 1059 (K.B. 1826).
12 Joseph Story, Commentaries on the Law of Agency (Boston: Little, Brown and
Co., 1839), 233–​34.
66

66 Speculation

grounds,—​as against law, and against all mercantile convenience.” In


the United States, Story’s proposal never caught on. Courts had already
declined to invalidate contracts to sell items the seller did not own, and
they continued to do so afterward. “Many moralists doubt the policy
of permitting a party to contract for the sale of goods which he does
not own at the time of making the contract, because it partakes of the
nature of a gambling transaction,” one New York judge acknowledged,
“but it is now well established” that such contracts were lawful. They
were simply too useful to be proscribed. “The mercantile business of the
present day could no longer be successfully carried on, if merchants and
dealers were unable to purchase or sell that which as to them had no
actual or potential existence,” explained the Michigan Supreme Court,
in upholding a contract for the sale of corn futures by a seller who pos-
sessed no corn on the contract date. “A dealer has a clear right to sell
and agree to deliver at some future time that which he then has not,
but expects to go into the market and buy.”13 By the middle of the nine-
teenth century, judges in England and the United States had decisively
rejected the idea that the common law prohibited selling what one did
not own.
Meanwhile, however, two developments were putting increasing
pressure on the courts to find some alternative way to forbid betting
on prices. One was the continued growth of organized markets in op-
tions and futures. The Chicago Board of Trade, which would become
the leading American commodity exchange, was formed in the 1840s,
followed by similar boards of trade in many other cities. A combination
of technological changes in the 1850s—​the construction of railroads,
grain elevators, and telegraph lines, along with the implementation of
a grading system for dividing grain into standardized quality classes—​
provided an infrastructure for the trading of options and futures in
grain.14 There were soon comparable infrastructures for speculative

13 Hibblewhite v. M’Morine, 151 Eng. Rep. 195, 197 (Exch. 1839); Frost v. Clarkson,
7 Cow. 24 (N.Y. Sup. Ct. 1827); Eastman v. Fiske, 9 N.H. 182 (1838); Noyes
v. Spaulding, 27 Vt. 429 (1855); Stanton v. Small, 3 Sandf. 230, 237–​38 (N.Y.
Super. Ct. 1849); Gregory v. Wendell, 39 Mich. 337, 340 (1878).
14 Jonathan Lurie, The Chicago Board of Trade 1859–1905: The Dynamics of Self-​
Regulation (Urbana: University of Illinois Press, 1979), 25; William Cronon,
  67

Betting on Prices 67

transactions in timber and livestock. The New  York Stock Exchange


continued to grow, as did stock exchanges in other cities, all of which
permitted brokers to trade options and futures.
The second development was the gradual shift in the common law’s
treatment of gambling, discussed in the previous chapter. In the early
nineteenth century, wagers had been enforceable contracts, except
where they were contrary to public policy. By the middle decades of the
century, gambling contracts were generally unenforceable. This change
in the legal background had important consequences for speculative
transactions. In 1800, if one classified a speculative purchase as a wager,
that meant, with certain narrow exceptions, that it was an enforceable
contract. In 1860, by contrast, to classify a speculative transaction as a
wager meant that it was not an enforceable contract.
These factors contributed to the state-​by-​state emergence, between
the 1850s and the 1870s, of a new common-​law rule prohibiting the
enforcement of “difference contracts”—​that is, contracts in which the
seller had no intention of delivering the item ostensibly being sold, and
the buyer had no intention of receiving it, but both parties intended
the contract to be settled by the payment of an amount of money repre-
senting the difference in the item’s price between the contract date and
the settlement date. The first step in this direction was an English case
from 1851 called Grizewood v. Blane. On August 3, Blane had agreed
to deliver stock to Grizewood on September 13 at an agreed-​upon
price. But on September 11, two days before the nominal delivery date,
Grizewood agreed to deliver to Blane the same amount of the same
stock, at the price then in effect. When the two agreements netted out,
all that was left was a contract to pay the difference in the stock price
between the two dates. Was it enforceable? Or was it a wager, and thus
unenforceable? The court determined that the answer turned on the in-
tentions of the parties. “If neither party intended to buy or to sell” any
shares, reasoned one judge, and if “the whole thing was to be settled by
the payment of differences,” then “it was no bargain, but a mere gam-
bling transaction.” A second judge agreed that if the parties had never

Nature’s Metropolis: Chicago and the Great West (New York: W. W. Norton, 1991),
97–​147.
68

68 Speculation

intended to deliver or receive any stock, “the transaction was clearly


gambling, and a practice which every one must condemn.”15
The new rule jumped to New York a few years later. In May 1856,
Elisha Hinmann agreed to sell five hundred barrels of pork to George
Cassard at $17 per barrel, to be delivered in September. Cassard was
in the meatpacking business—​he bought pork and beef; preserved
the meat in brine, which was necessary for long-​term storage in the
era before refrigeration; and then sold it.16 By purchasing meat sev-
eral months in advance, Cassard assured himself of a supply and pro-
tected himself against future price increases. When September came,
however, Hinmann refused to deliver the pork. Pork prices had risen
from $17 to $23 per barrel, so Hinmann, who would have to buy pork
at the current market price before he could deliver any to Cassard,
stood to lose money on the transaction.17 Cassard sued, and Hinmann
defended on the ground that the contract was illegal because it was
nothing more than a wager on the price of pork. He was not a dealer in
pork, Hinmann declared, nor did he ever possess any pork. He never
intended to deliver any pork to Cassard, and Cassard never expected to
receive any. Rather, Hinmann argued, both men intended the contract
to be settled by the payment of money in the amount of the difference
between the value of the pork in May and its value in September. As
wager contracts were unenforceable under New  York law, Hinmann
concluded, Cassard had no right to collect any money or any pork
from him. This defense was probably wrong as a factual matter, be-
cause Cassard most likely expected to receive pork from Hinmann.
Cassard genuinely needed pork for his business. But the litigation
did not progress far enough to find out, because at the pleading stage
Cassard argued that Hinmann’s defense was not a good one—​that
is, Cassard contended that even if all the facts alleged by Hinmann

1 5 Grizewood v. Blane, 138 Eng. Rep. 578, 584 (C.P. 1851).


16 Cassard described his business as a witness testifying in Adams v. Capron, 21
Md. 186, 194–​95 (1864). A price quotation for Cassard’s hams in brine can be
found in New York Daily Times, 3 May 1855, 1.
17 I am inferring that pork prices rose $6 per barrel, because the order was for five
hundred barrels and Cassard alleged a loss of $3,000. See Cassard v. Hinman,
14 N.Y. Super. 207 (1857).
  69

Betting on Prices 69

were true, the contract would still be valid. As a result, Justice Murray
Hoffman of New York’s Superior Court confronted a pure question of
law: if neither a buyer nor a seller expects delivery of the item being
sold, but both merely intend to bet on whether its price rises or falls,
is their contract void as a wager?18
Hoffman decided that the contract was indeed void as a wager. After
quoting extensively from Grizewood v. Blane, the English case from a
few years before, he made no secret of his distaste for “the unbridled
and defiant spirit of speculation.” Hoffman was a prominent layman in
the Episcopal Church. He had recently authored a lengthy treatise on
Episcopal canon law, and he would later write books about ecclesiasti-
cal law more broadly. His opinion suggests that his religious convictions
played a role in forming his view of speculation. Hoffman declared:
“I rejoice that a court of justice is able to do this at least—​to con-
demn the offence—​to annul the contract; and to clear the law from the
stain of enduring a practice teeming with temptation and disgrace to
those engaged in it, and with baneful influences upon the efforts of the
honest and just.” The decision was affirmed on appeal, by judges who
likewise quoted Grizewood at length.19 In New  York, a contract was
now an unlawful wager on prices if neither party expected the delivery
of the item supposedly being sold.
This new rule would eventually spread throughout the United States,
but for several years New York was the only state to adopt it, because
the new rule was not easy to reconcile with two well-​accepted legal
principles. One principle was that a party to a contract was normally
not allowed to claim the existence of an oral understanding at variance
with the written terms of the contract. If a contract said it was for the
sale of pork, then it was for the sale of pork, and a person who had
signed the contract could not evade his obligation by asserting that the
contract was really about something else. The second principle was that

18 Cassard v. Hinmann, 14 How. Pr. 84 (N.Y. Super. Ct. 1856).


19 Murray Hoffman, A Treatise on the Law of the Protestant Episcopal Church in
the United States (New York: Stanford and Swords, 1850); Murray Hoffman,
Ecclesiastical Law in the State of New York (New York: Pott and Amercy, 1868);
Murray Hoffman, The Ritual Law of the Church (New York: Pott, Young &
Co., 1872); Cassard, 14 How. Pr. at 91; Cassard, 14 N.Y. Super. at 211.
70

70 Speculation

the normal remedy for a breach of contract was the payment of money
from the breaching party to the innocent party, in an amount equal
to the difference between the contract price and the market price at
the time of the breach, precisely the outcome that the New York rule
condemned as a wager. When a pork seller failed to deliver pork to the
buyer, a court normally would not order him to deliver the missing
pork; rather, it would order him to compensate the buyer with enough
money to enable the buyer to obtain equivalent pork from another
vendor.20 New  York’s new rule appeared to contravene both of these
principles. The new rule allowed parties to wriggle out of contracts that
were lawful as written by claiming that the written contract was merely
a cover for what was truly a wager. And the new rule seemed to prohibit
contracting parties from settling their differences by the method that
judges routinely employed in contract cases—​the payment of money
from the breaching party to the nonbreaching party.
For these reasons, the first American court outside of New York to
consider the question rejected New  York’s new rule. The case before
Judge Thomas Drummond of Chicago involved an April contract to
sell fifteen thousand bushels of corn in June at 48 cents a bushel. By
June, corn was at 63 cents a bushel, and the seller had lost interest in
delivering it for 48 cents. When the buyer sued for breach of contract,
the seller tried the defense that had worked in New York: he argued
that neither he nor the buyer had any expectation of possessing any
corn, but that the contract had been a pure wager on the price of corn
and was thus unenforceable. Judge Drummond would have none of it.
The contract was for the purchase of corn, he observed, and “the rule
is well settled that when two men make a contract, and reduce it to
writing, and sign it that it is the contract between them. It cannot be
shown verbally that something different was intended at the time from
what appears in the writing.” Drummond conceded that the outcome
of the case—​a judgment requiring the seller to compensate the buyer
for the difference in the price of corn between April and June—​would
be identical to the outcome of a wager on prices. “So it is in any case of

20 Theodore Sedgwick, A Treatise on the Measure of Damages (New York: John


S. Voorhis, 1847), 260.
  71

Betting on Prices 71

this kind, when a party does not perform his contract,” he noted. “But
that circumstance does not make the contract the same” as a wager.21
The new rule nevertheless spread from state to state in the 1860s
and 1870s. The reason is not hard to find. Judges often laid their cards
on the table: they straightforwardly declared their moral opposition to
speculation. Pennsylvania became the second state to adopt the rule in
1867, when the Pennsylvania Supreme Court heard a case involving a
short sale of two hundred shares of Harlem Railroad stock. “It is said
the form in which this contract appears enters largely into the business
of stock brokerage,” observed Chief Justice James Thompson. But he
meant to put a stop to it. “The fewer licenses that are issued for such a
business the better,” Thompson continued. “Anything which induces
men to risk their money or property without any other hope of return
other than to get for nothing any given amount from another, is gam-
bling, and demoralizing to the community, no matter by what name it
may be called. It is the same whether the promise be to pay on the color
of a card, or the fleetness of a horse.” Thompson concluded with a sum-
mary of the dire consequences of speculation:  “I apprehend that the
losses incident to the practice disclosed in this very case, within the past
five years, have contributed more to the failures and embezzlements by
public officers, clerks, agents and others acting in fiduciary relations,
public and private, than any other known, or perhaps all other causes;
and the worst of it is, that in the train of its evils, there is a vast amount
of misery and suffering by persons entirely guiltless of any partition
in the cause of it.” Thompson was careful to distinguish “bona fide
time contracts about subjects of actual purchase” from mere wagers on
prices. The former, he reluctantly acknowledged, “seem from custom
necessary in our country.” But he had no doubt that the parties before
him were gamblers through and through. Their contract took the form
of a sale of stock, but, as in famous historical episodes like the South
Sea Bubble and the tulip mania in Holland, “the form served only as a
thin covering of the most frightful systems of gambling ever known.”22

21 Porter v. Viets, 19 F. Cas. 1077, 1078 (C.C.N.D. Ill. 1857).


22 Brua’s Appeal, 55 Pa. 294, 298–​99 (1867).
72

72 Speculation

The same sort of moralizing can be found among many judges of


the period. Oat options sold on the Chicago Board of Trade “are essen-
tially nothing but bets upon the price of oats,” insisted one of the city’s
federal judges, “and as it is obvious that the effect of such transactions
is to beget wild speculations, to derange prices, to make prices artifi-
cially high or low, as the interests, strength and skill of the manipulators
shall dictate, thereby tending to destroy healthy business and unsettle
legitimate commerce, there can be no doubt of the injurious tendency
of such contracts, and that they should be held void as against public
policy.” In the District of Columbia, a court refused to enforce a prom-
issory note signed by General George Custer just a few months before
he was killed at the Battle of Little Bighorn. Custer had been speculat-
ing on the New York Stock Exchange, and the promissory note had paid
for futures transactions in which neither Custer nor his counterparties
had ever transferred any stock. The court held that these transactions
were illegal, and that Custer’s note was accordingly void. “All observ-
ers agree that the inevitable effect of such dealings is to encourage wild
speculations; to derange prices to the detriment of the community; to
discourage the disposition to engage in steady business or labor, where
the gains though sure, are too slow to satisfy the thirst for gaming when
once aroused; and to fill the cities with the bankrupt victims of such
disasters as any ‘Black Friday’ may develop,” declared Justice Alexander
Burton Hagner. “The extent of this form of speculation now rife in our
country is unprecedented.” Such transactions were “immoral and ille-
gal,” sniffed the Nebraska Supreme Court. “Betting on a game of faro,
brag, or poker, cannot be more hazardous, dangerous or uncertain,”
agreed the Georgia Supreme Court. “Indeed, it may be said that these
animals are tame, gentle and submissive, compared to this monster.
The law has caged them, and driven them to their dens; they have been
outlawed, while this ferocious beast has been allowed to stalk about in
open mid-​day, with gilded signs and flaming advertisements, to lure
the unhappy victim to its embrace of death and destruction.”23 In state

23 Ex parte Young, 30 F. Cas. 828, 832–​33 (N.D. Ill. 1874); Justh v. Holliday, 13 D.C.
346, 348–​49 (1883); Rudolf v.  Winters, 7 Neb. 125, 130 (1878); Cunningham
v. National Bank of Augusta, 71 Ga. 400, 403 (1883).
  73

Betting on Prices 73

after state, judges clearly took great satisfaction in adopting a rule that
promised to limit practices they viewed as grave dangers.
This judicial attitude was almost certainly reflective of wider public
opinion. There does not appear to have been any significant effort, in
any state, to enact a statute to undo what the judges had done and
allow transactions in which neither party had the intent to deliver. The
prevailing view seems rather to have been that of the Bankers’ Magazine,
which declared that “in this evil time, when no power seems able to
stay the spirit of speculation, it is cheerful to note how strongly and
uniformly the Courts have set their faces against sustaining specula-
tive transactions.” In summarizing the decided cases, a pair of lawyers
approvingly remarked of the line the judges had drawn, a “distinction
between gambling with a commodity, and dealing there in by way of
legitimate speculation.”24
Occasional judges went so far as to declare that all options were il-
legal wagers. The transaction “called an ‘option sale,’ ” averred Judge
Thomas Gantt of St. Louis, was in fact “a wager respecting the rise and
fall in the market of mess pork.” Missouri law did not prohibit options,
he acknowledged, but “it is altogether inadmissible to infer that every-
thing is moral which the penal law does not forbid.” Gambling was
illegal and options were a form of gambling. A  commentator agreed
that “options” were lawful so long as the only choice open to the seller
was the precise time of delivery, but that options were illegal where
the seller had the option not to deliver the goods at all. But this was
an extreme view. As the New York lawyer John Dos Passos (the novel-
ist’s father) explained, “options frequently represent real transactions.”
Where the owner of stock desired to sell if the price rose high enough,
or to buy more if the price dropped low enough, it would be possible
to wait for the market to change and act accordingly, but one could
achieve the same result with more certainty simply by entering into
option contracts to buy or sell stock at the appropriate prices. In such
a case, Dos Passos observed, the stock owner was not wagering; he was

24 “Speculation and the Abuse of Trusts,” Bankers’ Magazine and Statistical


Register 38 (1884): 653; Lewis H. Bisbee and John C. Simonds, The Board of
Trade and Produce Exchange (Chicago: Callaghan & Co., 1884), 287.
74

74 Speculation

genuinely committing to buy or sell stock if prices moved in a certain


way.25 A rule classifying all options as wagers was simply too broad.
Most judges accordingly drew a more accommodating line be-
tween permissible speculation and impermissible gambling. In one
Pennsylvania case, for instance, the lawyer for one party argued that “all
purchases of stocks, with a view to re-​sell and make profit on their rise,
or contracts to furnish stocks on time, should be declared gambling
transactions and illegal.” This argument, the court observed, “would
make a great inroad into what has, for an indefinite period been re-
garded as a legitimate business.” The true distinction, the court con-
cluded, “is between a real transaction, where stock is actually bought
and purchased and delivered, and a transaction in which that is not
contemplated at all, and where the parties agree, from the beginning,
that the stock is not to be delivered, but that they are to settle their
mutual wagers upon the price of stock by paying the difference between
the sales at the different times.” A real transaction, involving the actual
purchase of something, could be as speculative as the parties desired
without thereby becoming illegal. “It is the business of all people who
engage in trade to speculate,” the court concluded; “all trade is based
upon speculation, and no contracts are obnoxious to legal objection
merely because they are speculations.” But if the supposed buyer and
seller never intended any delivery of stock, the transaction would be an
unlawful wager.26
Court opinions of the 1870s are replete with lengthy discussions
of this issue. Nearly all of them drew the same distinction, between
speculative real transactions (which were lawful) and mere wagers on
prices (which were not). “We must not confound gambling, whether
it be in corporation stocks or merchandise, with what is commonly
termed speculation,” the Pennsylvania Supreme Court explained in

25 Waterman v. Buckland, 1 Mo. App. 45, 46–​48 (1876); C.H.W., note appended


to Sawyer v.  Taggart, American Law Register 27 (1879):  229–​ 30; John
R.  Dos Passos, A Treatise on the Law of Stock-​Brokers and Stock-​Exchanges
(New York: Harper & Brothers, 1882), 445–​46.
26 Smith v. Bouvier, 70 Pa. 325, 331, 328 (1872) (the latter two quotations are from
the jury charge of the trial judge, which was approved by the Pennsylvania
Supreme Court).
  75

Betting on Prices 75

1872. “Merchants speculate upon the future prices of that in which


they deal, and buy and sell accordingly. … But when ventures are
made upon the turn of prices alone, with no bona fide intent to
deal in the article, but merely to risk the difference between the rise
and fall of the price at a given time, the case is changed.” “The law
does not undertake to prevent speculation,” agreed a federal judge in
Wisconsin. “The truth is, men are speculative creatures as certainly
as they are eating and sleeping ones. And, although it is undoubtedly
true that much harm comes to the community from over-​speculation,
it is more than doubtful if the world would be better off without
speculators; or, if it would be, that the law can do much in the way
of abolishing them.” The important thing was not whether a contract
was speculative, or whether it included an option. What was impor-
tant was whether the contract was for a real sale. As New York’s high-
est court concluded, if “there is no intention on the one side to sell or
deliver the property, or on the other to buy or take it, but merely that
the difference should be paid according to the fluctuation in market
values, the contract would be a wager.”27
By the 1880s, the intent-​to-​deliver rule was firmly established in
American law. “The generally accepted doctrine in this country,” the US
Supreme Court declared in 1884, was that “a contract is only valid when
the parties really intend and agree that the goods are to be delivered by
the seller and the price to be paid by the buyer; and if, under guise of
such a contract, the real intent be merely to speculate in the rise and
fall of prices, and the goods are not to be delivered, but one party is to
pay the other the difference between the contract price and the market
price of the goods at the date fixed for executing the contract, then the
whole transaction constitutes nothing more than a wager, and is null
and void.”28 Over the preceding three decades, case by case, state by
state, judges had crafted a new limitation on speculative transactions.
As a doctrinal matter, the judges were reconciling two developments
that pulled in opposite directions—​on one side, the rise of a blan-
ket common-​law disapproval of gambling; on the other, the growth

27 Kirkpatrick & Lyons v. Bonsall, 72 Pa. 155, 158 (1872); Clarke v. Foss, 5 F. Cas.
955, 960 (W.D. Wisc. 1878); Bigelow v. Benedict, 70 N.Y. 202, 206 (1877).
28 Irwin v. Williar, 110 U.S. 499, 508–​9 (1884).
76

76 Speculation

of organized futures and options trading, which facilitated forms of


speculative trading that looked an awful lot like gambling. As a policy
matter, the judges were drawing a line between two kinds of risky
transactions—​on one side, those they considered customary and useful,
and on the other, those they deemed novel and dangerous. “Two con-
flicting tendencies exist” in the court opinions on the subject, observed
the prolific legal writer Francis Wharton. One was “laissez faire,” the
view that “business should be left free” as much as possible. The other
“starts from an ethical or police basis,” Wharton noted, according to
which certain contracts were immoral by their nature and deserved cen-
sure even if businesspeople found them useful.29 The intent-​to-​deliver
rule was the judges’ way of mediating between these two goals.
But why exactly did judges consider transactions more dangerous
when parties intended to settle their contracts through the payment of
differences? What was it about the intended delivery of an item that
made judges more willing to classify transactions in that item as ac-
ceptable business practice? The answers to these questions shed some
light on why the rule was adopted so quickly and with so little recorded
dissent.
In the summer of 1880, a man named Melchert in Davenport, Iowa,
sold fifteen thousand bushels of rye on the Chicago Board of Trade,
to be delivered any time in September, at prices ranging from 65½ to
68½ cents per bushel. When September came, rye was at 80 cents per
bushel and rising. Melchert frantically telegraphed his broker, Erick
Gerstenberg, to “cover immediately”—​that is, to limit his losses by
buying fifteen thousand bushels of rye at the current market price, before
rye became even more expensive. But the direct telegraph line between
Davenport and Chicago was not working. The telegraph company had
to reroute the telegram through Omaha and St. Louis, which added a
few hours to the time it took for Gerstenberg to receive it. In those few
hours, rye jumped from 80 cents to 85 cents per bushel. Because of the
faulty telegraph line, Melchert had lost an additional $750. He sued
the telegraph company to get it back. One of the telegraph company’s

29 Francis Wharton, note appended to Melchert v. American Union Telegraph Co.,


11 F. 193, 201, 203 (C.C.D. Iowa 1882).
  77

Betting on Prices 77

defenses was that Melchert had not in fact lost anything, because his rye
contracts were unenforceable wagers. Melchert did not have to buy any
rye, the telegraph company argued, because neither he nor his counter-
parties ever intended that any rye would be transferred between them.
They were merely betting on rye prices.30
The law was clear by 1880: if no one involved in these contracts con-
templated the delivery of any rye, the contracts were unenforceable,
and Melchert would lose his suit against the telegraph company. Judge
James Love accordingly focused his attention on what was in Melchert’s
mind when he entered into the contracts. Did Melchert actually own
any rye in the summer, when he agreed to sell fifteen thousand bushels
of it? No, he hadn’t owned any rye whatsoever. When Melchert in-
structed Gerstenberg to “cover immediately,” was it likely that Melchert
actually meant for Gerstenberg to go out and purchase fifteen thousand
bushels of physical rye? To purchase fifteen thousand bushels of rye
at 80 cents per bushel would have cost $12,000, but Melchert did not
send any money to Gerstenberg, and it was not likely that Gerstenberg
would have lent so much money to Melchert. In fact, Gerstenberg did
not buy any physical rye. Rather, he entered into a contract to purchase
fifteen thousand bushels of rye from a firm called A.  M. Wright &
Co., which turned out to be the very same firm to which Melchert had
sold the fifteen thousand bushels back in the summer. And did A. M.
Wright & Co. actually own any rye? No. Like Melchert himself, the
firm owned no rye at all. Melchert’s “purchase” from A.  M. Wright
served merely to cancel out his “sale” to A. M. Wright from two months
before. Neither side ever had any rye, and neither ever intended to de-
liver rye to the other.31
These facts prompted Judge Love to consider why such an arrange-
ment should be illegal, when the very same transactions would have
been lawful had Melchert and A. M. Wright simply intended to trans-
fer ownership of some real rye. The answer, Love suggested, was that re-
quiring the intent to transfer property was a way of limiting speculation
to those who could best afford it. Someone who actually owned fifteen

30 Melchert v. American Union Telegraph Co., 11 F. 193 (C.C.D. Iowa 1882).


31 Id. at 197–​99.
78

78 Speculation

thousand bushels of rye would have to be a person of some means. At


80 cents a bushel, that much rye was worth $12,000, about thirty times
the average annual wage. But Melchert’s trades “required no capital,
except the small sums demanded to put up margins and pay differ-
ences.” His mode of dealing was open to “any impecunious gambler,”
Love reasoned. “It enables mere adventurers, at small risk, to agitate the
markets.” Love contrasted “the unscrupulous speculator, with little or
no capital,” with “the honest and legitimate trader,” a person with capi-
tal, who could afford to buy and sell real commodities. The intent-​to-​
deliver rule was not just a way of barring certain kinds of speculation; it
was also a way of barring certain kinds of speculators from the market.
The danger of letting the nonaffluent speculate was not just that they
might bankrupt themselves, but that they would harm the respectable,
by causing “sudden fluctuations in values” that could not be foreseen
even by the steady merchant.32
The intent-​to-​deliver rule thus had a sociological as well as an eco-
nomic significance. “The theory upon which these cases proceed,”
explained the New  York lawyer T.  Henry Dewey, is “that a man of
small means would not undertake to perform contracts requiring large
amounts of money to buy the commodities which he had sold short,
or to pay for those which he had purchased, but that he would make
a bet on the price, as he would then have to pay only differences if he
lost.” Respectable men of business “exhibit high mental grasp, and great
knowledge of business, and of the affairs of the world,” explained Justice
Daniel Agnew of the Pennsylvania Supreme Court. “Their speculations
display talent and forecast. … Such speculation cannot be denounced.”
But when actual commodities were not being bought and sold, specula-
tion assumed a darker character. “No money or capital is invested in the
purchase, but so much only is required as will cover the difference—​a
margin, as it is figuratively termed,” Agnew observed. “The difference
requires the ownership of only a few hundreds or thousands of dollars,
while the capital required to complete an actual purchase or sale may be
hundreds of thousands or millions. Hence ventures upon prices invite
men of small means to enter into transactions far beyond their capital.”

32 Id. at 195.
  79

Betting on Prices 79

Such a man happened to be in Agnew’s court. He was a Pittsburgh


resident named Sterling Bonsall, who had options to buy forty thou-
sand barrels of oil for $168,000. But what would Bonsall do with so
much oil? As he admitted at trial, he was not a refiner of oil. He had
no plans to consume any oil. He had no prospect of selling the oil to
someone else. And even if he had some purpose for the oil, he had
no chance of coming up with a sum anywhere near $168,000. “What
intent would reasonably be inferred from these circumstances?” Agnew
asked. “Certainly the belief induced that there was not an intent to
bargain for oil for a real business purpose.” Bonsall was a man of small
means. One purpose of the intent-​to-​deliver rule was to shut such a
man out of the market.33
Because the enforceability of transactions depended entirely on the
intent of the parties, the rule required courts to make some difficult
factual determinations. By the time of trial, the party who wanted to
enforce the contract had every incentive to claim that he intended to
deliver or accept some real commodity, while the party who wanted
to get out of the contract had every incentive to claim the opposite,
that it had been a wager all along. “It is by no means an easy matter,”
one commentator lamented, to figure out who was telling the truth.
The problem was that speculators typically did not wager on prices
explicitly, in contracts spelling out that the loser of the bet would pay
the difference in prices to the winner. Contracts were normally worded
as if the parties contemplated that a seller would transfer something
to a buyer; the parties’ intent to pay differences was merely implicit.
“Undisguised gambling can easily be proven,” a Chicago lawyer ob-
served, “but disguised gambling is more difficult of proof.” Such proof
was necessarily circumstantial. One treatise explained: “it becomes nec-
essary, to distinguish the real character of a contract, to look not merely
at its terms, but to the action of the parties to it, in fulfilling it; that
is, to regard not merely the agreement, but the transaction.” Judges
could not read contracts literally; they had to look behind the words,
to guess at what was in the minds of the signatories. What made this

33 T. Henry Dewey, A Treatise on Contracts for Future Delivery and Commercial


Wagers (New  York:  Baker, Voorhis & Co., 1886), 120; Kirkpatrick & Lyons
v. Bonsall, 72 Pa. 155, 158–​59 (1872).
80

80 Speculation

task so difficult was that there was nothing illegal about paying price
differences to settle a contract, so long as the parties had originally in-
tended to transfer some item at the time they signed the contract. “If a
contract for future delivery be made in good faith,” explained the New
Orleans lawyer Julius Aroni, “its validity cannot be affected by the fact
that afterwards the parties may agree to settle it by waiving delivery and
settling differences.”34 The fact that one party paid money to the other,
rather than delivering or accepting the item being sold, was thus not
enough to establish that the contract was unlawful. The payment of
money was relevant only if it revealed what the intent of the parties had
been back when they signed the contract.
One clue to someone’s intent was his past conduct. If the person
purporting to buy grain were a grain dealer, who had received ship-
ments of grain in the past, one might take his grain contracts at face
value, whereas if he had engaged in a series of gambling transactions in
the past, one might suppose that his contracts were mere vehicles for
wagering on prices. But “it would be very dangerous to permit the pre-
vious illegal conduct of one party” to invalidate a contract, cautioned
the Michigan judge Thomas Cooley. A contract was only void if both
parties intended to wager. And even an experienced bettor might genu-
inely buy grain, Cooley acknowledged, “for gamblers may make lawful
contracts as well as others.”35 Past conduct was a clue, but that’s all it
was. There was no easy way for judges to figure out what was in the
minds of people who entered into speculative contracts.
Further complicating matters was the circumstance that in all these
cases, someone was trying to wriggle out of a contract by claiming that
it had been illegal all along. Strictly speaking, the law paid no regard to
whether a person’s conduct was consistent with norms of appropriate

34 Francis A. Lewis, Law Relating to Stocks, Bonds, and Other Securities, in the
United States (Philadelphia: Rees Welsh & Co., 1881), 95; James C. McMath,
“Investment, Speculation and Gambling on the Fluctuations of the Market
Prices of Corporate Stocks and Other Commodities,” Central Law Journal 93
(1921): 225; Arthur Biddle and George Biddle, A Treatise on the Law of Stock
Brokers (Philadelphia: J. B. Lippincott & Co., 1882), 313; Julius Aroni, Futures
(New Orleans, La.: James A. Gresham, 1882), 10.
35 Gregory v. Wendell, 40 Mich. 432, 438 (1879).
  81

Betting on Prices 81

commercial behavior; if a contract was an unenforceable wager, the


loser did not have to pay off the winner. But this was often an intui-
tively uncomfortable outcome. To let a party off the hook, when one
suspected that the same party would have earnestly tried to enforce
the contract had the market gone in the opposite direction, seemed
to reward the most devious litigants. Because intent was so hard to
discern, and because a strict application of the law often favored the un-
ethical, judges tended to strain to find an intent to deliver. “Frequently
a court has failed to enforce the rule, for the avowed reason that in the
case at bar the wagering feature of the transaction was not clear to the
court,” one lawyer complained. “It must be confessed that some of the
cases in which this conclusion is reached are painfully unsatisfactory.”36
A case that arose in Maine provides a good example of the dilemma.
In 1872, Nicholas Berry of Bangor traveled to Chicago and visited the
board of trade, where he met a broker named Israel Rumsey. Berry
ordered Rumsey to sell ten thousand bushels of wheat, to be delivered
the following month. Berry was not in the wheat business and had no
wheat to deliver. Brokers on the board of trade, as on other exchanges,
protected themselves against loss by requiring clients to deposit a
“margin,” or a sum of money the broker could draw upon if the client
lost from the transaction. Berry left a margin with Rumsey, and when
wheat prices began to rise, Rumsey called upon Berry for an additional
margin, which Berry provided. Wheat prices continued to rise, but
when Rumsey demanded a further margin, Berry refused. Rumsey then
followed the normal practice at the board of trade. He repurchased the
wheat at the current price, at a loss of around $3,000. Berry’s margin
was only $700, so Rumsey sued Berry to recover the difference. Berry
defended on the ground that his supposed sale of wheat could not be
the basis for Rumsey’s suit, because it was an unenforceable wager.37
The Maine Supreme Court, by a vote of four to three, held that
Berry’s wheat sale was a valid contract. Justice Charles Danforth, writ-
ing for the majority, admitted that “it may indeed appear a little singu-
lar and even suspicious that a man residing in Bangor, having no wheat

36 “Option Sales,” Central Law Journal 10 (1880): 221.


37 Rumsey v. Berry, 65 Me. 570, 571 (1876).
82

82 Speculation

of his own, should undertake to sell and deliver wheat in Chicago.”


Nevertheless, Danforth concluded, “we cannot assume that anyone has
violated the law and been guilty of immoral and corrupting practices
in his business transactions, without proof, even though he may ask it
himself ”—​and here Danforth likely revealed the true reason for the
decision—​“for the purpose of being relieved from the obligation of a
losing contract.” In choosing whether to place the loss on Berry, who
had stiffed his broker, or Rumsey, who had done nothing wrong, it was
hard not to sympathize with Rumsey, even if that meant squinting a bit
to find an intent to deliver wheat.38
Of course, the difficulty of discerning intent also allowed judges the
same leeway to invalidate transactions when their instincts pointed in
that direction. A few years after the Maine court held Nicholas Berry
to his contracts, an Illinois court considered the seemingly similar case
of a young Chicago office clerk named McCurdy, who got in over his
head speculating at the board of trade. The Illinois court showed con-
siderable sympathy for McCurdy, who the court described as “a young
man … who was not, and never had been, engaged in buying or sell-
ing wheat, corn, pork, lard, or other like commodities” until a three-​
month binge at the board of trade left him owing his brokers $840,
nearly a year’s salary. The court held that his brokers must have known
that McCurdy had no intent to deliver or accept any commodities.
“It matters not how nicely the forms adopted may be adjusted to legal
requirements,” the court lectured, “if the transaction, when stripped of
its covering, discloses what is in substance a bet or wager on the future
price of grain.” The brokers thus could not collect on the debt.39
In cases like these, which presented common fact settings, courts
could plausibly impute to the parties either the intent to deliver or the
intent to gamble on prices. One suspects that the outcome often turned
on the judges’ sympathies for the particular parties before them, or—​as
in Ira Foote’s two cases discussed at the beginning of this chapter—​on
the judges’ unarticulated policy views as to the value of speculation
more broadly.

3 8 Id. at 573.
39 Beveridge v. Hewitt, 8 Ill. App. 467, 479, 482 (1881).
╇ 83

Betting on Prices 83

The Wickedness of€the Chicago Board of€Trade


Despite the difficulty of applying the intent-╉to-╉deliver rule, it was well
entrenched as a matter of common law by the late 1870s. In later years,
many states enacted statutes codifying the rule. Some of these laws
strengthened the rule by invalidating contracts even where only one
party had no intent to deliver. A few states restricted speculation even
more. The first was Illinois, which in 1874 banned the sale of options.
California’s constitution of 1879 prohibited stock futures. Georgia pro-
hibited short sales. Arkansas and Mississippi enacted statutes banning
all futures.40 To get a sense of the political forces motivating these pro-
visions, it is worth taking a close look at the two that were the most sig-
nificant, the Illinois statute barring options (significant because options
were staple fare at the Chicago Board of Trade) and the California con-
stitution’s ban on stock futures (likewise for the San Francisco Mining
Exchange). Both were enacted in response to local concerns, but both
were early appearances of issues that would become nationally salient
in later years.
The Illinois statute was enacted during a period of agrarian discon-
tent with the intermediaries between the farmer and the consumer—╉
primarily railroads and grain warehouses, which farmers believed
charged far too much. During this period, farmers in several states,
including Illinois, successfully lobbied for state regulation of railroad
and warehouse prices. Dealers were intermediaries in the grain trade
as well, and they too were viewed with suspicion by many farmers.
“We often hear the Boards of Trade, or what is known as dealing on
‘Change,’ denounced as a system inimical to the interests of farmers,”
declared M.  L. Dunlap at the 1873 convention of the Illinois State
Farmers’ Association. Dunlap, a farmer from Savoy, in the central part
of the state, was expressing the common view among farmers that trad-
ing in futures and options, especially short selling, had the effect of
depressing commodity prices. Dunlap himself was skeptical about this.
Nevertheless, he contended, “trading in options is no doubt a species of
gambling highly reprehensible, and with which the farmer has nothing

40 T. Henry Dewey, Legislation Against Speculation and Gambling in the Forms of


Trade (New York: Baker, Voorhis & Co., 1905), 15–╉50.
84

84 Speculation

to do.” As Dunlap saw it, “it is a practice that should not be permitted
among a class of men whose integrity is a large part of their capital.”41
In 1874, when the Illinois legislature convened to revise the state’s
statutes, agrarian pressure produced a bill to prohibit contracts for the
sale of any commodity that the seller did not own and actually have in
his possession at the time of the contract. This was the same rule first
suggested in the United States by Joseph Story thirty-​five years earlier,
when it had been roundly rejected. The proposal was directly targeted
at short selling on the Chicago Board of Trade, the single most vivid
circumstance in which a trader necessarily contracted to sell something
he did not yet own.
The bill was promptly criticized, for the same reason that Story’s
proposal had been criticized—​it banned good transactions, as well as
bad. “So far as such a law could be made to prevent wild speculations,
‘scalping’ options and their resultant ‘corners,’ it would prove benefi-
cial,” the Chicago Tribune argued, “but, unfortunately, the law is so
sweeping.” Farmers could not survive without making contracts to sell
grain before the harvest, the Tribune pointed out. “With such a law in
operation, it would require an amount of capital to transact business
which would simply be unattainable, and the farmers would soon find
themselves ruined.” And the law would yield even more absurd results
in other industries. A meat packer could not contract to deliver meat in
advance. An opera manager could not sell tickets in advance, because
the singers were not yet in the theater. A newspaper could not sell a
year’s subscription, because the publisher did not yet possess the entire
year’s news. “Such a law,” the Tribune lamented, “would put a complete
embargo on all business.”42
The bill was especially unpopular at the board of trade. Many of the
contracts made at the board were for large quantities, like a thousand
hogs or ten thousand bushels of grain. The board’s president wondered
how sellers could fit so many hogs or so much grain into their offices to
satisfy the “possession” requirement. The broker John Bensley thought

41 Lurie, The Chicago Board of Trade, 52–​56; Proceedings of the Illinois Farmers’
State Convention, Held at Bloomington, Ills., Jan. 15 & 16, 1873 (Chicago: Inter-​
Ocean Print, 1873), 22–​23.
42 Chicago Daily Tribune, 15 Jan. 1874, 4.
  85

Betting on Prices 85

the bill would only hurt farmers by reducing prices. “If the grain was
shipped in here, and left for sale to only those men who would come
forward and buy it for shipment, they would only buy when it had ac-
cumulated here to such extent as to be sold cheap,” he argued. “As it
is now, there is a constant market, and if they are not ready to take it,
speculators are, and they keep it up to a price higher than the country
shippers would be able to get for it.”43
These concerns sparked extended debate in the Illinois legislature.
On one side were rural members who, as the Tribune put it, “expatiated
on the wickedness of the Chicago Board of Trade, and the way in which
country innocents were fleeced.” On the other were urban legislators
who could see value in the board’s activities. The resulting compromise
yielded a weaker bill, one that prohibited options but preserved fu-
tures trading in commodities one did not yet own or possess. This was
the bill that became law in 1874. It made the sale of options a crime
punishable by fine or prison sentence up to a year, and declared that
options “shall be considered gambling contracts, and shall be void.” As
the Illinois Supreme Court explained several years later, the purpose of
the law was “to break down the pernicious practice of gambling on the
market prices of grain and other commodities.” There was “nothing
illegal or even immoral in an option contract within itself,” the court
noted. “The evil aimed at nevertheless grew out of such contracts,” and
for that reason Illinois abolished them.44
In California at approximately the same time, a different political
issue yielded a different prohibition. The Nevada mining boom of the
1870s was financed in large part by the sale of mining company stock
in California, especially in San Francisco, which was the tenth-​biggest
city in the country and home to the region’s only stock exchange.
“Californians, as a class, gamble in mining stocks,” the San Francisco
Chronicle observed in 1875. “From the lady leader of fashion to the
poorest seamstress, from the dignified bishop and the august Judge,
down to the boy who blacked their boots, each and all owned shares.”

43 Chicago Daily Tribune, 16 Jan. 1874, 2.


44 Chicago Daily Tribune, 21 Jan. 1874, 5; 23 Jan. 1874, 5; The Revised Statutes
of the State of Illinois (Springfield: Illinois Journal Co., 1874), 372; Schneider
v. Turner, 22 N.E. 497, 499 (Ill. 1889).
86

86 Speculation

As on other exchanges, people of modest means could speculate with


little capital, either by buying stock on margin or by entering into fu-
tures contracts to be settled on price differences, and as a result some
were ruined. “Mining is a legitimate business, while dealing in stocks
is gambling,” the Chronicle lectured. “Twenty thousand people have
lost where twenty have made money.” Worse, with all these sales came
ample opportunities for fraud. Sometimes promoters simply claimed
a mine was more productive than it really was, to induce purchasers
to overpay for stock. The Chronicle warned its readers not to believe
hucksters. “If the mine is worth more than its selling price in stock,”
the paper pointed out, “they would keep it” rather than selling. But the
operators of a mine had a more insidious way of fleecing sharehold-
ers, one that shareholders had scarcely any way of preventing. A mine
could deliberately be made less valuable. “It has very frequently hap-
pened,” the Chronicle charged, “that the managers of a mine, know-
ing very well that good ore was easily obtainable, would so direct the
work to lead away from the direction of the valuable deposits and into
the worthless rock. Sometimes the mine has been allowed to become
flooded.” Before doing such things, the mine operators would sell the
stock short, so when news of the mine’s failure reached the market, the
share price would plummet and the operators would profit at share-
holders’ expense.45
Fraud in the sale of corporate shares was thus an impor-
tant issue during the California Constitutional Convention of
1878–​79. Among the hundreds of clauses stuffed into California’s
lengthy Constitution of 1879 was one that voided “all contracts
for the sale of shares of the capital stock of any corporation or as-
sociation, on margin or to be delivered at a future day.” This was
the “one provision in the instrument which will commend it as
a whole to the mass of the people,” a Los Angeles newspaper de-
clared, because “it will do away with stock gambling.” “For this
alone,” the Chronicle agreed, the new constitution “will deserve the

45 Charles A. Fracchia, “The Founding of the San Francisco Mining Exchange,”


California Historical Society Quarterly 48 (1969): 3–​18; San Francisco Chronicle,
15 Sept. 1875, 2; 14 May 1873, 2; 10 May 1873, 2; 3 Mar. 1872, 1.
  87

Betting on Prices 87

support of every good citizen who wishes to see confidence restored


among businessmen.”46 For many years, California would be the only
state to prohibit futures and margin contracts in corporate stock.
These Illinois and California provisions had little effect in prac-
tice. Illinois’s ban on options must have been frequently flouted at
the Chicago Board of Trade, because there were perennial complaints,
both from within the board and outside, about how often options
were being bought and sold. In 1886, for example, approximately one
hundred members of the board found it necessary to petition their
board of directors to disallow options. The trading of options was
“dangerous and hurtful to our legitimate business, and prejudicial to
our good name as a commercial organization,” they declared. They
urged the board of directors to “use every possible means to suppress
this business.” A few years later, the Chicago Tribune likewise deplored
the prevalence of options at the board of trade. “The put and call busi-
ness is about on a level with the so-​called ‘trading’ in a bucket-​shop,”
the paper charged. Options were defended as a form of hedging, but
the Tribune was skeptical. “It is a notorious fact that the great majority
of those who buy privileges [that is, options] have nothing to insure,”
the Tribune observed. “The men who pay out their money for either
puts or calls do so merely as bets on the course of prices.” In California,
speculators continued to buy stock on margin despite the constitu-
tional ban on such purchases. The main effect of the constitutional
provision, recalled the broker John Percy, was to put every broker “ab-
solutely at the mercy of his client, who, if he lost, could simply turn
on the broker and recover all margins put up in the transaction.” The
situation was so “intolerable from the broker’s standpoint” that the
brokers pressed for an amendment to the state constitution, one they
finally procured in 1907. The amendment replaced the ban on margins
and futures with a new provision that simply restated the common-​law
intent-​to-​deliver rule.47

46 California Constitution of 1879, art. IV, sec. 26; Los Angeles Express quoted
in San Francisco Chronicle, 25 Mar. 1879, 1; San Francisco Chronicle, 29 Apr.
1879, 2.
47 Petition, June 1886, CBT, box 13, folder 29; Chicago Tribune, 11 Feb. 1892, 4;
John Percy to A. M. Clement, 25 June 1910, CBT, box 353, folder 8.
88

88 Speculation

The propriety of betting on prices was not just a legal question. The
experience of Illinois and California showed that under the right circum-
stances, it could be a significant political question as well. Before long, as
we will see in Â�chapter 3, it would assume a large role in national politics.

The Poor Man’s Stock Exchange


From the 1870s on, contracts were unenforceable—╉because they were
classified as wagers on price changes—╉when neither party intended
that the item ostensibly being purchased would actually be delivered.
Such was the rule on paper, but in a great many of the transactions on
the nation’s leading exchanges, and probably in most of them, neither
party had the slightest intention of delivering anything.
To be sure, contracts were carefully written to appear to comply with
the law. One member of the Chicago Board of Trade declared that his
grain contracts “all provide for delivery in Chicago.” When the Illinois
Supreme Court castigated the board of trade for selling “imaginary
wheat” that neither buyer nor seller ever intended to have delivered,
the board immediately protested. The court’s decision “grossly reflects
upon its members by characterizing contracts between them as gam-
bling contracts,” the board resolved at its next meeting. “The decision
appears to be based upon the erroneous inference” that board members
did not intend to fulfill contracts by delivery, the resolution insisted,
“when the contrary is the case.” In practice, however, delivery was a rare
event. William Warren, the president of the board of trade, estimated
that in the previous year he had bought or sold, on behalf of his clients,
five million bushels of wheat, two or three million bushels of corn, and
more than a million bushels of oats. He could not recall any of it having
been delivered. Walter Haugh, who was a member of the board of trade
for twenty-╉five years, reported that in his experience “there wouldn’t be
one per cent of the stuff delivered.”48
Most of the people buying and selling at the board of trade did not
actually want to deliver or receive the commodities in which they were

48 CBT, box 351, folder 1, page 3; J.B. Lyon & Co. v. Culbertson, Blair & Co., 83 Ill.
33, 40 (1876); CBT, box 3, folder 6; box 352, pages 935–╉36; box 352, page 182.
  89

Betting on Prices 89

transacting. Some were speculators, who wanted to receive (or pay, if


on the losing end) the difference between the price of the commodity
at the contract date and its price at the nominal delivery date. They had
no grain to deliver and no desire to acquire any.
Many of the others transacting on the board of trade were hedging, or
insuring against losses from price fluctuations. John Bradenbaugh, for
instance, was a grain merchant in Kansas City. His business consisted
of buying grain from farmers and then selling it at a profit to shippers.
Sometimes Bradenbaugh had to hold grain for several months before
selling it. If grain prices declined during that period, he would lose
money. When he bought grain, therefore, Bradenbaugh would simul-
taneously sell grain futures on the Chicago Board of Trade—​that is, he
would enter into a contract in which he nominally promised to deliver
grain at a future date for a stated price. That way, if grain prices fell, any
losses he suffered in his business of selling real grain would be offset by
his gains on the grain futures. (He would have gains on the grain futures
because if grain prices fell, he would have the right to acquire grain at the
now-​lower market price, and the right to receive the older higher price
when he delivered it.) Of course, by insuring against losses from a fall in
grain prices, Bradenbaugh was giving up the possibility of gains from a
rise in grain prices, because any gain he stood to make from a rise in the
price of grain would be similarly offset by losses on the grain futures. But
that was precisely the purpose of hedging. Bradenbaugh did not want
to speculate in grain prices. He wanted to conduct his business without
having to worry about price movements. “It simply means protection,”
Bradenbaugh explained, “against ups and downs of the market. If you
are short you buy and if you are long you sell.” Hedging was thus useful
to anyone in the chain of agricultural distribution who had to hold a
commodity for a period of time or who had to make commitments
to deliver a commodity in the future. “I never saw a firm that didn’t,”
Bradenbaugh observed. “They would be doing a very dangerous busi-
ness and treading on very dangerous ground if they didn’t.”49 Hedging
was the very opposite of speculating. Speculators sought to bear the risk
of price changes, while hedgers sought to avoid that risk.

49 CBT, box 351, folder 3, pages 3–​4, 13.


90

90 Speculation

Farmers also confronted the risk of price changes, so they too could
benefit from hedging. A  farmer had to decide whether to plant seed
long before there would be any crops to sell. A  sensible decision at
planting season could turn disastrous by the harvest if grain prices fell
in the interim. One way a farmer could avoid this risk was by find-
ing a customer who was willing to lock in a price and buy his crop
far in advance. But another method was to hedge. Like a grain dealer,
the farmer could simply sell futures in the planting season that would
come due at the harvest. By doing so, the farmer was, in effect, insuring
against the risk of price changes.
Hedgers, like speculators, normally did not intend to settle their fu-
tures contracts by delivering or receiving any grain. The grain futures
that Bradenbaugh sold at the board of trade nominally obligated him to
deliver a certain quantity of grain on a given day for a stated price. Once
Bradenbaugh no longer needed to hedge, however, he simply returned
to the board of trade and closed the transaction by purchasing grain
futures in the same amount as he had previously sold. When commodi-
ties were sold as hedges, Walter Haugh recalled, they were never actu-
ally delivered, and when they were bought as hedges they were never
actually received. Such contracts were settled the way Bradenbaugh set-
tled his, by a second transaction that was a mirror image of the first.50
The second transaction did not have to be with the same counterparty
as the first. When Bradenbaugh closed out his position by purchasing
grain futures, he could buy them from someone other than the person
to whom he had sold grain futures months before. At the end of each
day, when all the transactions on the board of trade were tallied up,
long and short positions would mostly cancel out, and all that would
be left was a list of who owed money to whom.
The board of trade’s rules explicitly permitted parties to settle their
contracts by the payment of money rather than the transfer of the com-
modities nominally purchased.51 The rules could hardly have stated

5 0 CBT, box 351, folder 3, page 6; box 352, page 201.


51 See, e.g., Act of Incorporation, Rules, Regulations and By-​Laws, of the Board of
Trade, Chicago, Ill. (Chicago: Dunlop, Sewell & Spalding, 1865), 14–​15, CBT,
box 43, folder 1; Act of Incorporation, Rules, By-​Laws and Regulations of the
Board of Trade of the City of Chicago (s.l.: s.n., 1896), 38, CBT, box 44, folder 1.
  91

Betting on Prices 91

otherwise without making speculation and hedging extraordinarily


cumbersome. The board of trade, along with the nation’s other com-
mercial exchanges, thus occupied a curious legal position. Most of the
transactions that took place there were unenforceable in court, because
in most, neither the buyer nor the seller contemplated delivery of the
commodity being sold.
One consequence was that members of the exchanges had to enforce
contracts themselves rather than relying on the courts to do it. This
was nothing new. The New York Stock Exchange had long operated a
miniature legal system of its own to enforce all the contracts that could
not be enforced in the New York courts under the New York stockjob-
bing statute of 1792, which voided contracts for the sale of stock the
seller did not own. When stock exchange members breached their con-
tracts, the exchange punished them by suspending or expelling them.
Exclusion from the exchange was a significant penalty, one that was at
least as capable of deterring breaches as anything a court could have or-
dered. The stock exchange was able to thrive despite the fact that most
of the promises made within its walls were not recognized as binding
by the legal system. Even after New  York repealed the stockjobbing
statute, the exchange continued to operate its own enforcement system,
which was probably faster and cheaper than resorting to the courts, and
which was staffed by fellow exchange members with expertise in the
ways of buying and selling stock.52
The Chicago Board of Trade adopted a similar system. The board
created an arbitration committee, made up of members elected for
two-​year terms, with the power to resolve contract disputes. Members
who disregarded the arbitration committee’s orders were subject to
suspension or expulsion. As in New York, disputes could be resolved
much more quickly by the committee than in the courts, in part be-
cause committee members knew the board of trade’s business much
better than any judge could. “It is not now, nor has it ever been the
policy of our firm to litigate in the courts,” declared one member of
the board of trade, “but on the contrary to submit our differences to”
arbitration within the board, where members could have their cases

52 Banner, Anglo-​American Securities Regulation, 270–​80; Constitution and By-​


Laws of the New-​York Stock Exchange (1865), 21, NYSE.
92

92 Speculation

decided by “less annoying and expensive tribunals.” For example, in


one 1877 case, a dispute between R. M. Weaver and the firm of Talbot
& Eckley, the arbitration committee heard all the evidence and issued
its order within a single day. The committee ordered Weaver to pay
Talbot & Eckley $178.75 in damages and an additional $10 in costs.
Much of the work of the board of trade’s board of directors consisted
of ruling upon such disputes. In May 1873, one typical month, three
members wrote letters to the board of directors complaining of con-
tract breaches on the part of other members, all of which most likely
involved contracts that were unenforceable in the court system.53 The
intent-​to-​deliver rule thus did not prevent the emergence and growth
of the board of trade and similar exchanges in other cities, organiza-
tions facilitating transactions in which scarcely anyone had any intent
to deliver anything.
The primary difficulty the intent-​to-​deliver rule created for the ex-
changes was not that it curtailed their ability to trade, but rather that
it greatly complicated their decades-​long battle with their primary
competitors, the bucket shops.54 A bucket shop was a simulated bro-
kerage, where customers could go to “buy” or “sell” stocks or com-
modities. The invention of the stock ticker in 1867 allowed anyone
with a telegraph connection to receive price information nearly in
real time. A bucket shop operator would buy a stock ticker and allow
customers to place wagers on price movements. Unlike a real broker,
who acted as the customer’s agent in transacting with a third party,
the operator of a bucket shop was on the other side of the customer’s
trade; the bucket shop stood in the same relation to its clients as a
casino to gamblers. “A legitimate broker actually buys or sells for his
customers,” explained Munsey’s Magazine, a popular monthly. “The

53 Lurie, The Chicago Board of Trade, 29; CBT, box 13, folder 31; box 4, folder 34;
Letters to Board of Directors from Howard Priestley (8 May 1873), J. K. Fisher
& Co. (24 May 1873), and Dupee & Hammond (29 May 1873), all in CBT,
box 1, folder 16.
54 The best account of bucket shops is David Hochfelder, “ ‘Where the Common
People Could Speculate’:  The Ticker, Bucket Shops, and the Origins of
Popular Participation in Financial Markets,” Journal of American History 93
(2006): 335–​58.
  93

Betting on Prices 93

bucket shop, on the contrary, simply bets its patrons that stocks,
grain, cotton, provisions, or any other speculative commodity will
go up or down.”55 Brokerage houses earned commissions on their
clients’ trades, whether the clients made or lost money. Bucket shops
took commissions too, but they profited mainly by winning whatever
their clients lost.
By the later decades of the nineteenth century, there were bucket
shops in towns all over the country. Some offered honest gambles
on prices without any pretense of doing anything else. Some falsely
claimed to be real brokers with membership in an established exchange.
Some were pure swindles, with concocted prices that always seemed to
move against the customer. All of them hurt the exchanges. The honest
bucket shops took legitimate clients away from the exchanges, while
the dishonest ones brought the entire enterprise of speculation into
disrepute.
Dishonest bucket shops used a variety of tricks. Prices came by tele-
graph from the cities where the exchanges were located. In small towns,
customers without telegraph connections of their own had no way of
knowing whether the prices posted by the bucket shop were the real
ones, so a bucket shop could make its prices rise or fall nearly at will.
“The city trader can compare bucket-​shop prices with the regular Board
of Trade figures, and thus to a degree protect himself against whole-
sale extortion,” the Chicago Tribune cautioned—​under the large head-
line “How Bucket Shop Men Fatten on Dupe Speculators”—​“but the
countryman is entirely at the mercy of the firm with which he deals.”
Or a bucket shop could “hold the market”—​that is, post prices only
after a delay, so the bucket shop operator would know the market’s
direction before his clients did. A bucket shop could even fleece its cus-
tomers with real prices. Customers typically purchased on very small
margins, putting down as little as 1 percent of the value of the stock
or commodity nominally being purchased, so a price movement of 1
percent in the wrong direction was enough to wipe out a customer’s ac-
count. A bucket shop operator seeking to make prices move by 1 percent

55 Patton Thomas, “The Bucket Shop in Speculation,” Munsey’s Magazine 24


(1900): 68.
94

94 Speculation

could buy from a confederate on the real exchange at the desired price,
which would be duly reported by telegraph. Once the customers’ ac-
counts had been emptied, the confederate would buy back at the same
price from the bucket shop operator. As the financial journalist Edwin
Lefèvre recalled in his thinly fictionalized biography of the speculator
Jesse Livermore, on the New York Stock Exchange “they had what were
frankly referred to as ‘bucket-​shop drives,’ when a stock was offered
down two or three points in a jiffy just to establish the decline on the
tape and wipe up the myriad shoe-​string traders who were long of the
stock in the bucket shops.” And if all else failed, and the market turned
in favor of the customers, the bucket shop operator “simply locks his
door at night and disappears from the town before morning,” one critic
charged, “leaving the traders who have taken profits, to bewail their
losses and their cupidity.”56
It was commonly observed that dishonest bucket shops harmed the
reputation of the established exchanges, because many customers could
not tell the difference between the two. As a special committee of the
New York Stock Exchange lamented, “the ignorant people who have
been robbed by the bucket-​shop feel that they have been robbed by the
Stock Exchange.” This was one reason the exchanges tried for decades
to stamp out the bucket shops. Dishonest bucket shops also harmed
the reputation of honest bucket shops, who thus had an interest in sup-
porting the exchanges’ efforts, so long as those efforts could be narrowly
targeted. An agent of the board of trade reported that the proprietor
of a Cincinnati bucket shop “says that he would be glad if the B. of
T. would go after several b[ucket] s[hop] houses who are doing business
on the bunco plan, as they are [the] kind who are giving houses who
do business on strictly business principles a black eye, and he would
like to see them all put out of business.”57 The dishonest bucket shops

56 Chicago Daily Tribune, 19 Sept. 1897, 37; Deposition of A. H. Sheldon, 1


June 1907, CBT, box 331, folder 6; Merrill A. Teague, “Bucket-​Shop Sharks,”
Everybody’s Magazine 15 (1906): 34; Edwin Lefèvre, Reminiscences of a Stock
Operator (1923) (Cutchogue, N.Y.: Buccaneer Books, 1976), 192–​93; “Bucket
Shops,” The Ticker 2 (1908): 22.
57 “The Doom of the Bucket Shop,” Independent 79 (1914):  452; “Bucket-​
Shops:  Attacked by the Stock Exchange,” Outlook, 6 Oct. 1915, 298; “The
  95

Betting on Prices 95

had no defenders. Everyone agreed that fraud was wrong and should
be stopped.
The exchanges had a harder time fighting off the honest bucket shops,
which took customers away by offering similar services at lower prices.
Bucket shops normally allowed trading in smaller lots than the ex-
changes did. They required smaller margins. They typically took a lower
rate of commission. Bucket shops could thus be an attractive alternative
to speculators and hedgers alike. John Bradenbaugh, the Kansas City
grain merchant, often hedged on the Chicago Board of Trade, but he
also often used a Kansas City bucket shop for that purpose. “So far as
we are concerned it was identically the same,” Bradenbaugh explained.
“Exactly the same in every particular.” From his point of view, “the
whole Chicago Board of Trade has been a bucket shop for the last ten
years.”58
Attitudes like Bradenbaugh’s were of great concern to the exchanges.
“The idea is assiduously cultivated by the proprietors of these places
that the bucket-​shop is the poor man’s stock exchange,” the New York
Stock Exchange complained. People who might otherwise have been its
own customers “feel that the bucket-​shop performs a useful function
in no way different from that of the great central Exchange.”59 The ex-
changes accordingly went to great lengths to suppress this competition.
They sponsored state legislation to outlaw bucket shops. They helped
prosecutors prepare cases against bucket shops. They took to the press
with public relations campaigns aimed at dissuading customers from
resorting to bucket shops. In all of these efforts, the first step was to
establish that the exchanges were fundamentally different from bucket
shops. The exchanges were sites of legitimate commerce, the argument
went, while the bucket shops were not, because the business of bucket
shops was betting on prices.

Bucket Shop Failures,” Bankers’ Magazine 104 (1922): 623; Special Committee


on Bucket Shops, “Digest of the Preliminary Work of the Special Committee
of June 25, 1913,” 7, NYSE; “Tom” to “Owen,” 17 July 1908, CBT, box 324,
folder 13.
58 New York Times, 2 Mar. 1878, 8; CBT, box 351, folder 3, pages 12–​13.
59 Special Committee on Bucket Shops, 7.
96

96 Speculation

The Chicago Board of Trade hammered this point home in a series


of pamphlets about bucket shops that it distributed to journalists and
legislators. “They are gambling houses, pure and simple,” the board in-
sisted. “Under the guise of a contract to buy or sell,” the board declared,
“the proprietor of the ‘shop’ will wager any comer that the price will
advance before it declines, or will decline before it advances.” On a real
exchange, the board argued, “mind meets mind, a contract is made,
enforceable under the law, just as bona fide a transaction as buying a
carload of lumber.” But in a bucket shop, “no transaction is made. It is
simply a bet.” An enforceable contract required at least one of the par-
ties to intend the delivery of the item ostensibly being sold, but that
could never be possible in a bucket shop. “Bucket shopping,” the board
concluded, was nothing but “gambling in prices.”60
This was a distinction that could hardly hold up to scrutiny, be-
cause the participants in most of the transactions on the established
exchanges likewise had no intention of delivering the items that were
nominally being sold. If betting on prices was the distinguishing fea-
ture of a bucket shop, then the exchanges were in large part bucket
shops too. “If the bucket-​shop man, or independent, violates the law
of Illinois, the Chicago Board-​of-​Trade member is equally guilty,” re-
sponded the bucket shop proprietor C. C. Christie. “The members of
that institution who climb upon a pedestal and assume a ‘holier-​than-​
thou’ expression at the mention of bucket-​shops or independents, are
confirmed lawbreakers.” In Christie’s view, the board of trade was “the
biggest bucket-​shop on earth.” The humor magazine Puck acknowl-
edged that the operators of bucket shops “simply bet on quotations”
but observed that the same was true on the exchanges, “for nineteenth-​
twentieths of Stock and Produce Exchange transactions are pure and
simple betting and gambling.” So long as a bucket shop dealt honestly
with its customers, agreed the Wall Street Journal, “we confess ourselves
unable to distinguish any difference in the moral standing of such con-
cerns and that of an exchange.” The Journal concluded that it could not

60 What Is the Bucket Shop?, CBT, box 326, folder 8, page 3; The Grain Exchanges
Versus Bucket Shops, CBT, box 326, folder 8, page 3; Evils of the Bucket Shop
System, CBT, box 326, folder 8, page 3; Why Bucket Shops Should Be Outlawed,
CBT, box 326, folder 8, n.p.
  97

Betting on Prices 97

“see what indictment can be sustained in the matter of principle against


the ‘bucket shop’ without it being likewise sustained as against the stock
exchange.” The exchanges emphasized gambling in their attacks on the
bucket shops because gambling was what made bucket shop transac-
tions unlawful, but by emphasizing gambling, the exchanges were only
inviting this sort of criticism, which hit them at their weakest point. As
the former broker Frederick Dickson pointed out, the pot was calling
the kettle black. “The bucket-​shop man makes money, much money,
out of the vanity and folly of fellows,” Dickson chortled, “and herein
is the vital point of difference between the two; for the stock-​exchange
man deems it quite irregular for any one to do this save a member of an
accredited exchange.” Indeed, the congruence between the two kinds
of organizations was close enough that some bucket shops were also
members of the board of trade, while many of the men who worked at
the New York Stock Exchange frequented bucket shops after hours.61
There were real differences between the exchanges and the bucket
shops, but betting on prices was not one of them. Some of the traders
on an exchange intended to deliver what they were selling or to receive
what they were buying, unlike in a bucket shop, where no one did.
Bucket shop operators had a greater incentive to cheat their customers
than did brokers on an exchange, because bucket shop operators gained
whatever their customers lost, dollar for dollar. Perhaps most important
for the broader economy, trades made on an exchange influenced the
prices of stock and commodities and in the long run could be expected
to cause their prices to reflect their true values, while trades made in a
bucket shop did not affect prices at all. But these distinctions were too
subtle for the exchanges to use in their campaign against the bucket
shops. The exchanges argued instead that the bucket shops should be
outlawed because they were gambling.

61 Ann Fabian, Card Sharps, Dream Books, & Bucket Shops:  Gambling in
19th-​Century America (Ithaca, N.Y.:  Cornell University Press, 1990), 198;
C.  C. Christie, “Bucket-​Shop vs. Board of Trade,” Everybody’s Magazine 15
(1906):  707, 708; “A Plea for Bucket-​Shops,” Puck, 10 Sept. 1884, 23; Wall
Street Journal, 10 Aug. 1904, 1; Frederick S.  Dickson, “The Poison of the
Street,” Everybody’s Magazine 20 (1909): 230; Chicago Daily Tribune, 31 Oct.
1895, 6; New York Times, 7 Oct. 1878, 2.
98

98 Speculation

Whenever the board of trade sued a bucket shop, this strategy left
the board’s leaders stammering under cross-​examination as they strug-
gled to articulate why bucket shops were any different from the board
of trade. “What is a bucket shop?” one lawyer asked William Warren,
the board’s president. Warren replied, “It is a place where dealings are
had upon the fluctuations of the market without any bona fide transac-
tion.” The lawyer continued: “Would it be a bona fide transaction if the
parties did not intend to receive or deliver?” Warren said it would not.
And here the lawyer moved in for the kill. He and Warren both knew
that the transactions he was describing were common on bucket shops
and the board of trade alike. “But if they did intend to receive or de-
liver,” the lawyer asked, “and did then settle on the difference in market
prices, that would be a bona fide transaction?” All Warren could do was
feebly reply: “Not in a bucket shop.” The lawyer: “Would that trans-
action have to be conducted on the Board of Trade in order to make
it bona fide?” Warren: “I think it would.” The president of the board
of trade had been forced to concede that there was no real difference
between the business of his organization and that of a bucket shop. In
the board’s view, the very same transaction was lawful on the board and
illegal anywhere else.62
When the Chicago Board of Trade and the New  York Stock
Exchange tried to help prosecutors go after bucket shops for violat-
ing the laws against gambling, they faced the same problem of dis-
tinguishing the bucket shop business from their own. The board of
trade sent George Burmeister all over the country to explain to dis-
trict attorneys and grand juries why bucket shops were different from
exchanges. “I take pleasure in telling you that I done my little stunt
before the Grand Jury today,” Burmeister reported from Detroit. “I
was first and had to give them a wide-​scope outline of a legitimate
house and a bucket-​shop, the way we execute orders and the way they
are filled in a bucket-​shop. Several very inquisitive fellows on it but
I gave a most satisfactory explanation; the two assistant prosecutors
both complimented me after we all got out saying they both learned
a heap.” From Cleveland, Burmeister reported: “I have spent all my

62 CBT, box 352, pages 926–​29.


  99

Betting on Prices 99

time in the District Attys office here while they are writing indict-
ments.” He lamented that “this case is entirely strange to all of them,
the Dist Atty, Asst Dist Attys and the Inspector have never made a
trade or even been in a Brokerage office or Bucketshop and practically
had to learn the entire business.”63
The same problem arose when the exchanges lobbied state legisla-
tures for statutes banning bucket shops. “I went to Sacramento for
the purpose of securing the introduction of a bill to put the bucket-​
shops out of business,” John Percy reported to the board of trade. “I
was personally acquainted with a large number of the members of
both houses, and interviewed them on this subject.” But Percy had no
success, because he could not persuade legislators that bucket shops
were any different from the board. “In practically every instance,” he
despaired, “I found that the members were entirely ignorant as to
the difference between bucket-​shops and legitimate brokerage houses,
and all seemed to have the idea that a marginal trade through a le-
gitimate broker was, to all intents and purposes, as fictitious as the
so-​called ‘purchases’ and ‘sales’ in the bucket-​shops. The general im-
pression seemed to be that the legislation I  was urging was simply
an attempt on the part of the big fellows to squeeze out the little
ones.” The grain dealer Harry Kress was a board of trade member who
lived in Piqua, Ohio. He tried to persuade his state legislature to ban
bucket shops as gambling houses, but he was not optimistic, because
most people thought the board’s transactions were gambles too. “The
average individual having no connection with the Grain Trade, does
not understand the necessity of trading in futures,” Kress complained,
“and right here is where we may expect trouble.” The proponent of
an Indiana bill to prohibit bucket shops acknowledged that “the first
question that will naturally arise in the minds of my colleagues is,
what is the difference between a bucket shop and a legitimate board
of trade[?]‌” The bucket shops themselves did much to promote the
impression that bucket shops and exchanges were identical. As one
Illinois legislator observed, the bucket shops’ standard argument

63 George Burmeister to H.  A. Foss, 16 Dec. 1912, CBT, box 329, folder 1;
George Burmeister to J.  C. F.  Merrill, 13 Oct. 1914, CBT, box 329, folder
6. Burmeister’s reports fill several folders in CBT, box 329.
100

100 Speculation

against this sort of legislation was “We are bad of course, rotten in
fact, but no worse than the board of trade.”64
The exchanges were walking a very thin line. They were trying to
stamp out a business that was not very different from their own, based
on the one attribute that exchanges and bucket shops shared in abun-
dance. Despite this obstacle, several states did enact statutes prohibiting
bucket shops in the late nineteenth and early twentieth centuries. But
this only converted the problem into one of legislative drafting, be-
cause any such statute had to distinguish bucket shops from exchanges.
Some of these laws did not even try to define the term bucket shop.
In Iowa, for example, the legislature simply declared that its purpose
was to prohibit “places commonly known and designated as bucket-​
shops.” The statutes that defined the term did so by restating the fa-
miliar common-​law ban on transactions where neither party intended
delivery. Wisconsin, for instance, declared that “a bucket-​shop, within
the meaning of this act, is defined to be an office, store or other place”
in which transactions were made “for the purchase or sale of any such
commodity wherein the parties do not contemplate the actual or bona
fide receipt or delivery of such property, but do contemplate a settle-
ment thereof based upon differences in the prices at which said prop-
erty is, or is claimed to be, bought and sold.” The statutes added new
penalties to bucket shopping, but they did not make anything unlawful
that was not already unlawful. They merely replicated the problem of
how to distinguish bucket shops from legitimate exchanges, because by
this definition, the exchanges were bucket shops too. “We do not see
how any legal distinction can be made between the two operations,” the
New York Times noted shortly after New York enacted its statute. Was
the business of a bucket shop “betting on future prices? In form it is
not, and in substance it is no more so than the ‘regular’ speculation” on
the New York Stock Exchange. After more than a decade of ineffective
enforcement, the Times concluded that no law could be drafted to ban
bucket shops without also banning the exchanges. “Bucket shops in a

6 4 John A. Percy to A. M. Clement, 25 June 1910, CBT, box 353, folder 8; Harry
W. Kress to A. H. Schuyler, 19 Feb. 1908, CBT, box 331, folder 7; CBT, box
353, folder 3; “Anti-​Bucket Shop Bill,” CBT, box 353, folder 9.
  101

Betting on Prices 101

financial community are something like sorrel in one’s front lawn,” the
paper mused. “No general methods can be taken to exterminate them
without danger to the legitimate growth surrounding them.” Even the
New York Stock Exchange lost faith in the ability of the legal system to
stamp out bucket shops. A stock exchange committee appointed to rec-
ommend some course of action regarding bucket shops concluded that
“these laws are not enforced and never can be enforced.” The bucket
shops were too popular, and they were popular because they provided
the same service as the stock exchange itself. New York and Pennsylvania
had enacted statutes prohibiting bucket shops, largely at the stock ex-
change’s behest, but “there are more bucket-​shops in New York State
and Pennsylvania today than there ever were before.”65
The exchanges eventually won the war against the bucket shops, not
by persuading anyone that they were fundamentally different, but by
cutting off the bucket shops’ access to the exchanges’ price quotations.
A bucket shop could not exist without timely price information, so if
the exchanges could prevent the bucket shops from receiving price quo-
tations, they could put the bucket shops out of business. But that was
easier said than done. The exchanges repeatedly tried to persuade the
telegraph companies to deny service to nonmembers, without much
success, because the bucket shop business was simply too lucrative for
the telegraph companies. In 1890, the board of trade took the extreme
step of stopping the telegraphic distribution of price quotations, an ex-
traordinary inconvenience to the board’s many out-​of-​town members
who were accustomed to receiving prices by telegraph. But the board
quickly discovered that the bucket shops were getting the quotations
anyway, apparently because someone inside the board was flashing sig-
nals out the window. The next day the board tried soaping the windows,
but even that desperate measure only delayed the bucket shops by about
fifteen minutes. The board resumed telegraphing its prices. In 1898,
when the board suspected that someone was using the new telephones

65 Morris J.  Cashel, “Bucket Shops,” American Political Science Review 2


(1907): 48–​50; T. Henry Dewey, Legislation Against Speculation and Gambling
in the Forms of Trade (New York: Baker, Voorhis & Co., 1905), 51–​65, 54, 63;
New York Times, 13 June 1889, 4; 22 May 1904, FS3; Special Committee on
Bucket Shops, 154, 157.
102

102 Speculation

on the exchange floor to leak price quotes to bucket shops, there was a
proposal to remove them, which drew complaints from board members
all over the country that without telephone contact they could not com-
pete with the bucket shops, because the telephone was the only way they
could get prices faster than the bucket shops did.66
Toward the end of the century, the exchanges turned to a different
tactic. They began filing suits against bucket shops, alleging that the
price quotations were their property and that the bucket shops had
stolen it. These suits had a mixed record at first, but in 1905 the ex-
changes scored a final victory in the US Supreme Court, which affirmed
that the prices resulting from transactions on an exchange were prop-
erty belonging to the exchange. The case was brought by the Chicago
Board of Trade against the Christie Grain & Stock Company, a Kansas
City bucket shop that was presumably obtaining board price quota-
tions by telegraph, despite a provision in the board’s contracts with the
telegraph companies not to furnish price information to any bucket
shop. The board’s “collection of quotations is entitled to the protection
of the law,” Justice Oliver Wendell Holmes wrote for the court. “It
stands like a trade secret. The plaintiff has the right to keep the work
which it had done, or paid for doing, to itself.” By sending prices over
the telegraph wires to members throughout the country, Holmes con-
tinued, the board had not given up this privilege of secrecy. “The plain-
tiff does not lose its rights by communicating the result to persons, even
if many, in confidential relations to itself, under a contract not to make
it public,” the court held. “Strangers to the trust”—​that is, the bucket
shops—​“will be restrained from getting at the knowledge by inducing a
breach of trust, and using knowledge obtained by such a breach.”67 The
exchanges had established their right to control their price quotations.
The Christie Grain & Stock Company’s primary defense was that
the board of trade “itself keeps the greatest of bucket shops,” that its

66 Chicago Daily Tribune, 18 Nov. 1890, 6; CBT, box 37, folder 25.


67 Kiernan v. Manhattan Quotation Telegraph Co., 50 How. Pr. 194 (N.Y. Sup. Ct.
1876); New York & Chicago Grain & Stock Exchange v. Board of Trade, 19 N.E.
855 (Ill. 1889); National Telegraph News Co. v. Western Union Telegraph Co., 119
F. 294 (7th Cir. 1902); Chicago Board of Trade v. Christie Grain & Stock Co., 198
U.S. 236, 250–​51 (1905).
  103

Betting on Prices 103

price quotations were thus the product of illegal conduct, and that even
if prices were property when created lawfully, the board should not be
allowed to profit from its own illegality. In the course of rebuffing this
defense, Holmes launched into an extended discussion of the legitimacy
and importance of speculative trading on organized exchanges, using
words that would be quoted liberally by the exchanges in the years to
come whenever the opportunity arose. “In a modern market, contracts
are not confined to sales for immediate delivery,” Holmes began. “People
will endeavor to forecast the future, and to make agreements according
to their prophecy. Speculation of this kind by competent men is the self-​
adjustment of society to the probable. Its value is well known.” Even if
some of the contracts on the board of trade were unenforceable wagers, he
continued, “there is no doubt that a large part of those contracts is made
for serious business purposes.” The idea that the board’s transactions were
mere wagers “seems to us hardly consistent with the admitted fact that
the quotations of prices from the market are of the utmost importance
to the business world.”68 Of course, many bucket shop transactions were
made for business purposes that were just as serious, but unlike transac-
tions on the board of trade, they did not set prices for the country.
Holmes’s opinion in the Chicago Board of Trade case was a strong
endorsement of the exchanges, but it did not change the law. Most
of the transactions on the exchanges were still unlawful, just as they
had been since state courts began adopting the intent-​to-​deliver rule
a few decades before. The Supreme Court had ratified the prevailing
view among the lower courts that price quotations were the exchanges’
property, but that did not stop the bucket shops from continuing to
use every means at their disposal to get the quotations anyway. It would
be another decade before the exchanges finally won the war. The key to
victory was probably not the spate of anti–​bucket shop statutes or the
Supreme Court’s formal recognition of a property right in price quota-
tions, but rather the increasing size and thus clout of the exchanges,
which eventually enabled them to force the telegraph companies to
restrict the distribution of price information. When the bucket shops
were starved for prices, most of them died.

68 Chicago Board of Trade v. Christie, 246, 247–​49.


104

104 Speculation

“Time was when the bucket shop was almost as the locust in a sev-
enth year,” the Saturday Evening Post recalled in 1916. “Every consid-
erable town had from one to forty concerns where the inexperienced
and credulous could bet away their money on variations in the price
of grain, cotton, stocks.” But not any longer. “Our impression is that
the species has been pretty nearly exterminated. Certainly it persists
only sporadically and in a meager fashion.”69 Bucket shops never disap-
peared completely, but they ceased to be serious competition for the
exchanges.
The intent-​to-​deliver rule had not disappeared, however. It was still
a part of the common law throughout the country, and it had been
enacted into statute in many states as well. The rule drew a line—​a very
fine line, as many saw it—​between legitimate and illegitimate spec-
ulation, between speculation considered as business and speculation
considered as gambling. Meanwhile, even before the bucket shop war
ended, the exchanges had to confront another challenge, in the form
of a concerted national effort to redraw the line so as to prohibit even
more forms of speculation.

69 “A Scientific Reform,” Saturday Evening Post, 13 May 1916, in CBT, box 353,
folder 10.
3
Q
The Anti-​Option Era

Speculation in agricultural commodities was a major national


political issue from the early 1890s through the early 1920s. Year after
year, Congress considered—​and sometimes nearly passed—​bills that
would have drastically curtailed speculation. These were popularly
known as “anti-​option” bills, but they addressed more than options.
Their real targets were the commodity exchanges in cities throughout
the country, which would likely have had to shut down had any of the
bills before the 1920s become law. The controversy did not subside until
1921, when Congress established a new federal regulatory framework for
commodity trading, the basic features of which remain in place today.
The political support for these measures came primarily from farm-
ers, many of whom were suffering from sharp drops in the prices they
received for their crops. Between 1888 and 1894, wheat fell from 93 cents
a bushel to 49 cents, and cotton from 8.5 cents a pound to 4.6 cents.
Corn declined from 50 cents a bushel in 1890 to 21 cents in 1896, while
oats dropped from 42 cents a bushel to 18 cents. The resulting agrarian
discontent took various forms, including most famously the creation of
a new political party, the People’s (or Populist) Party, which achieved
considerable success in congressional elections during the 1890s.1

1 Susan B. Carter et al., eds., Historical Statistics of the United States, Millennial
Edition On Line (Cambridge University Press, 2006), tables Da719, Da757,
Da697, Da672; Lawrence Goodwin, Democratic Promise: The Populist Movement
in America (New York: Oxford University Press, 1976).

105
106

106 Speculation

One of the farmers’ grievances involved commodity speculation.


Crops had once been sold locally, and thus prices had been determined
in a local market, but by the late nineteenth century crops were shipped
around the world, and prices were determined in a global market, me-
diated through the commodity exchanges newly established in the na-
tion’s major cities. Commodities were traded at the Chicago Board of
Trade, at the New  York Produce Exchange, and at similar exchanges
in Minneapolis, Duluth, Milwaukee, St. Louis, Kansas City, Toledo,
Philadelphia, and other cities. “Exchanges have sprung up in many
places,” the Bankers’ Magazine marveled in the 1880s. “Within a few
years speculation has spread out enormously in many directions.”2
When farmers sought to understand why they had been impoverished
by falling prices, these exchanges were natural institutions to blame.
Farmers and their supporters could draw upon the long tradition of
thought critical of speculation.
The result was a three-​decade national debate about speculation. The
arguments on both sides scarcely changed from one year to the next.
In 1921, John Hill observed that “the subject of future trading has been
almost constantly before Congress since 1890 and hearings such as those
had in recent months are mere repetition.” Hill had been defending the
Chicago Board of Trade for years, and he was tired of it. “There is no
variation in the charges, no variation in the arguments of the defense,”
he complained. “All the claims of both sides can be found in the hear-
ings of 1892.” In 1908, when the New York Cotton Exchange mounted
a public relations campaign against proposed legislation, it merely had
to reprint the open letters it had published in the early 1890s. By the
end of this period, these recurring rural calls to restrict speculation were
a source of some amusement, and no small amount of condescension,
on the part of urbane defenders of the status quo. “There breaks forth
periodically from prudish but well-​meaning people a protest against
all forms of speculation,” noted the editor of the New York Financial
World. “The word ‘speculation,’ to the impassioned social reformer, is

2 Jerry W. Markham, The History of Commodity Futures Trading and Its Regulation
(New York: Praeger, 1987), 7–​8; “Speculation and Money,” Bankers’ Magazine
and Statistical Register 37 (1883): 811.
╇ 107

The Anti-Option Era 107

like red to a bull,” agreed the philosopher Robert Hutcheon. “All his
latent belligerency against things as they are is roused into strenuous
action when he hears the word or reflects on the crazy mentality and
the reckless activities it stands for.”3
But there was much more than ignorance behind the farmers’ call to
prohibit speculation in their products. The anti-╉option movement was
part of a broader battle over the organization of agriculture, fought by
an occupational group who could see their relative numbers dwindling
and their influence in the market slipping away. Many of the farmers’
arguments against speculation may have been wrong as a matter of
economics, but there was nothing irrational about them as a matter of
politics.

Hayseed Legislation
William Vilas served only a single term as a senator from Wisconsin, but
it lasted from 1891 to 1897, when the agitation for restricting commod-
ity speculation was at its peak, so he found himself bombarded with
appeals from both sides of the struggle. Political considerations pulled
Vilas in both directions. There were many farmers in Wisconsin, but
Milwaukee was already a big city, and the Milwaukee Grain Exchange
was a center of grain speculation, so any position Vilas took would be
sure to alienate a substantial fraction of voters and a good number of
the state legislators who still had the power to select Wisconsin’s sena-
tors. Vilas was a Democrat, in an era when the Democratic Party, espe-
cially in the West, was identified with the small farmer, but Vilas him-
self was about as far from the farm as a Wisconsinite could be. He had
been a lawyer, a law professor at the state university in Madison, and
then postmaster general and secretary of the interior during the first
Grover Cleveland administration. He had much more in common with
the bankers and brokers of Milwaukee than with the state’s farmers.

3 John Hill Jr. to E. F. Ladd, 17 June 1921, CBT, box 328, folder 3; Bills in Congress
Affecting Cotton Contracts for Future Delivery (New  York:  Latham, Alexander
& Co., 1908); Louis Guenther, Investment and Speculation (Chicago: La Salle
Extension University, 1921), 157; Robert J. Hutcheon, “Speculation, Legitimate
and Illegitimate,” International Journal of Ethics 32 (1922): 289.
108

108 Speculation

Throughout 1892 and early 1893, Vilas received a steady stream of


letters in support of the pending bills to curtail speculation. All were
from farmers or their representatives. George Stowe, the vice president
of the Wisconsin Farmers’ Alliance, urged Vilas to “stop the gambling
in grain & provisions.” Stowe reported that the state’s farmers “feel that
the gamblers on the Boards of trade have no right to sell produce they
do not have nor ever expect to have and by selling so many millions
of bushels of grain more than is in the whole world it is a very great
wrong.” The publisher of the Western Rural and American Stockman
declared that although the cities opposed the bills, “the country is prac-
tically a unit in favor of anti option legislation,” which was supported
by “the farm organizations of the northern states, five in number, rep-
resenting a membership of upwards of one million members, practi-
cally all voters.” As another rural correspondent pointed out, “it is quite
noticeable that the chief opposition to these bills comes almost wholly
from money centers and capitalists, and from organizations of grain
and produce or stock speculators.”4
Vilas received an equally large set of letters from opponents of the
legislation. Many came from the Milwaukee Chamber of Commerce,
whose more than six hundred members voted unanimously to oppose
the bills. Vilas heard arguments against restricting speculation from
grain traders, from bankers, from the Chicago Board of Trade, and
from the Toledo Produce Exchange. “I do not wonder,” he confessed
to the president of the Milwaukee Chamber of Commerce, “that there
is a rising feeling on the part of farmers and producers that something
ought to be done in their interest.” But he thought the legislation would
do little to help farmers, so he opposed it, after trying to mollify rural
voters by declaring on the Senate floor that his opposition was based
not on the bill’s merits but on his view that it was unconstitutional.5

4 G. E. Stowe to Vilas, 9 May 1892, box 14, folder 1; Milton George to Vilas, 6
July 1892, box 14, folder 1; N. F. Fox to Vilas, 29 Feb. 1892, box 13, folder 1, all
in WFV.
5 E. P. Bacon to Vilas, 5 Mar. 1892, 29 Apr. 1892, and 3 Dec. 1892, and “A Protest”
(27 Apr. 1892), all in WFV, box 14, folder 1; Robert Eliot to Vilas, 15 Jan. 1892,
WFV, box 13, folder 1; John Johnston to Vilas, 17 Feb. 1892, WFV, box 13, folder
1; George Stone to Vilas, 30 Dec. 1892, WFV, box 15, folder 1; Denison B. Smith
  109

The Anti-Option Era 109

The rural–​urban divide that Vilas encountered was representative of


the nation as a whole. The urban northeast, from Massachusetts down
to Maryland, was the only part of the country from which a majority
of members of Congress voted against restricting speculation in the
early 1890s. The idea was popular everywhere else. In 1890, the House
of Representatives first considered a bill to impose a prohibitive tax
on people who sold agricultural products they did not own, a bill that
would have put an end to commodity trading on all the boards of trade
and produce exchanges. Petitions of support poured in from farmers’ al-
liances in Iowa; from a group of farmers in Indiana who castigated “the
gigantic gambling devices known as short selling”; from the citizens of
Wyandotte County, Kansas, and Buffalo County, Wisconsin; from the
farmers of Spokane County and Lincoln County in Washington; and
from a group of Indianans who declared that “the so-​called business of
dealing in options as it is commonly practiced on boards of trade is as
clearly a mere gambling device as a ‘wheel of fortune’ or faro lay-​out.”
The primary opposition came from the New York Produce Exchange,
the New York Cotton Exchange, the Chicago Board of Trade, and the
New Orleans Cotton Exchange.6
House opponents of the 1890 bill were able to keep it from coming
to a vote, but the legislation returned in the next session of Congress,
in the summer of 1892, with the same configuration of supporters and
opponents. The farmers were in favor, while the commodity exchanges,
joined by many banks, were against. This time, bills were passed by both
the House and the Senate. An election was looming, and both parties
were anxious not to lose votes to the Populists, who nominated their own
presidential candidate, as well as candidates in many congressional dis-
tricts. “The passage of the Hatch Anti-​Option bill in the House is simply
a Democratic bid for the favor of the Farmers’ Alliance,” charged the St.
Louis Globe-​Democrat. “It does not mean that the men who spoke and

to Vilas, 11 Mar. 1892, WFV, box 13, folder 2; Vilas to E. P. Bacon, 7 Mar. 1892,
WFV, box 14, folder 1; New York Times, 6 Jan. 1893, 3.
6 Cedric B. Cowing, Populists, Plungers, and Progressives: A Social History of Stock
and Commodity Speculation, 1890–​1936 (Princeton, N.J.: Princeton University
Press, 1965), 22; Journal of the House of Representatives of the United States, 51st
Cong., 2nd Sess. (1890), 69, 118, 114, 169; New York Times, 12 Apr. 1890, 3.
110

110 Speculation

voted for it are anxious to promote the interests of the agricultural class,
but only that they think to help their party by humoring a silly preju-
dice against a certain form of trading.” The Pittsburg Dispatch sarcasti-
cally referred to the bills’ House and Senate sponsors as “Farmer Hatch”
and “Farmer Washburn,” despite the fact that neither William Hatch of
Missouri nor William Washburn of Minnesota was a farmer; Hatch
was a lawyer and Washburn engaged in a variety of businesses. The bill
was “hayseed legislation,” cracked one member of the New York Produce
Exchange. “The farmer, as a rule, imagines everyone in the city is his
enemy, and there are just enough demagogues lying around loose to en-
courage him in this opinion.”7
The Senate’s version of the bill was slightly different from the
House’s, so it had to return to the House, but by then the 1892 election
had already taken place, so there was less need to make a conspicuous
show of supporting the farmers. Opponents of the bill were able to stall
until the session expired. “By persistently postponing a vote on the bill
to prohibit dealings in options and futures, the minority of the mem-
bers of the House accomplished its defeat,” a relieved New York Times
reported.8
Proponents of a national ban on commodity speculation would
never come so close again. Hatch returned with a similar bill in 1894;
it passed the House but never came to a vote in the Senate. Another
bill in 1895 could not get through the House. Commodity prices lev-
eled off and even rose a bit over the next few years, so there would be
no more antispeculation bills until 1903, but then they started appear-
ing in increasing numbers. The Panic of 1907 elicited no fewer than
twenty bills to prohibit futures trading, followed by thirteen more in
1909 and another eight in 1911. In 1912, the Democratic Party platform

7 Journal of the House of Representatives of the United States, 52nd Cong., 1st Sess.
(1892), part II, 251, 255, 267, 270, 275, 277, 278, 282; Dealings in “Options” and
“Futures”: Protests, Memorials and Arguments Against Bills Introduced in the Fifty-​
Second Congress (s.l.: New York Cotton Exchange et al., 1892); Globe-​Democrat
quoted in Chicago Daily Tribune, 13 June 1892, 4; Pittsburg Dispatch, 20 Feb.
1892, 1; New York Times, 18 July 1892, 10.
8 New York Times, 2 Mar. 1893, 5.
  111

The Anti-Option Era 111

even included a pledge to “suppress the pernicious practice of gambling


in agricultural products by organized exchanges.”9
Low prices were the immediate cause of the farmers’ anger, so many
of their complaints about commodity speculation focused on its effects
on prices. Critics made a variety of claims about the relationship be-
tween speculation and prices, not all of which were consistent, but all
of which proposed mechanisms by which speculation harmed produc-
ers in the end.
One common contention was that speculators had the power to
manipulate prices to their own advantage. “Prices are almost entirely
artificial and are controlled by powerful cliques of speculators,” one
critic charged when cotton and wheat prices began to fall. When
prices rose, the Michigan Farmer blamed “the power of speculators to
force the market of a great staple commodity up.” Speculators were
alleged to manipulate prices by conspiring to buy and sell at prices
that varied from the true value of a commodity, and by securing the
publication of false crop reports in the newspapers. Some critics con-
tended that speculators’ goal was to depress the prices paid to the
producers of commodities. “There can be no doubt that the price of
our food products is fixed by the sales of futures or options,” insisted
Representative Edward Funston of Kansas, who was a farmer himself.
“The bull and the bear work together to keep prices down for unless
they are down, there is no profit to the buyers and sellers.” Other crit-
ics claimed that speculators’ real interest was to engineer fluctuations
in prices, because that is how they made their profits. “Whether it was
up or down, it was always to the detriment of the farmer,” lamented
Representative Jasper Tincher, another Kansas farmer in Congress. “If
it was up it was to the detriment of the farmer who sold his grain the
day before, and if it was down it was to the detriment of the farmer
who was going to sell some the next day.” Whether speculators were
depressing prices or making them fluctuate, critics were certain that

9 H.R. Rep. No. 845, 53rd Cong., 2nd Sess. (1894); H.R. Rep. No. 1999, 53rd
Cong., 3rd Sess. (1895); Carl Parker, “Governmental Regulation of Speculation,”
Annals of the American Academy of Political and Social Science 38 (1911):  141–​
44; Democratic platform at http://​www.presidency.ucsb.edu/​ws/​index
.php?pid=29590.
112

112 Speculation

speculation caused prices to diverge from the true values they would
have had in a market free from manipulation. “We neither ask for our
produce, and are unwilling to pay more than the actual value,” de-
clared T. J. Kimbrough, master of the Georgia State Grange. “Supply
and demand should control.”10
Opponents of legislation replied that while speculators might be
able to manipulate prices for very short periods of time, they lacked
the resources to do so for long. “With regard to the present price of
cotton,” argued the cotton speculator Daniel Sully, “no plan of ma-
nipulation, unless financed on a scale sufficient to take over a large
portion of the crop,” could keep prices higher or lower. “No clique of
speculators could be found with money enough and daring enough to
attempt such a thing.” As the economist Richard Ely put it, “buying
and selling in themselves cannot raise prices. … If it were possible thus
to raise prices, there would be an easy road to fortune for everyone.”
Speculators thus could not be blamed for the long-​term decline in com-
modity prices. Others repeated the familiar observation that the effect
of speculation was not to increase fluctuations but to smooth them out,
because speculators bought in a declining market and sold in a rising
one. Without them, reasoned the journalist Harold Howland, “booms
would tend to be wild and unrestrained, with inevitable disaster when
the break finally came,” while “declines would be sudden, rapid, and
extensive, with equally disastrous consequences.” The financial writer
Charles Conant, anticipating a way of thinking that would become
orthodoxy later in the twentieth century, provided a second method by
which speculators smoothed price changes. “Every event in the world
which affects price is felt upon the exchanges; and it comes to be said
that future events, when they occur, have been already ‘discounted,’ ”

10 “Speculation and Prices,” Independent 34 (1882): 22; “The Fruit of Speculation,


Michigan Farmer 45 (1904):  301; “When Speculation Is Immoral,” Current
Literature 44 (1908): 67; “Newspapers as Aids to Speculation,” Ohio Farmer
64 (1883): 328; Omaha Daily Bee, 9 Apr. 1890, 1; J. R. Dodge, “The Discontent
of the Farmer,” Century 43 (1892):  454; Richard Wheatly, “The New  York
Produce Exchange,” Harper’s New Monthly Magazine 73 (1886):  217; Future
Trading: Hearings Before the Committee on Agriculture, House of Representatives,
67th Cong., 1st Sess. (1921), 5–​6; “Gambling in Futures,” Southern Cultivator
48 (1890): 496.
  113

The Anti-Option Era 113

he noted. “Many important events which would cause a sudden break


or rise in prices, if they came like thunderbolts from a clear sky, are
thus ‘discounted’ before they actually take place. This fact is the highest
compliment to the organization of the exchanges, and the best proof
of their usefulness. The mere suspicion in well-​informed quarters that
wheat is to be scarce six months hence leads to its gradual rise in price.”
The grain trader Egerton Williams concluded that without speculators
to support the market, the fall in grain prices that took place in the
early 1890s would have been even worse.11
A second line of criticism was that price declines were caused not by
manipulation but by short selling, the practice of selling, for future de-
livery, a commodity one did not yet own. A short seller hoped that the
price would drop between the contract date and the delivery date, so
that he could purchase it on the delivery date at a price lower than the
seller had agreed to receive. “The man who sells hundreds of thousands
of bushels of wheat, although he does not possess a bushel, affects the
market,” argued Benjamin Butterworth, the Ohio representative who
sponsored the 1890 bill that would have put an end to short selling.
“For every bushel of grain grown a hundred are sold, and I maintain
that the effect is to lower the market price. … When the market supply
is 100 times greater than it really is you drive the price down.” C. Wood
Davis, the Kansas farmer turned activist who devoted years to the anti-​
option cause, offered an analogy. Davis asked the House Agriculture
Committee, “If the offering of too many coats will reduce the price of
coats, will not the offering of wheat or cotton or any other article in like
excess of requirements, reduce the price of wheat or cotton, or other
articles?” Such sellers of fictitious crops were competing with farmers
selling real crops, Davis explained, and so long as “the broker’s lung
power is good, they can continue to offer 10,000 bushels every minute
in competition with the 10,000 bushels of wheat which we produce.”

11 Daniel J. Sully, “Is the High Price of Cotton the Result of Manipulation?,”
North American Review 178 (1904): 194; Richard T. Ely, “Land Speculation,”
Journal of Farm Economics 2 (1920): 130; Harold J. Howland, “Gambling Joint
or Market Place?,” Outlook 104 (1913): 437; Charles A. Conant, “The Uses of
Speculation,” Forum 31 (1901): 701; Egerton R. Williams, “Thirty Years in the
Grain Trade,” North American Review 161 (1895): 33.
114

114 Speculation

John Whittaker, a commodity merchant in St. Louis, testified that


in New York in January 1892 alone, three million bushels of physical
wheat were sold in the spot market, while eighty-​three million bushels
of theoretical wheat were sold in the futures market. “I do not think
it will occur to any reasonable man that the selling of that 83,000,000
bushels did not depress the value of the actual wheat that was sent to
the market for sale,” Whittaker concluded. “Unqualifiedly I call that
the greatest outrage committed on the American people.”12
The idea that short selling depressed prices was widely believed but
simply mistaken, as representatives of the exchanges pointed out year
after year. Every short seller had to be a buyer eventually, so sellers and
buyers would cancel each other out. “The moment a man sells prop-
erty he has not got he is a buyer from necessity,” explained Charles
Hamill, the president of the Chicago Board of Trade. “They are the
most anxious of buyers.” And of course all those bushels of wheat sold
on the futures market were simultaneously being purchased on the fu-
tures market by the person on the other side of the transaction, so the
downward pressure from the sales had to be exactly counterbalanced by
the upward pressure from the purchases. “The bear is seen offering to
sell a commodity,” one defender of speculation noted, “but the specula-
tive buyer is not seen, as he bids for even greater quantities of ‘wind.’ ”
The New York Times offered a simple demonstration that the supposed
link between short selling and price declines was “of the post hoc ergo
propter hoc kind.” Prices had dropped for commodities like wheat and
cotton that were sold short on the exchanges, but they had dropped
just as much for commodities that were not traded on exchanges, like
wool, pig iron, and the animal hides used for leather. The Times con-
cluded: “Trading in ‘futures’ has no more been the cause of depression
in cotton, wheat, and hog products than its absence has been the cause
of it in wool, hides, and pig iron.”13

12 Chicago Daily Tribune, 17 Feb. 1890, 5; Fictitious Dealing in Agricultural


Products: Testimony Before the Committee on Agriculture (Washington, D.C.:
Government Printing Office, 1892), 12, 15, 47.
13 Fictitious Dealing, 169; Harrison H. Brace, The Value of Organized Speculation
(Boston: Houghton Mifflin Co., 1913), 100; New York Times, 28 May 1892, 4.
  115

The Anti-Option Era 115

Agrarian critics had a third reason for believing that speculation


reduced the farmer’s earnings, one that was not as easy to dismiss.
Speculators were intermediaries between the producers and consumers
of agricultural products. The ultimate price to the consumer was capped
by what consumers were willing to pay, which meant that every dollar
earned by a speculator was taken directly from the farmer’s pocket. The
idea that speculators were parasitic middlemen was a staple of the ag-
ricultural press. “We believe it necessary to have some middlemen to
keep the wheels of trade moving properly,” explained the Ohio Farmer,
“but do not believe it necessary for farm products to go through three,
four, five or six hands, each reaping a profit, before reaching the manu-
facturer or consumer, and causing the farmer to sell for less than cost
of production.” The Indiana Farmer’s Guide agreed that “between the
man who produces food today and the man who consumes the food,
operates a long and complicated system of buying and selling which
has nothing to do with real service to either of these primarily inter-
ested parties.” The Farmer’s Guide cited a local example: “a carload of
Indiana eggs sold to a New York wholesaler at 24 ½ cents a dozen, was
resold nine times without leaving the cold storage warehouse, until the
St. Regis hotel paid 43 cents a dozen for the lot.” While farmers were
suffering, “a lot of fellows make millions of dollars every year by so-​
called business transactions in food-​stuffs which contribute nothing at
all toward bringing the food from the producer to the consumer.” As
the Southern Cultivator complained, “the farmer gets a small price, the
manufacturer pays a good one, and the intervening speculator pockets
the difference.”14
From the farmers’ perspective, all this speculation in agricultural
commodities stood in conspicuous contrast with the lack of specula-
tion in other goods, which consumers bought directly from producers.
“No one who is about building a house or block goes into the option
pit to make his contract,” the Wichita Daily Eagle reasoned, while the
Butterworth bill was before Congress in 1890. “He looks for the actual
owners and laborers with whom to make his contract. Merchants who

14 “Wool,” Ohio Farmer 59 (1881): 421; “Oust the Speculators,” Indiana Farmer’s


Guide 31 (1919): 36; “The Panic,” Southern Cultivator 31 (1873): 409.
116

116 Speculation

wish to contract for the future supply of an article go to the manufac-


turer and engage with him to supply the article when needed in the
future. He does not contract for it with people who do not have it.
Why should farm products be on a different basis?”15 If all these other
industries could flourish without speculators to take a cut of the pro-
ducer’s profit, why couldn’t agriculture?
Implicit in this question was the ancient view, still very much alive,
that speculation was a nonproductive activity, one that added no value
to the commodities being bought and sold. “If A goes to Chicago and
buys wheat and transports it to New York and sells it, he is a producer,”
reasoned one critic. “But the speculator never enhances the value of
anything.” Merchants and speculators both took risks, another ex-
plained, but the difference between the two was that the merchant’s
profit “is sought as a fair equivalent for services rendered to the com-
munity in bringing some needed article to market,” while the specula-
tor’s profit was simply taken from someone else’s pocket.16 The farmers’
anger was not directed at all middlemen indiscriminately, but only at
those, like the traders on the commodity exchanges, who seemed to be
amply rewarded for no good reason.
Defenders of speculation had an answer: commodity speculation did
add value to commodities, and farmers were accordingly gainers, not
losers, from the chain of intermediaries who stood between them and
the consumer. The difference between agriculture and other industries,
they argued, was that agricultural products were harvested during only
one season of the year but were consumed all year round. Someone
had to hold commodities for the remainder of the year, which meant
that someone had to bear the risk that commodity prices would decline
before they could be sold to the consumer. That was the function served
by speculators. The speculator “performed a distinct service to the ag-
riculturalist,” explained John Hill of the board of trade, “by providing
a market for one million bushels of wheat at a time when the con-
sumer was in no hurry to buy, and when the wheat was accumulating

1 5 Wichita Daily Eagle, 4 Sept. 1890, 4.


16 “The Bank as a Promoter of Speculation,” Bankers’ Magazine and Statistical
Register 6 (1872): 657–​58; “The Tragic Side of Commercial Speculation,” Zion’s
Herald 52 (1875): 292.
  117

The Anti-Option Era 117

in the warehouses.” Speculators were often said to “carry” commodi-


ties throughout the year, a term that had two meanings. Sometimes
speculators literally possessed commodities—​they purchased commod-
ities when farmers needed to sell but consumers were not yet ready to
buy. Without them, one journalist pointed out, “the farmer could not
sell except throughout the year as the demand arose.” But speculators
“carried” commodities in a metaphorical sense as well. They relieved
the farmer of the risk that prices would fall between the harvest and
the sale. Without speculators, suggested the Chicago Tribune, farm-
ers would have to “borrow money with which to pay their way and
take their chances of a falling market.” The broker Alexander Hudnut
pointed out that someone had to bear the risk that commodity prices
would decline. If it wasn’t a speculator it would be the farmer, and in
that case the farmer would be speculating just as much as any trader on
an exchange. If there were no speculators in the chain of distribution,
farmers would have to become speculators themselves.17
Farmers were not the only ones who faced the risk that prices
would fall while commodities were in their possession. The same risk
confronted all the useful intermediaries between the farmer and the
consumer—​the miller, the shipper, and so on—​who bought commodi-
ties at one time and sold them at another. These market participants
used the commodity exchanges to hedge that risk; they sold futures to
offset the risk of price declines. A Minneapolis trade journal explained
that “there is no grain or milling concern in the northwest with capital
enough to stand ‘pat’ on its share of purchases without some sort of
assurance of the character supplied by future sales.” W.  C. Brown, a
miller in Fostoria, Ohio, testified that each fall he took possession of
approximately seven hundred thousand bushels of wheat. “As soon as
we buy it we sell a future against it,” he reported. “Then it makes no
difference to us whether the price of wheat goes up or down, the mill
is absolutely protected and we do not gamble in grain.” Of course, it

17 John Hill Jr., Gold Bricks of Speculation (Chicago: Lincoln Book Concern, 1904),
420–​21; “Business and Speculation,” Independent 48 (1896): 21; Chicago Daily
Tribune, 25 Dec. 1891, 4; Alexander M. Hudnut, “Dealing in Futures Is Not
Gambling,” Town and Country, 20 May 1905, 37; L. D. H. Weld, “High Food
Prices, Middlemen, and Speculation,” North American Review 206 (1917): 589.
118

118 Speculation

would not be possible to sell futures unless someone stood ready to


buy futures. One could not sell a risk without a counterparty willing to
assume it. Speculators thus performed the same function for millers and
other intermediaries that they performed for farmers. “This is where the
speculator comes into play by being willing to take a chance that we
will not take,” explained E. A. Duff, who operated thirty grain elevators
in Nebraska and Kansas. “It is mathematically true that hedging can
not exist without some speculation,” agreed George McDermott of the
Kansas Grain Dealers’ Association.18
Members of exchanges accordingly liked to analogize their role to
that of an insurance underwriter:  they were compensated for assum-
ing risks others wished to shed. The grain trade was a “vast system of
insurance,” declared Walter Fitch, the former president of the Chicago
Board of Trade. In its promotional material, the board described itself
as “an insurance underwriter for all of these elements of the trade.”
F. B. Wells, the vice president of a grain warehousing firm and thus a
frequent hedger, thought commodity exchanges the American equiv-
alent of Lloyd’s of London, the venerable English insurance market.
“The only difference between those men and speculators in the grain
exchange,” Wells reasoned, “is that they received a fixed amount for
assuming the risk, whereas the speculator in the grain pit backs his
own opinion that he will receive something for assuming the risk. The
principle involved is the same.”19
Critics found the analogy absurd. Insurance companies “have no fi-
nancial interest in the risks except in decreasing them,” observed the
Illinois Agricultural Association, an organization of farmers. “Fire in-
surance companies spend large sums of money in educating the public
to prevent fires. They are certainly not benefited by any increase in
the number of fires. Grain speculators, on the other hand—​the men

18 Daily Market Record, 8 Jan. 1892, 1; Fictitious Dealing, 213; Future Trading in
Grain: Hearings Before the Committee on Agriculture and Forestry, United States
Senate, 67th Cong., 1st Sess. (1921), 109–​10, 207.
19 Hearings Before the Committee on Agriculture During the Second Session of the
Sixty-​First Congress, vol. II, Hearings on Bills for the Prevention of “Dealing in
Futures” on Boards of Trade, Etc. (1910), 438; Board of Directors Resolution, 28
Feb. 1921, CBT, box 334, folder 8; Future Trading in Grain, 393–​94.
  119

The Anti-Option Era 119

who provide this so-​called grain insurance in the form of hedging—​are


interested in increasing the risk.” There was no hedging, and no fu-
tures trading, in most markets. “There is none on hay, there is none on
iron, there is none on coal, and there is none on an enormous amount
of important products,” noted T. J. Brooks, speaking on behalf of the
Farmers’ Educational and Cooperative Union. “There are no futures on
farm machinery,” he pointed out, and “a manufacturer may produce a
machine five years before it is sold.” Hedging, in his view, was no more
necessary for cotton or for wheat than it was for anything else. After
all, the price of any item could fall while someone was holding it; how
could hedging be needed only for certain commodities? “The obvious
answer” to the claim that speculators assumed risk, argued the Populist
theologian John Ryan, was that “traders in produce should take the
risks of fluctuating prices themselves.” If traders bore their own risks,
“many of them would doubtless go to the wall, but the community
would be the gainer through the elimination of the unfit.” Why should
farmers pay for an elaborate infrastructure of speculation just to prop
up the weakest grain traders?20
George Norris, the long-​serving Nebraska senator, grew up on a
farm in Ohio. At the hearings on the bill that would eventually become
the Grain Futures Act of 1922, Norris summed up the farmers’ skepti-
cism about the insurance function supposedly served by speculators.
“If I buy a piece of land because I think it is cheap, why should there
not be some organization on which I could rely in that transaction?”
Norris asked Rollin Smith, the Agriculture Department’s supervisor of
grain markets. There was no hedging in the real estate market, despite
its evident risks. “Why should not the man who buys wheat be subject
to the same rule of commerce that the farmer is, or the local dealer who
buys hogs in a town lot?” Such people, like most Americans, bore their
own risks. “Why should there be any difference; why does not the same
rule of trade apply to all of them?” Smith recognized the implication of
Norris’s question. “That touches on a sort of nerve center of the whole
system,” he responded. “Why should a vast volume of speculation by

20 Future Trading in Grain, 176; Hearings Before the Committee on Agriculture, 7;


John A. Ryan, “The Ethics of Speculation,” International Journal of Ethics 12
(1902): 338.
120

120 Speculation

the public, the professionals, the cash grain men, elevator companies
and others, continue to go on just to create an insurance market for a
few grain dealers? That is what it amounts to.” Norris replied, “If you
ask me, I say I do not know why.” “Neither do I,” admitted Smith, “but
that is the situation.”21
Norris and Smith were likely being disingenuous, because the ex-
changes and their supporters had been providing an answer to their
question for three decades:  if the intermediaries in the distribution
system had to bear their own risks, they would have to be compensated
for doing so, in the form of a bigger spread between the price they paid
to farmers and the price they charged to consumers. Without specula-
tors, the farmers’ earnings would be even lower than they already were.
Traders “would certainly be compelled to buy on a wider margin to
protect themselves against declining markets while the grain is in tran-
sit,” the grain merchant S. W. Tallmadge advised William Vilas, “and
in so doing the producer receives less pay for his grain.” “If the farm-
ers succeed” in their efforts to ban the exchanges, the Chicago Tribune
editorialized, “they will find themselves even worse off than now, the
remedy being even more bad than the disease they seek to have cured.”
The broker S. V. White wondered: “Is there a man so blind as not to see
that if the buyer had to take the risk of a two weeks’ fluctuation in the
grain market he could not pay within many cents per bushel of the ad-
vanced prices?” R. C. Clark, a member of the Chicago Board of Trade,
agreed that “there isn’t a farmer in the West who could get as much for
his grain by two to five cents a bushel as he gets now if trading in futures
were prohibited.” “Before trading in grain for future delivery became
a custom,” recalled John Messmore of the St. Louis Grain Exchange,
grain exporters “made from 10 to 20 cents per bushel on every bushel
they exported.” The exchanges had driven this margin sharply down.
“The exporter of today works on a very small margin, frequently only 1
cent per bushel profit,” Messmore explained. “The millers also work on
a small margin. They are able to do this because they can protect their
purchases or sales in the terminal or primary markets, by buying or sell-
ing the grain for future delivery.” Speculation thus left more money for

21 Future Trading in Grain, 52–​53.


  121

The Anti-Option Era 121

farmers. Messmore concluded that “to pass a law abolishing the trading
in grain for future delivery would be a calamity to the farmers.”22
Even worse, exchange members suggested, an inability to hedge
might make the grain trade too risky for all but the largest dealers,
which would leave the farmer with a constricted market for his prod-
ucts. “The result of the bill,” predicted Murry Nelson of the Chicago
Board of Trade, “would be to crush out the small dealers and place the
farmer at the mercy of the large capitalist, who can buy cash grain; but
they will buy it at such a price that they can hold it until Gabriel blows
his horn without sustaining loss.”23 Without competition among many
potential purchasers for grain, farmers would be forced to sell at lower
prices.
Of course, it would be possible for farmers to bypass the interme-
diaries altogether and sell grain directly to the consumer as in the old
days, but this was yet another route to lower prices. “If the producer
should sell direct to the consumer,” board of trade member Charles
Counselman expected, “the farmers of Kansas and Nebraska, instead
of receiving 10 to 15 cents a bushel for corn this winter, which was little
enough, would not have received five cents a bushel.” Counselman’s col-
league J. H. Norton pointed out that “the speculator doesn’t buy merely
what he himself can consume, as does the farmer’s neighbor, but he
buys what a nation or part of a nation may need. In this way he gives
the farmer the benefit of the world’s prices instead of being limited to
local demand.” Benjamin Hutchinson, one of the leading grain specula-
tors of the 1880s, posed the problem in more concrete terms. “If there
were no speculation,” he suggested, “the farmers could only sell their
grain to local buyers, who would be liable to get full and stop buying,
and then the farmer would be compelled to wait for customers; and in
the meantime a mortgage might be foreclosed on his farm, even while
the wheat in his bins would more than satisfy the mortgage, if converted
into cash. But speculation, flashing its news over the wires from one
side of the world to the other, keeps the market always open to him.”

22 S. W. Tallmadge to William Vilas, 12 Mar. 1892, WFV, box 14, folder 1; Chicago
Daily Tribune, 6 Mar. 1892, 12; 26 Aug. 1890, 5; Jan. 1891, 7; Hearings Before the
Committee on Agriculture, 470.
23 Chicago Daily Tribune, 12 Apr. 1890, 5.
122

122 Speculation

Hutchinson concluded that “if farmers are hostile to boards of trade and
futures, in my opinion they are mistaken.”24
On this view, the farmers were miscalculating their own interests
and agitating for measures that would leave them worse off. “The de-
struction of the system, which is a perfectly legitimate development of
trade, would produce immediate disaster” for farmers, the New York
Times editorialized. “Their demand for its abolition comes from a
lack of understanding of the function performed by the system and
a misconception of its effects.”25 But how could millions of farmers
have made such a big mistake about a matter so central to their own
well-╉being?26

Where Are All These Horses?


William Vilas visited the Chicago Board of Trade in January 1893 for
a meeting with the board’s leaders. The Senate had already passed the
Washburn anti-╉option bill, despite Vilas’s opposition, and it was pend-
ing in the House. Board president Charles Hamill and his colleagues
welcomed the opportunity to discuss legislative strategy with an ally,
but Vilas had a more precise purpose in mind. “I was a little anxious,”
Vilas explained to Hamill, “to see if there was not something in the way
of figures that we could obtain that would satisfy the farming commu-
nity generally that the idea was an erroneous one that speculation in
wheat necessarily depressed prices, or indeed at all.” Vilas was getting

24 Chicago Daily Tribune, 10 Apr. 1890, 6; 21 Jan. 1890, 3; B.  P. Hutchinson,


“Speculation in Wheat,” North American Review 153 (1891): 414, 416.
25 New York Times, 28 Apr. 1892, 4.
26 For answers to this question I find less plausible than the one offered here, due
to the absence of evidence that anyone was thinking along these lines during
the debates of the late nineteenth and early twentieth centuries, see B. Peter
Pashigian, “Why Have Some Farmers Opposed Futures Markets?,” Journal of
Political Economy 96 (1988): 371–╉82 (arguing that farmers would have gained
from abolishing futures if the owners of grain elevators were using futures prices
to facilitate the formation of buying cartels); Roberta Romano, “The Political
Dynamics of Derivative Securities Regulation,” Yale Journal on Regulation 14
(1997): 307–╉11 (arguing that farmers would have gained from abolishing futures
if farmers had been able to form selling cartels of their own).
  123

The Anti-Option Era 123

ready to defend his vote back home in Wisconsin, where the farm-
ers believed that the board of trade was responsible for their financial
distress. “We want to say answer a fool according to his folly and pass
him by. But you can’t do it,” Vilas lamented. “You have got to begin pa-
tiently at the bottom and build up this argument.” As Vilas saw it, the
farmers “have been in a certain sense hoodwinked into the support of
something that was not to their interest.” Vilas hoped to set the farmers
straight, but he knew it would not be easy. “I should have no apprehen-
sion of meeting any farmers in this world,” he explained, “but there is
a class of men who make a very loud outcry—​what I call the political
farmers; they are presidents of alliances, and all sorts of agricultural
organizations.” These were “men who are continually making political
headway for themselves,” in Vilas’s view, rather than looking out for the
real interests of the farmer.27
On the exchanges there was little doubt that “the farmers have been
misled on this subject,” as a member of the Toledo Produce Exchange
put it. “I do not wonder that the uninstructed person” could be con-
vinced that speculation reduces prices, agreed Charles Hamill of the
board of trade.28 But who would take the trouble to deceive so many
farmers about commodity trading? Who had an interest in making
them believe it was harmful? On the exchanges and in the urban press,
the answer seemed clear: the real backers of this legislation, the men
pulling the strings behind the scenes, were the millers.
Millers bought grain from farmers, just like speculators did. If they
could push the speculators out of the market, they would be the only
buyers left. “The only competition the millers now have is in the ex-
changes in the buying of grain for future delivery,” declared H. A. Foss
of the board of trade. “Therefore, the abolition of time dealings would
eliminate this competition and in the end surely would reduce the
number of buyers of cash cereals from the farmer.” The Chicago Tribune
recognized that the bill “will benefit the millers more than any other
class,” because it would allow them “to beat down the farmer when he
wants to sell his wheat.” In Congress, the New York Times reported, the

2 7 “Minutes of Conference,” 2 Jan. 1893, WFV, box 15, folder 2, 1, 8–​9, 20, 49.
28 Denison B. Smith to Vilas, 15 Dec. 1892, WFV, box 15, folder 1; “Minutes of
Conference,” 1.
124

124 Speculation

bill was known as “the millers’ friend.” The Times alleged that the peti-
tions of support ostensibly streaming in from farmers’ organizations
throughout the country were in fact written in Washington and sent
out for signatures, as part of a public relations campaign funded by the
millers. “There isn’t any question about it,” William Vilas told the lead-
ers of the board of trade. “They have worked up a fictitious, apparently
agricultural sentiment to support their interests.”29
Several of Vilas’s correspondents likewise singled out the millers.
“It is promoted by the millers,” insisted the grain merchant Robert
Eliot, “who seem to wish to get a monopoly of the trade, and rule
out the speculator, who is a competitor and who aids in keeping up
prices.” E. C. Wall, writing on behalf of the Milwaukee Chamber of
Commerce, called the legislation “an absurd bill” that “will do the
farmers infinite harm,” but one that “is a good thing for wealthy mill-
ers; for they can then buy the farmers’ grain at their own price.” Even
some of Wisconsin’s farmers recognized who stood to gain. “I am a
farmer,” declared E. H. Harris of Beloit, “and while I am opposed to
the Chicago methods of playing wheat ‘Poker,’ still I  prefer it rather
than take the chances of being dumped bodily into the Mill Elevators
at Minneapolis.”30
Indeed, William Washburn, the Senate’s leading proponent of leg-
islation, was in the milling business himself, as the urban press missed
no opportunity to point out. His family owned a large milling com-
pany in Minneapolis, the predecessor to the firm that would later be
called General Mills. The Tribune accused him of “pushing this bill for
mercenary and not philanthropic motives.” Washburn provided only
the lamest of defenses. “I know there has been a good deal of this talk
running in the newspapers, assuming that this bill is in the interest of
the millers,” he acknowledged on the Senate floor. “I should like to un-
derstand under what provision of this bill the miller gets any advantage

29 Untitled typescript (1909), CBT, box 328, folder 2; Chicago Daily Tribune, 26
Aug. 1890, 5; New York Times, 31 Jan. 1893, 3; 25 June 1892, 5; 28 Feb. 1893, 4;
“Minutes of Conference,” 20.
30 Robert Eliot to Vilas, 15 Jan. 1892, box 13, folder 1; E. C. Wall to Vilas, 3 Mar.
1892, box 13, folder 2; E. H. Harris to Vilas, 3 Jan. 1893, box 15, folder 2, all
in WFV.
  125

The Anti-Option Era 125

over anybody else in the world; how he occupies any different position
from anybody else who wants to buy wheat.”31 Of course, other than
speculators, no one but a miller had any occasion to buy wheat straight
from the farmer.
For opponents of the legislation, the millers’ incentive to push the
speculators aside provided the answer to the puzzle of how so many
farmers could be so wrong. It was all a nefarious plot: the millers had
tricked the farmers into believing that speculators were taking their
money, when in fact the speculators were the very people who were
protecting the farmers from the millers themselves, who would take
even more of the farmers’ money if speculation were prohibited.
In assessing the truth of this theory, it will be useful to separate the
incentives of the millers from those of the farmers. The largest millers
almost certainly did stand to gain from the abolition of commodity
speculation. Smaller millers needed to hedge on the exchanges, but the
bigger millers may well have been able to bear the risk of price declines
themselves, if they could pay a low enough price to farmers. Prohibiting
commodity speculation might have cleared the field, not just of the
speculators, but of the smaller millers as well. At harvest time, the big
millers would have had the farmers over a barrel. “Gentlemen, if you
put the exchanges out of business, who would profit?” the cotton dealer
Solomon Cone asked the House Agriculture Committee. “Do you see
the representatives of the big mill interests of the South?” Cone was
there to testify against the legislation, but, he explained, “I guarantee
that my own brother, who buys for his different mills over 100,000
bales of cotton, if he could control me, would not have me here.”32 The
millers would likely have been the main beneficiaries of prohibiting
commodity speculation.
But the millers hardly invented the notion that speculation hurt
farmers. As we saw in ­chapter  1, there was a long tradition of belief
that speculators could manipulate prices to their advantage, and that
speculators were nonproductive middlemen who took money out of
the pockets of producers. These propositions would have been tenets

31 Boston Evening Transcript, 13 July 1892, 5; New  York Times, 16 July 1892, 9;
Chicago Daily Tribune, 23 July 1892, 4; Congressional Record 23 (1892): 6442.
32 Hearings Before the Committee on Agriculture, 95.
126

126 Speculation

of late nineteenth-​century agrarian populism even without the millers.


The millers were opportunistically exploiting beliefs that were already
widespread; they did not need to convince anyone of anything.
Farmers, meanwhile, had a few reasons to object to commodity spec-
ulation, regardless of whether it had any effect on prices. Commodity
prices had once been determined in local markets, in which farmers
participated themselves, but now prices were determined in a world-
wide market by a host of other actors. Farmers were left feeling power-
less, their livelihoods controlled by others. “It is the offering of these
immense quantities of fictitious products that fixes the price,” C. Wood
Davis complained. “The farmer has nothing to do with fixing the price,
he is the only man on earth that is not allowed to have anything to
do with the fixing of the value of the property that he has produced.”
Farmers were forced to accept prices dictated by exchanges in big cities,
where traders who had never worked the soil bought and sold crops,
most of which did not even exist. The farmers “have found that before
their crop is harvested, possibly before it is planted, that some one
without authority from them had contracted to sell it; and they tell us
that this is right and in the interest of the farmer,” declared an indig-
nant J. H. Brigham, the master of the National Grange. “We do not
believe that any man has any right to sell our products before we have
authorized their sale.”33
Representative Albert Burleson of Texas, who counted many cotton
farmers among his constituents, provided a useful parable to illustrate
how the exchanges looked from the point of view of a farmer. “Suppose
a man who had a horse to sell heard that there was a horse exchange
in town and should carry his horse to the exchange to dispose of him,”
Burleson began.

Arriving there he finds that on this exchange they are not selling
horses but future contracts for the delivery of horses, and that these
transactions were going to control the price he was to receive for
his horse. Suppose he should see a future contract for the delivery
of a horse sold for a given price, then another future contract for a

33 Fictitious Dealing, 14, 260.


  127

The Anti-Option Era 127

horse at a lower price, and then another at a higher price, and some-
body standing by, a member of the exchange, should say to him as
these contracts were being sold and bought: “Now, the price of your
horse has gone down, now it has gone up.” And if you asked why,
should answer, “Because these sale contracts for the delivery of horses
that are taking place, actual contracts, where there is an expectation
that delivery will take place, indicate the market value of your horse,
in fact control the price you are to receive for him.” When he saw
these transactions going on and on and on would it be surprising if
he finally asked, “Where are all these horses? I don’t see any horses
that are being traded in under these contracts; where are the horses?
I  have seen dozens of sales, I  have seen hundreds of contracts for
horses sold, but I have not seen a horse.” Would you blame him if
he reached the conclusion that they were phantom horses that were
being dealt in? Why, of course, you would not. Neither would you
blame him if he objected to having the market price of his horse
influenced or controlled by these transactions.

A Minneapolis trade magazine expressed the farmers’ unhappiness more


sharply: Farmers “are heartily sick of the reign of the gamblers. Under
the tyrannical rule of the few who neither toil nor spin, the grower
and the manufacturer find themselves slaves. The natural servant has
become an unnatural master. The beggar is on horseback and is riding
roughshod over … the farmer.” In such circumstances, banning specu-
lation was a way to restore power to the farmers, regardless of whether
it increased prices. “A revolt is in order,” the magazine concluded, and
a law terminating the commodity exchanges “is, perhaps, the first sign
of the coming warfare.”34
Believing their lives unfairly controlled by speculators, farmers had
little trouble summoning up all the traditional critiques of speculation.
Speculation still had a moral dimension. It was still widely seen as un-
christian. The Christian Observer and the Christian Union both edi-
torialized in favor of antispeculation bills in the 1890s, not because of

34 Hearings Before the Committee on Agriculture, 421; Weekly Northwestern Miller


29 (1890): 477.
128

128 Speculation

any economic effects the bills might have, but because, as the Observer
put it, “the principle that controls in this matter is set forth by the
Lord in the decalogue”—​the commandment that prohibits stealing.
The Union saw the bills as evidence of “the awakening moral sentiment
of the Nation.” The journalist George Muller pronounced commodity
speculation “unworthy of nineteenth century progress. It is selfish and
dehumanizing.” He thought it should be abolished on moral grounds
even “if in the process of reformation, the farmer temporarily gets less
for his product.” The minister George Hubbard agreed that “when
weighed in the balances of eternal justice, speculation is found want-
ing. Its character will not stand the supreme test. It is a moral wrong.”35
Speculation was still widely understood as a form of gambling, a
practice moralists considered wrong regardless of whether it was profit-
able to any particular participant. Speculation on the board of trade
“can hardly be justified in the light of the moral law,” one critic charged.
“With regard to the character of gambling there is no controversy. Every
one admits its immorality.” The Massachusetts Ploughman contrasted
the gambler, who knows he is doing wrong, with the grain speculator,
who erroneously “imagines he is doing a very respectable thing.” In
the late nineteenth century, the Louisiana State Lottery was a popu-
lar metaphor for all the evils associated with gambling, because it was
the only legal state lottery in the country. “But, after all, what a small
affair has the Louisiana lottery been,” declared one magazine in 1892,
“when compared with the ‘bucket shop’ and Board of Trade.” When
the lottery was abolished, a rural Minnesota newspaper declared that
it was “time that the option room of the Chicago Board of Trade was
catalogued with the Louisiana State Lottery and relegated to the shades
of an ancient and barbarous method of appropriating the unearned
increment from the pockets of the hardworking citizens of the great
northwest. By the side of comparatively honest methods in gambling,
trading in options will in the history of the near future take its place
as the co-​partner and twin fraud of bunco steering.” In the Senate,

35 “Is Speculation Right?,” Christian Observer 84 (1896): 1; “The Anti-​Option


Discussion,” Christian Union 47 (1893): 60; The Sunset Club (Chicago: Sunset
Club, 1892), 143; George H. Hubbard, “The Ethics of Speculation,” New
Englander and Yale Review 14 (1889): 44.
  129

The Anti-Option Era 129

William Washburn shrewdly capitalized on the widespread belief that


speculation was tantamount to gambling. His bill, he insisted, “is aimed
at a system of gambling the most unique, insidious, the most perni-
cious, and bringing with it the most widespread and disastrous results
of any scheme of gambling that the wit and skill of man has ever yet
been able to devise.” He concluded his speech with a biblically inflected
call to “drive the gamblers from our temples and reinstate therein the
genius of legitimate trade.”36
Opponents of regulation responded by distinguishing between spec-
ulation and gambling. Speculation, they argued, was the rational ap-
plication of intellectual effort to the estimation of future prices, while
gambling was pure chance. “Speculation is a venture with calculation,”
one explained, “gambling a venture without calculation.” But those who
saw the board of trade as a gambling den were hardly convinced. “What
shall we say about the Chicago grain speculators?” asked one proponent
of regulation. “What do they know about next year’s droughts, or next
month’s prairie fire, or next week’s Charleston earthquake[?]‌” These
supposed rational calculators “have no special ‘revelations’ in regard to
what may happen in the ‘future.’ ” Their enterprise was based on chance
just as much as the gambler’s.37
Speculators were still widely depicted as greedy, dishonest people who
cared more about money than about their fellow man. “Speculation is
the child of covetousness,” declared the Christian Observer. “Its tenden-
cies are grasping and selfish.” The Christian Advocate warned its readers
to “avoid speculation of every kind,” because “the life of the speculator
is one of consuming anxiety. He never knows where he is or what he
is worth.” Speculation “corrupts the speculator,” cautioned the Western
Christian Advocate. “It cannot be carried on in honor and honesty,

36 “Speculation in Business,” Chautauquan 4 (1884):  282; “Agricultural


Comment,” Massachusetts Ploughman and New England Journal of Agriculture
43 (1884): 1; “The Progress of the World,” Review of Reviews 6 (1892): 135; New
Ulm Review, 5 Oct. 1892, 3; Congressional Record 23 (1892): 5980, 5993.
37 Thomas L. Greene, “Legitimate and Illegitimate Speculation,” Independent
44 (1892): 7; “What Is Speculation?,” Outlook 74 (1903): 1006; D. G. Watts,
“Speculation as a Fine Art,” Cosmopolitan 10 (1891): 592; “Speculating and
Speculators,” Independent 38 (1886): 24.
130

130 Speculation

because it is essentially dishonest in its principles.” Phrenologists even


claimed that speculators were prone to baldness just above the right ear,
because that was the location of “the organ of Acquisitiveness, or love of
gain,” which speculators would rub when they became excited.38
Farmers thus had several reasons to be suspicious of commodity spec-
ulators, even apart from whether speculation lowered prices. The fall in
commodity prices was a catalyst that sharply increased the salience of
these older noneconomic critiques of speculation, but these critiques
would have had some force under any circumstances. As for whether
speculation caused the drop in prices, the answer depends on how
broadly the question was framed. The speculators were almost certainly
correct that shutting the exchanges would not bring higher prices to
the farmer, and indeed would probably bring lower ones. Commodities
were traded, and prices were set, in an international market. Farmers
had to sell their crops to someone; if it wasn’t speculators, it would be
millers, exporters, or other intermediaries, none of whom would pay
higher prices than the speculators had. Expelling a class of buyers from
the market was not a likely route to higher prices.
But we can get a sense of the farmers’ perspective if we frame the
question more broadly. The exchanges were just one component, and to
the farmer the most visible component, of the international network of
commodity distribution constructed in the nineteenth century. In 1800,
most crops were sold locally. Prices probably varied considerably from
region to region and even from town to town, reflecting local variations
in supply and demand. By 1900, crops were traded in a worldwide
market, and prices were much more uniform. When farmers lashed out
at the board of trade and other exchanges, these were just the closest
and most tangible symbols of a fundamental change in the organiza-
tion of agricultural life.39 If the exchanges had been abolished but the
rest of the system left in place, the farmers would not have benefitted.

38 “Honest Speculation,” New York Evangelist 53 (1882): 4; “What Is Speculation?,”


528; “The Temptations of Speculation,” Christian Observer 87 (1899): 2; “Letters
to Young Men,” Christian Advocate 59 (1884): 254; “Legitimate Business Better
Than Speculation,” Western Christian Advocate 25 (1874): 196; “How Stock
Speculators Become Bald-​Headed,” Phrenological Journal 57 (1873): 313.
39 David S. Jacks, “Populists Versus Theorists: Futures Markets and the Volatility
of Prices,” Explorations in Economic History 44 (2007): 358.
╇ 131

The Anti-Option Era 131

But the agrarian critique of speculation was often something more than
that. It was a critique of the entire system, a desire to return to ear-
lier ways of moving food from the farm to the consumer. If the entire
system could have been dismantled, commodity markets would have
been local once again, and prices would have once again varied widely
from one place to the next. The economic effect on farmers would have
varied widely too. Some would have been better off some of the time,
and some would have been worse off some of the time.
Of course, there was no possibility of dismantling the entire distri-
bution network. The only parts of it available to attack were the ex-
changes. The farmers had to focus on the exchanges, but by linking
speculation on the exchanges to the decline in commodity prices, the
farmers were stuck with an implausible economic argument. And while
lower commodity prices were bad for farmers, they were a great ben-
efit to consumers, so in calling for legislation on the ostensible ground
that it would raise food prices, the farmers were making it much less
likely that nonfarmers would join their cause. That Congress neverthe-
less came so close to shutting down the exchanges is a testament to the
political force of the agrarian movement in the 1890s.

Justifiable Gambling
The argument that commodity speculation hurt farmers was even less
plausible in 1920 than it had been in 1890. In the intervening years,
the issue’s political visibility had given rise to some thorough defenses
of speculation’s role in the economy. The American political econo-
mists of the first half of the nineteenth century had tended to classify
speculation as a nonproductive activity, one that added no value to the
goods being traded. In the later part of the century, however, American
economists began to take a different view. Speculation, they asserted,
was just as useful as manufacturing.
The leading voice in this school of thought was Henry Crosby
Emery, who spent most of his career in the economics department at
Yale. Emery’s doctoral dissertation, Speculation on the Stock and Produce
Exchanges of the United States, was published as a book in 1896, not
long after the peak of antispeculation agitation. “The criticism directed
against speculation is made from two somewhat conflicting points of
132

132 Speculation

view,” he argued. “The first is that speculation is merely gambling and


has no reference to actual trade, except that it consists in betting on the
course of prices. The second is that speculation is all-​powerful in trade,
which has become completely demoralized by its subjection to ficti-
tious speculative conditions.” He contended that both lines of criticism
were wrong. Speculators, he asserted, played a crucial role in absorbing
the risks that would otherwise deter traders from efficiently moving
products from the farmer to the consumer. “Speculative losses cannot
be met by mutual insurance,” he pointed out, “since they fall on all
members of the class at once. These are risks that inhere in ownership,
and they can be met only by a transfer of ownership.” By transferring
risk to those better able to bear it, the commodity exchanges allowed
the distribution system to operate at a lower overall cost. “The only
payment for the transfer of risk is the coincident transfer of the chance
of gain,” Emery argued. “To assume risks in this manner is the function
of the speculator.” The result, he concluded, was that speculation was
essential: “It is difficult to see how a great world trade in such staples as
grain and cotton would be possible without it.”40
The economist Albert Stevens offered another defense of speculation
with an equally modern ring. When prices fell as a result of speculation,
he argued, the speculator “was not unduly depressing the price. He was
putting quotations … where they should have been.” Speculators on the
commodity exchanges performed a service, in Stevens’s view, by caus-
ing the prices of commodities to reflect the commodities’ true value, a
service that no other institution could possibly perform. Speculators
“put the price of wheat where it belonged, from the point of view of
its statistical situation,” Stevens suggested; “they did that which nei-
ther the national Agricultural Department, the Vienna Congress,
Bradstreet’s, the Cincinnati Price Current, Beerbohm’s or Dornbusch’s
(London) Lists, or any other of the governmental and private statistical
bureaus of the world had been able to do.” Once these prices had been

40 Henry Crosby Emery, Speculation on the Stock and Produce Exchanges of the
United States (New  York:  Columbia University, 1896), 98; Henry Crosby
Emery, “Legislation Against Futures,” Political Science Quarterly 10 (1895): 62–​
64; Henry Crosby Emery, “The Place of the Speculator in the Theory of
Distribution,” American Economic Association Publications 1 (1900): 103.
  133

The Anti-Option Era 133

established, added the Chicago law professor Van Buren Denslow, they
were telegraphed around the world, “so that every producer and pur-
chaser gets his quotation with his morning paper and as often during
the day as he wishes.” The value of such timely and exact knowledge of
prices, Denslow explained, could be seen by contrasting grain markets
with markets in nonstandardized items like clothing, furniture, and
jewelry, where participants had to spend substantial time simply trying
to ascertain the true value of what they were buying and selling. With
the commodities traded on exchanges, market participants “know that
millions of dollars are staked in behalf of a rise in price, and an equal
number of millions in favor of a fall, and that the actual price quoted
is the precise point at which these millions balance. They know that
the published quotation is not one fixed by the arbitrary determina-
tion of any one dealer, but by the aggregate verdict of them all.” As the
Cornell economist James Boyle concluded in his book-​length defense
of the board of trade, speculation’s long-​term benefits greatly exceeded
whatever temporary dislocations it might cause.41
These new, sophisticated defenses of speculation joined the older,
more intuitive defenses that had been circulating for a century.
Opponents of regulation still argued that every merchant was in some
sense a speculator. They still pointed out that speculation was the driv-
ing force behind the establishment of new enterprises. And they still
contended that regulation would in any event be fruitless. “What a
bewildering effect Washington air must have on the human brain!” ex-
claimed one commercial newspaper while Congress was considering
the Hatch and Washburn bills in 1892. “So many men when they get
there acquire so exaggerated an idea of the power of a statute as panacea
for every commercial and financial irregularity.” The Chicago Tribune
agreed that “the amateur merchants who think they know it all suppose
themselves able to legislate the community into a Utopia.” In fact, the

41 Albert Clark Stevens, “The Utility of Speculation in Modern Commerce,”


Political Science Quarterly 7 (1892): 425; see also Albert C. Stevens, “ ‘Futures’
in the Wheat Market,” Quarterly Journal of Economics 2 (1887):  37–​63; Van
Buren Denslow, “Board of Trade Morality,” North American Review 137
(1883):  375–​76; James E.  Boyle, Speculation and the Chicago Board of Trade
(New York: Macmillan, 1921).
134

134 Speculation

Tribune concluded, “to legislate ‘options’ out of existence will be of no


more avail than the effort of the old lady to sweep back the waves of
the Atlantic with her broom.” In the view of another proponent of the
status quo, “prohibiting ‘speculation’ is babyish,” because “so long as
there is trade, there will be speculative trade.”42
Perhaps as a result of such defenses, the congressional debate over
speculation in commodities grew narrower. By the 1910s, rather than
trying to shut down the exchanges entirely on the theory that all
speculation harmed the farmer, critics tended to focus on particular
speculative practices they found harmful. These specific attacks pro-
duced some significant federal regulation. The first clear example was
the Cotton Futures Act of 1914. The Populists had once railed against
cotton speculation. In the previous three decades, Congress had con-
sidered several bills that would have imposed prohibitive taxes on
cotton speculators to drive the cotton exchanges out of business, and
some of those bills had come close to passing. The Cotton Futures
Act, by contrast, was directed at one specific practice, the rule on the
New  York and New Orleans Cotton Exchanges establishing what
were called “fixed differences” between the prices of different grades
of cotton. The rule was a bit arcane to anyone not in the cotton busi-
ness, but its effect was to allow traders to exploit divergences between
exchange prices and market prices outside the exchange, often to the
detriment of farmers. The Cotton Futures Act required exchanges to
allow the price of all grades of cotton to be determined by their actual
market values.43

42 Joseph C.  Ely, “Monopolies, Labor Unions, and Speculation,” Unitarian


Review and Religious Magazine 26 (1886): 537; “Speculation vs. Investment,”
Christian Union 14 (1876): 227; “Legitimate Speculation,” Christian Union 27
(1883):  3; Morris H.  Smith, “Morals of Speculation,” Christian Advocate 62
(1887): 287; “The Financial Situation,” Commercial and Financial Chronicle 55
(1892): 124; Chicago Daily Tribune, 21 Jan. 1890, 4; “The Offset to Speculation,”
Independent 37 (1885): 28.
43 Francis G. Caffey, “The United States Cotton Futures Act,” in US Department
of Agriculture, Service and Regulatory Announcements No. 5 (1915), 54–​55; S. Rep.
No. 289, 63rd Cong., 2nd Sess. (1914); 38 Stat. 696 (1914); I. Newton Hoffman,
“The Cotton Futures Act,” Journal of Political Economy 23 (1915): 465–​89; Luther
Conant Jr., “The United States Cotton Futures Act,” American Economic Review
  135

The Anti-Option Era 135

Unlike in earlier installments of the debate over speculation, even


the proponents of regulating the cotton exchanges acknowledged the
utility of speculation. “Any legislation … which eliminates from the
cotton trade the element of legitimate speculation and legitimate spec-
ulators, must, in the opinion of the committee, result disastrously to
the producer,” the House Committee on Agriculture declared. “The le-
gitimate speculator, operating through the exchanges, is the only buffer
standing between the helpless producer and the powerful buyer of his
product.” The Cotton Futures Act “is constructive and regulatory, not
destructive or oppressive,” pronounced Secretary of Agriculture David
Houston. “It recognizes that the exchanges, when they are properly
conducted, may benefit both the purchaser of raw cotton and the man-
ufacturer.”44 After decades of contention over speculation, the terms of
the debate had narrowed.
Futures trading on the exchanges was suspended in 1917 due to World
War I. When trading resumed in 1920, commodity prices, which had
been high during the war, promptly plummeted. Corn dropped from
$1.51 to 61 cents per bushel, wheat from $1.82 to $1.03 per bushel, and
cotton from 35 cents to 16 cents per pound. Farmers once again called
for regulating the exchanges, but in Congress, the climate was nothing
like it had been in the 1890s. Even most of those who favored legislation
acknowledged that futures trading was useful. “The only one absolutely
in favor of entirely abolishing the legitimate hedge, I think, is a man
who does not know what a legitimate hedge is,” declared Representative
Jasper Tincher of Kansas, one of the House sponsors of the bill. The
farmers’ organizations had once been the most outspoken critics of
the commodity exchanges, but now T.  C. Atkeson, the Washington
representative of the National Grange, testified to the importance of
preserving the ability to hedge. Speculators were necessary because they
took the other side of hedging transactions, Atkeson explained. The

5 (1915): 1–​11. The Cotton Futures Act was held unconstitutional in Hubbard


v. Lowe, 226 F. 135 (S.D.N.Y. 1915), but was promptly re-​enacted in constitu-
tional form, 39 Stat. 476 (1916).
44 H.R. Rep. No. 765, 63rd Cong., 2nd Sess. (1914), 6; Annual Reports of the
Department of Agriculture (Washington, D.C.: Government Printing Office,
1914), 23.
136

136 Speculation

speculators were gambling, he concluded, but “it is perhaps justifiable


gambling,” because it served a useful purpose. Not everyone agreed.
“It is not necessary for men to speculate in futures upon the boards of
trade,” declared James Strong of Kansas, who reported that the farmers
in his congressional district still wanted to “stop gambling in the staple
foodstuffs of the nation.”45 But this view seems to have been less widely
held than it had been a generation before.
Legislators still declared that they wanted to prohibit a certain sort
of gambling, the kind that had no useful purpose, but they had all but
given up hope of being able to do it without curtailing the speculative
transactions that most parties wanted to allow. “The point is how are
we going to get at the bad features and prohibit them and still permit
hedging,” wondered George Norris, the chair of the Senate Agriculture
Committee. “How can you tell the difference between a hedge and a
gamble?” As a representative of the Minneapolis Chamber of Commerce
admitted, “the difference between gambling and speculation or invest-
ment is a rather difficult thing to state.” The Senate requested guidance
from the Federal Trade Commission, which offered a few possible ways
of drawing the line between speculation and gambling but concluded
that “it is questionable whether any of these definitions is capable of
practical administrative application in distinguishing the speculator
from the gambler.”46
If the good speculation could not be distinguished from the bad,
commodity speculation could at least be required to take place in a
small number of easily observable arenas. In the Future Trading Act
of 1921, Congress accordingly imposed a prohibitive tax on futures
trades conducted by anyone other than the actual physical owner of a
commodity but exempted transactions on government-​approved com-
modity exchanges. To qualify for approval, an exchange had to comply
with several requirements: it had to report the terms of all transactions
to the government, it had to prevent its members from disseminating
false information that could affect prices, and it had to prohibit price

45 Future Trading, 7, 300–​301, 64–​65.


46 Future Trading in Grain, 123, 290; Report of the Federal Trade Commission on
Methods and Operations of Grain Exporters (Washington, D.C.: Government
Printing Office, 1922), I:xxii.
  137

The Anti-Option Era 137

manipulation. The goal, explained the Senate Agriculture Committee,


was “to correct some practices on the grain exchanges and to authorize
supervision of the grain-​futures markets, but not to disturb any of their
legitimate and useful functions.” Market participants would still be free
to hedge and to speculate. The law would merely “eliminate from the
market some of the undesirable practices of professional speculators.”47
The Supreme Court found the Future Trading Act an unconstitu-
tional exercise of the federal government’s taxing power, on the ground
that it was really a commercial regulation disguised as a tax, so Congress
promptly re-​enacted it as the Grain Futures Act of 1922, but this time as
a regulation of interstate commerce rather than a tax. Congress blessed
the new law the following year.48 Ever since, commodity futures trading
has been lawful on an approved exchange, but not otherwise.
In retrospect, the period from the early 1890s through the early
1920s saw the first sustained national debate over the propriety of par-
ticular forms of speculation. The debate forced both sides to articulate
theories distinguishing legitimate business from illegitimate gambling.
The result was a compromise:  commodity speculation could go on,
but only under the close supervision of the federal government. There
would never again be so much argument over speculation in agricul-
tural commodities. Speculation in financial assets, by contrast, would
remain controversial throughout the twentieth century. Indeed, while
the old debate over commodities was winding down, a new debate over
the government’s role in supervising the stock market was just getting
underway.

4 7 42 Stat. 187 (1921); S. Rep. No. 212, 67th Cong., 1st Sess. (1921), 4.
48 Hill v. Wallace, 259 U.S. 44 (1922); 42 Stat. 998 (1922); Board of Trade v. Olsen,
262 U.S. 1 (1923).
4
Q
Selling Blue Sky

It started more as a hobby than an official government activity.


But within a few years it turned into one of the fastest-​spreading and
farthest-​reaching regulatory programs the country had ever seen.
“Why, I had been in the banking business here in Kansas a good many
years before I  became banking commissioner,” Joseph Dolley recalled.
“Every now and then I would hear of one of these swindles—​that some-
body had lost his money through buying stock in a fake mine, or in a
Central America plantation that was nine parts imagination, or in some
wonderful investment company that was going to pay forty per cent divi-
dends.” As chairman of the state Republican Party, Dolley had achieved a
double victory in the 1908 election, when Walter Stubbs was elected gov-
ernor and the state’s electoral votes went to William Howard Taft. Stubbs
rewarded Dolley by making him the state banking commissioner. “After
I was appointed banking commissioner,” Dolley continued, “I heard more
reports and complaints of fake stock swindles than ever.” Times were good
for Kansas farmers, who were beneficiaries of high commodity prices and
thus had money to invest. “So reports of these stock swindles drifted to
me. I received complaints and inquiries direct from people who had been
swindled, wanting me to look up the company and see if they couldn’t get
their money back—​after they had parted with the money!” Dolley soon
realized that stock sellers, mostly from out of state, were keeping an eye
on real estate sales and probate courts, to identify inexperienced investors
who were coming into money and would be easy marks.1

1 Will Payne, “How Kansas Drove Out a Set of Thieves,” Saturday Evening Post,
2 Dec. 1911, 3–​4.
138
  139

Selling Blue Sky 139

So Dolley began providing free investment advice. He placed notices


in the state’s newspapers informing Kansans that if they were offered
stock for sale, they were welcome to write to him with inquiries about
the financial status of the company. The letters began streaming in. “In
the case of every one of these companies I can give exact and authen-
tic information about their financial standing and their investments
and their general character,” Dolley told the press. “We have access to
financial and industrial information of all kinds and know just where
to get in touch with any company that is promoting something.” Soon
Dolley was receiving telephone calls as well. One day in June, for ex-
ample, a Topeka widow called Dolley for advice. She had just received
the proceeds of her late husband’s life insurance when she was visited by
a stock salesman who promised a 30 percent annual dividend. Dolley
told her to buy municipal bonds instead.2
After several months of such informal consultations, Dolley re-
quested an appropriation from the state legislature to hire employees to
carry on the work. He also requested the enactment of a state law that
would require everyone wishing to sell stock in Kansas to get permis-
sion from his office, which would assess the merits of an offering before
allowing it to be sold. The legislature gave him everything he asked for.
In 1911, Kansas enacted the first of what would come to be called “blue
sky laws.” Before anyone could sell stock in Kansas, they would first
have to persuade Joseph Dolley and his staff that the stock was a sound
investment.3
The idea swept the country. Within two years, half the states had
blue sky laws. Within two decades, nearly all did.4 They were called

2 Kansas City Journal, 2 Apr. 1910; Topeka Journal, 8 Apr. 1910; Kansas City
Journal, 4 June 1910. The citations in this chapter to Kansas newspapers are
from the Kansas Historical Society’s Kansas Memory website, at http://​www.
kansasmemory.org.
3 Tenth Biennial Report of the Bank Commissioner of the State of Kansas
(Topeka:  State Printing Office, 1910), xiii; Rick A.  Fleming, “100 Years of
Securities Law:  Examining a Foundation Laid in the Kansas Blue Sky,”
Washburn Law Journal 50 (2011): 583–​609.
4 The blue sky laws, with dates of adoption, are usefully summarized in Paul
G.  Mahoney, “The Origin of the Blue-​ Sky Laws:  A  Test of Competing
Hypotheses,” Journal of Law & Economics 46 (2003): 232. For details of the blue
140

140 Speculation

blue sky laws because “blue sky” was a common metaphor for a worth-
less investment. “When a promoter by artful persuasions succeeds in
getting money for something which has no value except in the mind
of the credulous purchaser,” an 1895 newspaper explained, “he is said
to have been selling ‘blue sky.’â•›” In a 1906 advertisement for shares in a
Colorado mine, the promoter insisted, “I am not selling ‘Blue Sky.’â•›”5
Selling blue sky was what the blue sky laws were supposed to prevent.
The blue sky laws responded to a very old concern with speculation—╉
that speculators were likely to harm themselves and their families by
taking on risks they lacked the ability to evaluate intelligently. But if
the problem was an old one, the solution was something that had never
been tried before: to have the government evaluate the risks and guide
investors to safety. If the state could screen out the investments that
were too speculative or were likely to be frauds, one of the primary
dangers of speculation would be a thing of the past.

A Fair€Return
No one could have expected the blue sky law movement in 1910, when
Joseph Dolley opened his informal investment advisory service in the
office of the Kansas state bank commissioner. In retrospect, to be sure,
one can see the blue sky laws as emblematic of a broader concern for
consumer protection in the progressive era, the same concern that gave
rise to pure food and drug laws, occupational licensing, and antitrust
laws.6 More broadly, one can see the blue sky laws as an example of
the use of government to mitigate risk, like the establishment of work-
ers’ compensation and unemployment insurance, two other develop-
ments that were then in progress.7 But at the time it would have been

sky laws in effect as of 1919, see John M. Elliott, The Annotated Blue Sky Laws of
the United States (Cincinnati: W. H. Anderson Co., 1919).
5 Lawrence R. Gelber, The Gelberlaw Glossary, http://╉www.gelberlaw.net/╉Glossary
.html; Chicago Tribune, 25 Feb. 1906, 10.
6 Morton Keller, Regulating a New Economy: Public Policy and Economic Change
in America, 1900–╉1933 (Cambridge, Mass.: Harvard University Press, 1990).
For contemporary recognition of this point, see William B. Shaw, “Progressive
Law-╉Making in Many States,” Review of Reviews 48 (1913): 84–╉93.
7 David A. Moss, When All Else Fails: Government as the Ultimate Risk Manager
(Cambridge, Mass.: Harvard University Press, 2002).
  141

Selling Blue Sky 141

very difficult, probably impossible, to predict that the first serious stock
market regulation of the twentieth century would take the form of blue
sky laws. In the preceding decades, while there had been calls to regu-
late securities sales in just about every other conceivable way, no one
had proposed that government officials should decide, company by
company, which investments would offer a fair return.
All the old worries about speculation were still very much alive.
Speculation was still widely viewed as a form of gambling. “Of all the
varied types of gambling doubtless the most colossal is that of the great
stock exchanges,” declared one minister in 1895, “and the success of
every rich man’s corner is apt to be followed by the dishonesty or sui-
cide, or both, of some of his victims.” As the humor magazine Puck de-
scribed the successful speculator, “it is a common thing to see bestowed
upon a man who began with nothing of his own and ended with a great
deal of other people’s property the flattering commendation that he
was a self-​made man.” The press was still full of articles worrying that
speculators would bankrupt themselves and their families, that specula-
tion was a nonproductive activity, and that prices were controlled by
a small cabal of insiders. “The editor who fails to have a fling at Wall
Street as often as two or three times a week, especially if he live[s]‌in the
South or West, must be unusually well supplied with prize fights and
divorce cases,” smirked Brayton Ives, the president of the New  York
Stock Exchange. “And what demagogue addressing his constituents
omits in any speech to denounce it as the home of the monopolist and
the hotbed of injustice?” The investment banker Henry Clews agreed
that “it seems to be a genial pastime for men in various walks of life
who know very little about financial affairs, and the methods of doing
business in Wall Street, to denounce this great centre of the moneyed
interests, as the sum of all villanies, a kind of Pandora’s box, but with-
out any hope at the bottom.”8

8 “Gambling Speculations,” Bankers’ Magazine 14 (1879): 19; John F. Hume,


“The Heart of Speculation,” Forum, Oct. 1886, 130; Joseph Parker, “Gambling
and Speculation,” Independent 49 (1897): 6; C. H. Hamlin, “Gambling, or
Theft by Indirection,” Arena 2 (1895): 414; W. J. Henderson, “Speculators Not
Appreciated,” Puck 18 (1885): 227; “Money Matters,” New York Evangelist 61
(1890): 4; “Speculation and Business,” Bankers’ Magazine 37 (1882): 88; “Stock
Speculation,” Independent 38 (1886): 24; “To Amateur Speculators,” Christian
142

142 Speculation

Such concerns brought forward a wide variety of proposals for regu-


lation in the late nineteenth and early twentieth centuries. “The enter-
prising speculator rejoices and grows fat in that domain that lies just
beyond the common law, and has not yet attracted the attention of
statutory enactment,” the Bankers’ Magazine observed. “Here he can
give himself full swing until the slow foot of legislation overtakes him.”9
Toward the turn of the century, as legislatures passed more and more
statutes regulating various aspects of the economy, many suggested that
it was high time to regulate the speculators as well.
These proposals tended to avoid outright bans on speculative trans-
actions, because of the perennial problem of how to prohibit the bad
without also forbidding the good. As the New York investment banker
Adolph Lewisohn noted, “it is very difficult to draw the line between
where investment ceases and speculation commences.”10 The regulatory
proposals of the late nineteenth and early twentieth centuries were thus
primarily directed at other targets.
Perhaps the most common complaint about the stock market was
that corporate directors and officers had an unfair advantage, because
they had more information about their own firms than anyone else did.
“They know what is going to happen beforehand,” one critic alleged.
“There is no ‘future’ in the stock market to these men: they themselves
make the future.” The ordinary investor had to pit his “blank igno-
rance against billions of money and certain knowledge. He is guess-
ing where they are certain sure.” Worse, corporate insiders could de-
liberately spread false information about their own firm to profit when
the true information was revealed and the stock price moved back to
where it ought to have been all along. “Directors who are unscrupulous
speculators start and circulate rumors of the badness of the business,”
complained one lawyer. They could make money by selling short as the

Union 21 (1880): 242; “Speculation,” Ohio Farmer 69 (1886): 104; W. R. Givens,


“Does Wall Street Speculation Pay?,” Independent 59 (1905): 496; “The Spirit of
Speculation,” National Police Gazette, 12 Feb. 1881, 11; Brayton Ives, “Wall Street
as an Economic Factor,” North American Review 147 (1888): 555; Henry Clews,
“Delusions About Wall Street,” North American Review 145 (1887): 410.
9 “Agitation and Speculation,” Bankers’ Magazine 56 (1898): 808.
10 New York Times, 25 July 1909, 6.
  143

Selling Blue Sky 143

stock price declined, and then make even more by buying as it went
back up. “This is pure swindling,” he concluded.11
To level the playing field, some proposed requiring corporations to
disclose their financial status. “Make the seller of securities publish an
accurate analysis of the values, equities, and earnings of which these
securities are representative,” one writer urged. “Such reports should
be promptly published in order to permit investors to see exactly what
the assets and liabilities are,” agreed Charles Batchelder, a director of the
Boston Lumber Company. If the law could require food to be honestly
labeled, he suggested, it could do the same for securities. “Publicity,”
he concluded, was “the best safeguard [not only] for the investor, but
also for the community.” In the first decade of the twentieth century, a
handful of states began to require such disclosures, but only for public
utility companies and railroads, two kinds of firms the states already
regulated more intensively than other forms of enterprise. The econo-
mist Earl Howard even proposed a similar disclosure requirement for
brokers. If they had to reveal the name of the person for whom they
were trading, he argued, the manipulation of prices by insiders would
become impossible. “No manipulator would care to try the game with
all the cards on the table,” he concluded, “for the first essential of ma-
nipulation is secrecy.”12 But no state adopted this proposal.

11 “A Game of Speculation,” Albion 51 (1873): 328; “Hydraulic Pressure in Wall


Street,” North American Review 135 (1882): 60; Samuel Untermyer, “Speculation
on the Stock Exchanges and Public Regulation of the Exchanges,” American
Economic Review Papers and Proceedings 5 (1915):  47–​48; “The Confessions
of a Stock Speculator,” Independent 62 (1907):  672; “Directors as Stock
Speculators,” American Law Review 3 (1882): 919.
12 Theodore H.  Price, “Speculation:  The Wisdom of Restricting It by Law or
Otherwise,” Outlook, 16 May 1914, 128; Charles C. Batchelder, “The Character
and Powers of Governmental Regulation Machinery,” Journal of Political
Economy 20 (1912): 393, 386; Arthur U. Ayres, “Governmental Regulation of
Securities Issues,” Political Science Quarterly 28 (1913): 586–​92; Leo Sharfman,
“Commission Regulation of Public Utilities: A Survey of Legislation,” Annals
of the American Academy of Political and Social Science 53 (1914): 1–​18; William
Z. Ripley, “Public Regulation of Railroad Issues,” American Economic Review 4
(1914): 541–​64; Earl Dean Howard, “The Speculators and the Banks,” Bankers’
Magazine 77 (1908): 191.
144

144 Speculation

Other proposed laws were meant to address other perceived prob-


lems with speculation. To deter overspeculation, how about prohibiting
banks from lending money to speculators? To ensure that a speculator’s
distress did not spread to others, what about banning speculators from
investing on behalf of others? After the bank failures of the Panic of
1907, some suggested that banks should no longer be allowed to specu-
late with their depositors’ funds. The financial writer Charles Conant
thought the most constructive thing the government could do was “not
to hamper legitimate corporations by new laws, but to teach the public
to judge investments with discrimination.” If people could be taught
“that high returns almost inevitably mean risk,” Conant concluded,
they would be much less tempted by uninformed speculation.13 None
of these ideas was put into practice at the time.
During this period, the most thorough review of possible meth-
ods of regulation was undertaken in the wake of the Panic of 1907
by a “Committee on Speculation in Securities and Commodities”
appointed by New York governor Charles Evans Hughes. The com-
mittee was chaired by the financial journalist Horace White, and its
members were praised by one magazine as “men of large knowledge
both in practical affairs and in the theory of finance.” They included
Charles Schieren, a banker and the former mayor of Brooklyn; David
Leventritt, who had recently retired as a judge; Clark Williams, the
state’s bank superintendent; and John Bates Clark, the Columbia econ-
omist. “It is unquestionable that only a small part of the transactions
upon the Exchange is of an investment character; a substantial part may
be determined as virtually gambling,” the committee observed. “Yet we
are unable to see how the State could distinguish by law between proper
and improper transactions, since the forms and the mechanisms used
are identical.” The committee nevertheless offered several regulatory
proposals, including a ban on misleading or unsupportable promises

13 “Banks and Speculation,” Bankers’ Magazine 36 (1882):  572; “The Credit


System, and Speculation,” Oneida Circular 12 (1875):  284; “Speculation
With Bank Deposits,” Independent 68 (1907):  1063; “Bank-​Plundering by
Speculators,” Outlook, 23 Oct. 1909, 359; Charles A. Conant, Wall Street and
the Country (New York: G. P. Putnam’s Sons, 1904), 48–​51.
  145

Selling Blue Sky 145

in advertisements for stock.14 The committee’s proposals did not result


in any legislation.
Yet for all these proposed limits on speculation during the late
nineteenth and early twentieth centuries, the one thing that was not
proposed—​not until Joseph Dolley began putting it into practice in
Kansas—​was a government agency to evaluate whether a proposed sale
of stock would be a good investment. The blue sky laws came from out
of the blue.
Once the Kansas blue sky law went into effect in early 1911, stocks
and bonds could not be sold in Kansas unless the seller could persuade
Joseph Dolley that they promised “a fair return” on the investment.
“A man came in here yesterday with a stock proposition that wasn’t
worth a continental,” Dolley related a few days after the law took effect.
The man “said he had been selling it in Kansas, but that he had read
about the ‘blue sky’ bill and wanted to know what we were going to do
about it.” Dolley had his first request for permission to sell securities
in Kansas. “I looked at his stock and found that we had already inves-
tigated it thoroughly,” before the law had been enacted, when Dolley
was conducting these investigations on his own at the request of po-
tential purchasers. “I told him he couldn’t sell another dollar’s worth
in Kansas.”15
The volume of applications to sell stock quickly grew very high, but
his office approved very few of them, at least according to the figures
that Dolley regularly released to reporters. In October 1911, after only
seven months under the blue sky law, Dolley reported that five hundred
companies had applied to sell securities in Kansas, that his office had
approved only forty-​four of the applications, and that the office sup-
posed that many more would-​be fraudsters had been scared away by
the prospect of filing an application. By March 1912, Dolley’s numbers
were up to more than seven hundred applications, of which only forty-​
eight had been approved. In his September 1912 report to the governor,
Dolley claimed that between fourteen hundred and fifteen hundred

14 “Speculation and Gambling,” Outlook, 26 June 1909, 422; Report of the


Committee on Speculation in Securities and Commodities (Albany, N.Y.: J. B.
Lyon Co., 1910), 8, 28–​29.
15 Kansas Session Laws 1911, ch. 133, § 5; Topeka Capital, 18 Mar. 1911.
146

146 Speculation

companies had been investigated by his office, and fewer than one hun-
dred had been allowed to sell securities. Dolley also provided recurring
but widely variant estimates of how much the people of Kansas had
saved in investment losses by virtue of his guidance—​amounts ranging
from $2 million to $77 million in the first year alone. He often declared
with pride that whatever the annual figure, it was more than the cost of
running the entire state government for the year.16
How could an office consisting of only a few people conduct in-
vestigations of so many companies in such a short time? The task was
perhaps not as monumental as it might seem. By Dolley’s own account,
most of the firms “never get farther than making the application. It was
not necessary for us to turn them all down. As soon as they found out
what information we were going to ask them for and what the nature
of our investigation was going to be they suddenly changed their minds
and withdrew their applications.” When necessary, Dolley could also
seek advice from other government officials and from experts at the
University of Kansas. “For example,” he explained, “a company recently
came to the department with a new electric battery, which they claimed
was going to revolutionize the electrical world.” The firm had endorse-
ments from prominent bankers. “We at once presented the matter to
the electrical experts at our State University, and they advised us within
thirty days that the battery was an absolute fake, and would not do
what the company claimed for it.” Dolley refused to allow the company
to sell stock in Kansas.17
But even with many applications aborted, and even with the as-
sistance of experts, Dolley’s office still had to evaluate the merits of
hundreds of securities offerings, many of which were based on assets
located out of state. Dolley claimed to conduct thorough investiga-
tions of all of them. “If the financial statement does not satisfy him—​
and he is a trained banker—​he sends an expert auditor to go over

16 New York Times, 13 Oct. 1911, 10; Wall Street Journal, 2 Mar. 1912, 6; Eleventh
Biennial Report of the Bank Commissioner of the State of Kansas (Topeka: State
Printing Office, 1912), 6; William Allen White, “Free Kansas: Where the People
Rule the People,” Outlook, 24 Feb. 1912, 412; “A Blue Sky Law,” Independent 72
(1912): 375; Topeka Capital, 12 May 1912.
17 “The Kansas ‘Blue Sky Law,’ ” Central Law Journal 75 (1912): 222.
  147

Selling Blue Sky 147

the company’s books,” marveled one magazine, based on information


supplied by Dolley. “If the applicant is a gold-​mine, he sends a trained
engineer to the spot to investigate; if it is an industrial concern, the
best man available goes to look over the plant and its product.”18 This
would have been an enormous expenditure of time and money, if
Dolley’s accounts were true.
But they were probably not true. Dolley’s term as bank commissioner
ended with the election of 1912, when George Hodges, a Democrat,
became governor. Not long after, the Canadian lawyer and government
official Thomas Mulvey read through the files in Dolley’s old office,
because Canada was considering adopting a blue sky law and wanted
to learn from the experience of Kansas. According to Mulvey, from
March 1911 through March 1913, permits to sell securities had been
granted to forty-​nine companies and denied to only sixty-​two. Dolley
had claimed to have resolved fourteen hundred to fifteen hundred cases
in the first eighteen months of that period. Mulvey found no basis for
any of Dolley’s claims about how much money he had saved the people
of Kansas. The office did not keep such statistics. Mulvey also discov-
ered that Dolley had been granting temporary permits to firms that had
not been investigated, a practice the blue sky law did not authorize.
Some of these firms had committed serious frauds. The state’s blue sky
law was much less impressive than Dolley’s enthusiastic public relations
suggested.19
The Kansas blue sky law nevertheless attracted great interest all over
the country, and indeed in much of the world. Governor Walter Stubbs
received inquiries from Missouri to New York and many states in be-
tween, all asking for copies of the Kansas law. “Such a law is very much
needed in our state for the protection of its citizens from the confidence
sharks,” declared E. T. Merritt of Minneapolis. Dolley even heard from
the British and German consuls, who requested information they could
send home to their governments. By the spring of 1913, Connecticut,
Arizona, West Virginia, and Vermont had enacted blue sky laws. So had
Manitoba and New South Wales, which copied the Kansas law almost

18 Isaac F. Marcosson, “Barring Out the Stock Thieves,” Munsey’s Magazine 46


(1912): 680.
19 Thomas Mulvey, “Blue Sky Law,” Canadian Law Times 36 (1916): 39–​40.
148

148 Speculation

verbatim. By the end of 1913, twenty-​four states had blue sky laws. Most
of the remaining states would have blue sky laws by the end of the
decade. Even the US-​governed Philippines got one.20
Why were blue sky laws so popular? Some of their support no doubt
came from prospective investors, who were genuinely fearful of being
defrauded by fast-​talking salesmen. Public participation in the securities
markets increased dramatically in the early twentieth century.21 Many
people were buying stocks and bonds for the first time. They must have
welcomed some guidance from the state. Stock promoters “are said to
fleece Californians out of at least $20,000,000 a year,” gasped the San
Francisco Chronicle in an editorial urging California to give its citizens
the protection Kansans already enjoyed. In the paper’s view, the blue
sky law was simply “a good idea, floated on a good phrase.”22
But inexperienced investors were not the only supporters of blue
sky laws. In Kansas, Joseph Dolley’s scrutiny of securities sales had
been welcomed by the state’s business community, who recognized
that the blue sky law would be a barrier to out-​of-​state competitors.
One of Topeka’s leading bond brokers praised Dolley for making it
easier for in-​state firms to sell bonds. The treasurer of the Atchison,
Topeka & Santa Fe Railroad was just as happy with Dolley’s work. “I
believe the established corporations of the state … should receive the
support of the people of the state,” he explained. “A Kansan having

20 Letters to Walter Stubbs from John Sullivan, 7 Feb. 1912; W. C. Edson, 16 May
1912; H. J. Fitzgerald, 9 July 1912; Wade H. Barnes, 28 Aug. 1912; John Wright,
12 Oct. 1912; William Dinwiddie, 30 Dec. 1912; E. T. Merritt, 30 Dec. 1912, all
from the Kansas Historical Society’s Kansas Memory website, at http://​www
.kansasmemory.org; Topeka Capital, n.d.; C.  A. Dykstra, “Blue Sky
Legislation,” American Political Science Review 7 (1913):  231–​32; Los Angeles
Times, 5 Feb. 1916, 12.
21 Julia C. Ott, When Wall Street Met Main Street: The Quest for an Investors’
Democracy (Cambridge, Mass.: Harvard University Press, 2011); Mary
O’Sullivan, “The Expansion of the U.S. Stock Market, 1885–​1930: Historical
Facts and Theoretical Fashions,” Enterprise and Society 8 (2007): 489–​542. For
an argument that increased public participation in the stock market funda-
mentally changed the nature of speculation, an argument I find unpersuasive,
see Lawrence E. Mitchell, The Speculation Economy: How Finance Triumphed
Over Industry (San Francisco: Berrett-​Koehler Publishers, 2007).
22 San Francisco Chronicle, 11 Sept. 1912, 6.
  149

Selling Blue Sky 149

$100 or $500 to invest could buy preferred or common stock of the


Atchison Topeka & Santa Fe Railway Company.” Dolley himself felt
the same way. “We are putting in a foundation for a great work,” he
declared, “and it means Kansas money for Kansas investments.” Local
businesses had local reputations and their managers had local con-
tacts. If the officials administering a state’s blue sky law scrutinized
firms from other states more closely than they scrutinized local firms,
a blue sky law could be an effective means of protectionism, ben-
efitting in-​state banks and securities issuers by hindering out-​of-​state
competition.23
In state after state, local business groups were thus among the most
vocal supporters of blue sky laws. In Illinois, the push for a blue sky
law was led by the Illinois Bankers’ Association. In Ohio it was the
Ohio Bankers’ Association. In California, a group called the Credit
Men’s Association played the same role. “The law has the support of
the well-​known and reputable bond houses” of California, the San
Francisco Chronicle explained, because “although the law is general in
its terms, such houses have no difficulty in establishing their standing
with the Commissioner.” At least in some cases, this expectation of in-​
state favoritism was well justified. Of the firms Joseph Dolley allowed
to sell securities, most turned out to be from Kansas. In Nebraska, the
state official responsible for blue sky matters frankly declared that his
“Department is opposed to the sale of securities in this state by any
eastern financial house” and would accordingly “interpose every avail-
able legal obstacle to their sale.”24
The most significant opponents of blue sky laws were the elite urban
investment bankers, who sold securities in multiple states and were
thus the ones who lost out when states favored local firms. They broke

23 Topeka Journal, 25 Sept. 1910; Kansas City Journal, 30 Dec. 1910; Jonathan
R. Macey and Geoffrey P. Miller, “Origin of the Blue Sky Laws,” Texas Law
Review 70 (1991): 347–​97. For the view that protectionist concerns were a big-
ger influence on the form of blue sky laws than on whether blue sky laws were
adopted, see Mahoney, “Origins of the Blue-​Sky Laws,” 229–​51.
24 Chicago Daily Tribune, 18 Feb. 1912, 5; 22 Feb. 1912, 14; 26 Feb. 1912, 14;
Wall Street Journal, 22 Nov. 1912, 6; Los Angeles Times, 17 Oct. 1912, II2; San
Francisco Chronicle, 3 June 1915, 16; “The Kansas ‘Blue Sky Law,’ ” Central Law
Journal 75 (1912): 222; Lee Herdman to L. W. Berle, 6 Oct. 1932, AAB, box 20.
150

150 Speculation

away from the American Bankers Association in 1912 to form their


own group, the Investment Bankers Association of America, in large
part to lobby against blue sky laws. At the IBA’s first annual conven-
tion, held in New York in the fall of 1912, the Cleveland investment
banker Warren Hayden spoke at length on the danger “that legislation
may be brought forward by those who have no familiarity with the fi-
nancial mechanism of the country, or with the investment business in
particular, and that as a result of careless work, however well intended,
we might find ourselves obliged to modify our business.” He pointed
out that an established investment bank might have a list of two hun-
dred different securities for sale at any given time, adding perhaps five
new ones every day, each of which would have to be individually ap-
proved by the Kansas banking commissioner before it could be sold in
Kansas. If other states enacted similar laws, Hayden predicted, invest-
ment banks would be overwhelmed by the task of complying with all
of them.25
The IBA’s primary argument against the Kansas-​style blue sky laws,
and the argument that gained the most traction in the press, was that
they vested in government officials a dangerous degree of power over
the economy. “A state’s commercial and industrial progress is in the
hands of one man,” worried the Wall Street Journal. “Human fallibility
is not equal to the task. New companies must necessarily be constantly
formed if the country’s resources are to be developed. And to give to
one individual (or any set of individuals) the arbitrary power of saying
this company may sell its securities and that one may not, is dangerous
in the extreme. The solid investments of today were the speculations
of yesterday.” When the New  York Assembly passed a blue sky bill,
the New  York Times joked that state officials could not even manage
their own affairs competently, much less supervise the investment de-
cisions of all New  Yorkers. The Los Angeles Times thought the state’s
blue sky law ought to be renamed a “black-​sky law” or a “murky-​sky

25 Vincent P.  Carosso, Investment Banking in America:  A  History (Cambridge,


Mass.: Harvard University Press, 1970), 165–​73; Proceedings of the Organization
Meeting and of the First Annual Convention of the Investment Bankers’ Association
of America (Chicago: Investment Bankers’ Association of America, 1912), 140,
144–​45.
  151

Selling Blue Sky 151

law,” because it made politicians “the judges of the solvency … and the
honest intentions of corporations.”26
The IBA pursued a two-​part strategy. The first part was to propose
a milder blue sky law for states to adopt, one that required sellers of
securities to disclose information about the securities but did not em-
power a government official to determine whether they would be a
sound investment. The IBA wished to avoid giving the “impression that
the position of the Association is negative and opposed to the popu-
lar sentiment in favor of necessary legislation to prevent the promo-
tion of fraudulent securities,” explained Robert Reed, the IBA’s lawyer.
Instead, “we should try to work out an effective general law that will
meet the situation,” a law that “will settle the ‘blue sky’ frauds without
further resort to ‘blue sky’ or imaginative remedies.” Reputable sellers
of securities had an interest in driving the disreputable out of business,
so long as that could be accomplished without unduly burdening the
reputable. The IBA accordingly drafted a model blue sky law, which it
circulated to its members, who sought to have it introduced in their
home state legislatures. The IBA’s version of a blue sky law borrowed
from the British Companies Act of 1908, which required a company
selling securities to file a prospectus with the government, disclosing a
wide range of information about the company.27
The second part of the IBA’s strategy involved filing lawsuits seeking
to have the existing blue sky laws declared unconstitutional. The IBA
did not file the suits itself; rather, it organized them, by identifying
plaintiffs, retaining local counsel, and formulating legal arguments. The
plaintiffs were typically out-​of-​state investment banks and corporations.

26 “The ‘Blue-​Sky’ Laws,” Bankers’ Magazine 84 (1912):  636; Henry C.  Emery,
“Speculation on the Stock Exchanges and Public Regulation of the Exchanges,”
American Economic Review Papers and Proceedings 5 (1915):  80; Wall Street
Journal, 11 Apr. 1913, 1; New York Times, 27 Mar. 1913, 10; Los Angeles Times, 30
June 1913, II4; 1 Mar. 1913, II8.
27 “Report of General Counsel,” Investment Bankers’ Association of America
Bulletin 2 (1914):  8–​9; Robert R.  Reed and Lester H.  Washburn, Blue Sky
Laws: Analysis and Text (New York: Clark Boardman Co., 1921), x–​xi; “Pursuit
of Swindlers,” Independent 74 (1913): 486; Wall Street Journal, 21 Feb. 1913, 5;
Wall Street Journal, 26 Nov. 1912, 1; 27 Mar. 1913, 1; “Blue Sky Laws,” Columbia
Law Review 24 (1924): 79–​80.
152

152 Speculation

These efforts bore fruit quickly. In 1914, courts invalidated the blue sky
laws of Michigan, Iowa, and West Virginia. The Michigan judges seem
to have been taken aback by the breadth of the state’s law, which, as
in Kansas, prohibited the sale of securities if a state commission con-
cluded that investors were likely to lose their money. “It does not cover
fraudulent securities merely, but reaches the sale of securities that are
honest, valid, and safe,” the court huffed. “It does not simply protect
the unwary citizen against fraudulent misleading, but it prevents
the experienced investor from deliberately assisting an enterprise which
he thinks gives sufficient promise of gain to offset the risk of loss.” The
judges were equally alarmed by the commission’s “uncontrolled discre-
tion” to grant or deny approval based on its view as to the fairness of the
company’s business plan. “Broader and vaguer language could not be
chosen,” the court observed. That was enough for the court to find that
Michigan’s blue sky law unconstitutionally denied due process and in-
terfered with interstate commerce. The Iowa and West Virginia judges
had the same reaction. The Iowa court acknowledged that Iowa’s blue
sky law was enacted with a laudable purpose—​“to protect the humble,
honest citizens of the state, unlearned in the intricacy of business affairs
as conducted at this day from being plundered and despoiled of their
small earnings and property, acquired through years of patient toil, by
the alluring machinations and the deceptive, misleading, and fraudu-
lent devices which the unscrupulous, cunning, and deceitful ‘Get-​Rich-​
Quick-​Wallingfords’ of our day practice.” But a praiseworthy goal did
not allow the state “to punish the doing of such customary, everyday
transactions” as the sale of securities, because the prohibition exceeded
the state’s limited power over interstate commerce.28
The IBA’s winning streak continued in 1915 and 1916. Ohio’s blue
sky law met the same fate. After the Michigan legislature amended its
blue sky law, the court struck it down once again. In light of all these
decisions, concluded the lawyer Lee Perrin, while a state could prohibit

28 “Report of General Counsel,” 6–​7; Topeka Journal, 22 Oct. 1914; New York


Times, 16 Aug. 1913, 1; Wall Street Journal, 27 Aug. 1913, 8; “Blue Sky Laws,”
Independent 77 (1914): 244; Alabama and New Orleans Transp. Co. v. Doyle, 210
F. 173, 175, 181 (E.D. Mich. 1914); William R. Compton Co. v. Allen, 216 F. 537,
545–​46 (S.D. Iowa 1914); Bracey v. Darst, 218 F. 482 (N.D.W.Va. 1914).
  153

Selling Blue Sky 153

fraud in the sale of securities, the state lacked the power to substitute
its own judgment for that of the investor. “With the security and the
material facts once fairly before him,” Perrin advised, “he should be left
to his own discretion.”29
Meanwhile, in states with blue sky laws that were still in force, ad-
ministrators were reconsidering the wisdom of evaluating the sound-
ness of every security offered in the state. Even Joseph Dolley was
having second thoughts as he neared the end of his term in office. It
was simply impossible to investigate them all. He proposed a new law
that would allow reputable investment bankers to obtain licenses that
would entitle them to sell any security, without the state having to scru-
tinize each one. The Kansas legislature enacted this new law in 1913.30
As a result of these three developments—​lobbying by the IBA,
court decisions striking down several of the early statutes, and state
officials’ realization that a full merit review of all securities issues was
impractical—​the nature of blue sky laws changed in the mid-​1910s.
Most of the states in the first wave of blue sky laws, from 1911 to 1913,
sought to prohibit the sale of securities that were unlikely to offer a
fair return. But the second generation of blue sky laws, from 1915 on,
by and large abandoned the effort to evaluate the soundness of invest-
ments. The new laws tended instead merely to prohibit fraud and to
require sellers to disclose information. Even Kansas, which had pio-
neered merit review in 1911, gave it up in 1915. Kansas’s new law, like
its counterparts in other states, omitted any requirement that securities
offer a fair return.31 The IBA’s campaign had been a success.
By the time the constitutionality of blue sky laws reached the United
States Supreme Court, in three cases decided in 1917, all three of the laws
at issue were of the second-​wave type, prohibiting fraud but not oth-
erwise empowering state officials to judge the merits of an investment.
The court upheld all three. “It may be that there are better ways to meet
the evils at which the statute is directed,” Justice Joseph McKenna said

29 Geiger-​Jones Co. v. Turner, 230 F. 233 (S.D. Ohio 1916); N.W. Halsey & Co.
v. Merrick, 228 F. 805 (E.D. Mich. 1915); Lee J. Perrin, “The ‘Blue Sky Laws,’ ”
Bench & Bar 10 (1916): 496
30 New York Times, 18 Mar. 1913, 10; Fleming, “100 Years,” 604–​5.
31 Fleming, “100 Years,” 606.
154

154 Speculation

of Michigan’s revised law. “We can only reply that it is not our func-
tion to decide between measures.” In prohibiting fraud, the state was
doing what it had always done. All that was new was that the state was
policing fraudulent sales before the sales took place rather than after.
“The policy of a state and its expression in laws must vary with circum-
stances,” McKenna continued. “It burdens honest business, it is true,
but burdens it only that, under its forms, dishonest business cannot be
done.”32 Had one of the first-╉wave statutes reached the court, the result
might well have been different.
Only seven years had passed since Joseph Dolley had begun offering
investment advice to the citizens of Kansas. Blue sky laws had swept the
country, but they had weakened. Many states had briefly experimented
with a scheme in which government officials would serve as investment
advisors to the public, separating the good investments from the bad,
to ensure that state residents were not ruined by speculation. The ex-
periment had been a failure. The plan exceeded both the states’ author-
ity under the constitutional law of the era and the limited institutional
capacities of the officials charged with determining which investments
would offer a fair return.

A Ridiculous Situation
Joseph Dolley’s experiment had failed, but the nation was left with a
body of law that had not existed before he began offering investment
advice to Kansans. By the late 1910s and early 1920s, most of the state
blue sky laws had a family resemblance. They required sellers of securi-
ties to disclose information, and most authorized government officials
to scrutinize these filings for signs of fraud before granting permission
to sell. But this similarity on paper concealed some very large differ-
ences among states in practice.
Wisconsin, for example, had a typical second-╉generation blue sky law.
The law did not even apply to many classes of securities that were unlikely

32 Merrick v. N.W. Halsey & Co., 242, U.S. 568, 589, 586–╉87 (1917). The other
two cases were Hall v. Geiger Jones Co., 242 U.S. 539 (1917) (concerning Ohio’s
law), and Caldwell v. Sioux Falls Stock Yards Co., 242 U.S. 559 (1917) (concern-
ing South Dakota’s).
  155

Selling Blue Sky 155

to be out and out frauds—​government bonds, short-​term commercial


paper, securities of public utilities, securities listed on one of the major
stock exchanges, and securities of banks and trust companies. Nor did it
apply to the sale of securities by their owner, so long as the owner was not
in the business of selling securities. It placed only very mild restrictions
on the sale of stock in businesses that had operated at a profit for the
preceding two years. Authorized brokers could sell such stock without
requesting permission. Wisconsin’s law had real teeth only for securities
being sold by brand new companies or companies that were losing money.
These were the transactions with the greatest potential for fraud. These
companies had to file a detailed application with the Securities Division
of the Wisconsin Railroad Commission. In the application, a company
had to provide full information about the securities and how they would
be sold, as well as financial statements detailing the company’s income
and expenses over the past two years. If the company was a new one, the
application had to lay out its business plan in considerable detail. The
Securities Division then determined whether, in the words of the statute,
“the proposed plan of business … is not unlawful, unfair, unjust, or in-
equitable”; whether “the company intends to fairly and honestly transact
its business”; and whether the sale of securities at the proposed prices
“will work a fraud upon the purchaser thereof.” If nothing seemed amiss,
the Securities Division would approve the application.33
The Securities Division approved the large majority of applications. It
approved 651 in 1920–​21, for instance, while only 125 applications were
denied or withdrawn, and those figures grew even more lopsided in suc-
ceeding years.34 The division seems nevertheless to have been an aggressive
enforcer of the law, at least in those instances for which records survive.
In 1925, for example, the Securities Division investigated a number
of proposed bond issues intended to finance the construction of

33 Edwin L.  Schujan, “Wisconsin Blue Sky Legislation,” University Journal of


Business 1 (1923): 393–​405.
34 Annual Report of Securities Division July 1, 1920, to June 30, 1921
(Madison:  Railroad Commission of Wisconsin, 1921), 5; Annual Report of
Securities Division July 1, 1922, to June 30, 1923 (Madison: Railroad Commission
of Wisconsin, 1923), 3; Annual Report of Securities Division July 1, 1923, to June
30, 1924 (Madison: Railroad Commission of Wisconsin, 1924), 5.
156

156 Speculation

buildings. In one case, John Hunholz, the builder of two Milwaukee


apartment buildings, had inflated the building costs in his application,
in an effort to sell more bonds than the building was actually worth.
The division learned this by interviewing the contractors, who reported
receiving less than Hunholz claimed to be paying them. The division
denied Hunholz permission to sell the bonds and turned the case over
to the state attorney general, to prosecute Hunholz for filing false infor-
mation. In another case, the Civic Realty Company asked its contrac-
tors to sign contracts providing for payment greater than the amounts
called for by their bids, with the understanding that the contractors
would rebate the excess amount to the builder. The motive, again, was
to make the building appear more valuable so as to sell more bonds,
and once again the division discovered the fraud by interviewing the
contractors. There were many more such cases in Milwaukee at the
same time—​so many that for a while the Securities Division was func-
tioning as de facto auditor of construction costs, whenever construc-
tion was to be financed by a bond issue.35
The division also spent considerable time policing the sale of furry ani-
mals. In the 1920s, several promoters organized companies in Wisconsin
to sell rabbits, muskrats, and mink. The plan was to sell customers “units”
consisting of one male and two females, along with “ranching contracts”
under which the company would care for the animals and share with
the customers the profits from the sale of pelts. The division took the
position that this plan involved the sale of a security and thus required
compliance with the blue sky law. “We will not allow the sale of contracts
of this sort at any more than a fair market value for the animals,” the
division informed the proprietor of one such scheme, called the Hillside
Spring Fur Farm. “Your proposition, however, seems to us unfair on the
face of it,” because even on optimistic assumptions it “would leave the
investor a net return of $15 on an investment of $350. The investor could
do much better by investing that sum in a high grade bond.”36

35 Arthur Snapper to L. E. Gettle, 16 Dec. 1925, WDS, box 1, folder 1.


36 G. B. Brubaker to Securities Division, 21 Apr. 1928; Railroad Commission of
Wisconsin to William Speck, 31 Jan. 1929, both WDS, box 1, folder 2; H. W.
Harriman to file, 2 Sept. 1930, WDS, box 1, folder 4; Railroad Commission of
Wisconsin to Hillside Spring Fur Farm, 24 July 1930, WDS, box 1, folder 4.
  157

Selling Blue Sky 157

Much of the division’s work involved simply making sure that sell-
ers of securities complied with the rules. Greenbaum Sons, a Chicago
bond dealer, sold 182 different bond issues in Wisconsin before it both-
ered to apply for permission. The division informed Greenbaum Sons
that it would be barred from doing business in Wisconsin unless it
promptly qualified all those bonds. Nearly half of the issues had been
under $1,000, and for those the division allowed Greenbaum Sons the
alternative of repurchasing them from their holders for the sale price
plus accrued interest. But the division made clear that its authority
under the blue sky law concerned only the sale of securities, not the
ongoing conduct of corporations once the securities had been sold. “It
was not intended that stockholders should be freed from any necessity
of protecting their rights when this law was passed,” explained Garfield
Canright, the division’s director in the 1920s. “Our jurisdiction has to
do rather with the right to start the enterprise. After it is started the
stockholders must see that they elect proper directors and that the di-
rectors properly perform their duties.” So far as one can tell today, the
Wisconsin blue sky law seems to have succeeded in preventing at least
some of the more egregious forms of fraud in the sale of securities.
“It is, of course, impossible entirely to prevent violations of any law,”
Canright acknowledged. “We believe, however, that this situation is
gradually improving.”37
At the other end of the spectrum was Illinois, which had a blue
sky law that was similar to Wisconsin’s on paper but very different in
practice. “The Illinois blue sky law is practically null and void in its
operation,” admitted Walter Flint, who ran the state’s securities depart-
ment. “There is no appropriation for its enforcement.” Flint’s office
had allowed approximately seven hundred companies to sell securities
without conducting much of an investigation. “The trouble is that we
have no investigators,” Flint explained. “Our method is simply this. We
take the list of officers of the company applying for a license and ask
for their references. We then correspond with the persons referred to.

37 Railroad Commission of Wisconsin to Greenbaum Sons Investment Co., 19


Sept. 1924, WDS, box 1, folder 19; Garfield E. Canright to Lewis E. Gettle, 15
Jan. 1924, WDS, box 1, folder 7; Annual Report of the Securities Division July 1,
1920, to June 30, 1921, 8.
158

158 Speculation

The license is granted or withheld as a result of the correspondence.”


This method’s utter failure to prevent fraudulent stock sales came to
light in a 1919 lawsuit against the Consumers’ Packing Company filed
on behalf of shareholders who had bought the company’s worthless
stock. (The judge was Kenesaw Mountain Landis, who would leave the
bench a year later to become the first commissioner of baseball.) Flint
had granted a permit to the Consumers’ Packing Company because the
company’s officers and directors had adequate references, even though
Flint had reason to know that the company was a scam. One of the
references was a state politician who had been paid for his services. “We
have not a single investigator and the work of the office is handled by
a force of seven clerks,” Flint despaired. “There have been nearly 400
applications this year.”38
The Consumers’ Packing scandal prompted the Illinois legislature
to make appropriations for investigators, but the episode highlighted
how difficult it could be for a state agency to figure out ahead of time
which ventures were likely to be fraudulent. Daniel Holt, who has con-
ducted the closest study of the blue sky laws in operation, concludes
that many blue sky offices had only a handful of employees. Unable to
conduct thorough investigations of every proposed sale of securities,
they tended to adopt rules of thumb instead. They were extremely skep-
tical of mining companies, whose worth depended on the uncertain
value of what might be found underground. The Nebraska Bureau of
Securities, for instance, rejected 84 percent of the mining companies
that applied to sell stock, but only around 18 percent of the nonmining
companies. Aviation companies also met with high rejection rates. Blue
sky offices were likewise skeptical of corporate promoters who were not
sharing the risks with investors by putting their own money into the
business.39 Such determinations were not written into any law; they
were the last resort of state employees overwhelmed with the job they
had been given. Even the watered-​down second-​generation blue sky
laws, which tasked state agencies with ferreting out fraud rather than

3 8 Chicago Daily Tribune, 15 Mar. 1919, 1; 17 Mar. 1919, 5.


39 Daniel Stephen Holt, Acceptable Risk: Law, Regulation, and the Politics of
American Financial Markets, 1878–​1930 (Dissertation, University of Virginia,
Department of History, 2008), 173–​246.
  159

Selling Blue Sky 159

assessing the wisdom of an investment, were often impossible for the


states to enforce literally.
The young economist James Waterhouse Angell, writing in 1920, di-
agnosed two problems with this sort of haphazard enforcement. The
first was that blue sky bureaus had no real way to separate the fraudsters
from the people who genuinely wanted to raise money for new busi-
nesses that were speculative but honest. “If a fraudulent enterprise ap-
plies for qualification,” Angell pointed out, “the evidence upon which
its application is judged is, primarily, its own sworn statements, which
it can falsify at pleasure.” The second problem was that the blue sky
bureaus, aware of their own ineffectiveness but wanting to show the
public that they were enforcing the law vigorously, made the applica-
tion process cumbersome, so they would be able to reject a respectable
percentage of applications. As a result, Angell concluded, the blue sky
law “exactly inverts its supposed purpose; it burdens the class that is
honest with endless red tape, and guards it carefully from law-​break-
ing; while it calmly permits the more dubious classes to operate almost
without hindrance.”40
The investment bankers felt the same way. “The amount of good
accomplished by any or all of the so-​called ‘Blue Sky’ acts,” declared
the IBA’s committee on legislation in 1919, “is infinitesimal in com-
parison with the amount of expense and annoyance which they entail
upon legitimate business.” The laws were “worse than useless,” the com-
mittee complained, “in that reputable people do not need to be con-
trolled,” while the disreputable had been “granted licenses which they
have adroitly used to create the impression that what they were offering
had in some measure received the approval of the legally constituted
authorities.”41
Because of such concerns, some of the later blue sky laws entirely
gave up the effort to screen securities sales ahead of time. These laws
simply prohibited fraud, without requiring a permit before one could
sell securities. They were enacted primarily in the northeastern states,

40 James Waterhouse Angell, “The Illinois Blue Sky Law,” Journal of Political
Economy 28 (1920): 316, 314.
41 Proceedings of the Eighth Annual Convention of the Investment Bankers Association
of America (Chicago: Investment Bankers Association of America, 1919), 122.
160

160 Speculation

including the financial centers of New York and Pennsylvania, but also


in New Jersey, Maryland, Delaware, and Connecticut. Keyes Winter,
New York’s deputy attorney general, called his state’s law a “substitute
for a blue sky law,” because “it is not a license law.” Five thousand cor-
porations a month were organized in New York, and on some days three
million shares of stock were traded on the New York Stock Exchange.
“A license law restricting these transfers of securities in New York State
preliminary to a thorough investigation would choke business, not only
in the state, but in the entire United States,” Winter explained. Instead
of requiring a license, New York’s law gave the state attorney general
broad powers to subpoena witnesses and to compel the production of
corporate records, when there was reason to suspect fraud.42
Dissatisfaction with state blue sky laws gave rise to considerable sup-
port for a national blue sky law. The idea was appealing to an unlikely
coalition: on one side, proponents of regulation who hoped the fed-
eral government would be a stronger enforcer than the states could
ever hope to be, and on the other, business interests who preferred
a uniform national scheme of regulation over a hodgepodge of laws
that varied from state to state, both on paper and in practice. In the
1912 presidential election, the Progressive Party platform called for a
federal blue sky law. Theodore Roosevelt, the Progressive candidate,
praised the recently enacted Kansas law and urged it as a model for the
nation. Only a few months later, the Bankers’ Magazine also declared
its support for a federal blue sky law. “The sale of securities should, if
practicable, be brought under Federal regulation rather than be com-
mitted to the numerous States with their varying standards of legisla-
tion and administration,” the magazine insisted. “It would be compara-
tively simple to comply with a single Federal statute, but would prove
irksome to comply with the requirements of the laws of every State in
which it was desired to transact business.” The Bankers’ Magazine advo-
cated a federal blue sky law repeatedly over the next several years. Soon
the idea had received the support of virtually all the relevant regulatory

42 Keyes Winter, “State Regulation of Corporations by Policing Sales of Securities,”


Annals of the American Academy of Political and Social Science 129 (1927): 153, 152;
Keyes Winter, “Parasites of Finance,” North American Review 224 (1927): 520.
  161

Selling Blue Sky 161

and banking organizations. The National Association of Supervisors


of State Banks favored it. So did the Federal Trade Commission. The
Investment Bankers Association of America supported it too. “We need
a federal ‘blue sky’ law,” the IBA’s lawyer declared in 1920. “The solu-
tion of that problem lies with the federal government. Neither one nor
forty-​seven states can solve it and forty-​eight states never will without
federal compulsion.”43
Federal financial regulation was very much in the air at the time. In
the well-​publicized “Money Trust” hearings of 1912–​13, led by Louisiana
congressman Arsène Pujo, a House subcommittee heard extensive testi-
mony about the extent to which the nation’s financial system was domi-
nated by a small clique of bankers. The hearings were about banking
rather than the stock market (they took place just before the establish-
ment of the Federal Reserve System in 1913), but the underwriting of
securities was one of the things that bankers did, so part of the Money
Trust hearings involved the sale of speculative securities. The Pujo
Committee’s final report found that there was too much speculation
on the New York Stock Exchange and that powerful traders sometimes
manipulated prices by engaging in sham transactions. As a cure for
both problems, the committee recommended requiring corporations
listed on the stock exchange “to make a complete disclosure of their af-
fairs,” particularly any commissions they had paid to bankers out of the
proceeds of the sale of securities.44 But such matters were tangential to
the thrust of the Money Trust investigation, and Congress did not act
on these particular recommendations.
Once blue sky laws had been enacted in most states, and once the
world war was over, Congress took the issue up in earnest. In 1919, the

43 New York Times, 3 Sept. 1912, 2; Current Literature, Dec. 1912, 34; “Regulation
of the Sale of Securities,” Bankers’ Magazine 86 (1913): 418–​19; “ ‘Blue Sky’ Laws
Obscured by a Judicial Cloud,” Bankers’ Magazine 88 (1914): 283; “Federal Blue-​
Sky Legislation,” Bankers’ Magazine 104 (1922): 979; Los Angeles Times, 9 July
1914, I5; New York Times, 8 Oct. 1919, 29; Robert R. Reed, “ ‘Blue Sky’ Laws,”
Annals of the American Academy of Political and Social Science 88 (1920): 184, 183.
See also “Finance & Investment,” Current Opinion, 1 Nov. 1924, 652.
44 Money Trust Investigation: Investigation of Financial and Monetary Conditions
in the United States (Washington, D.C.: Government Printing Office, 1913),
part I, 6; H.R. Rep. No. 1593, 62nd Cong., 3rd Sess. (1913), 42–​54, 162–​63.
162

162 Speculation

House Judiciary Committee held hearings on a bill that was generally


referred to as a “federal blue sky law,” although its official name was
the Federal Stock Publicity Act. The bill, introduced by Representative
Edward Taylor of Colorado, required all corporations engaged in
interstate commerce to file financial statements with the Treasury
Department before selling stock. Unlike most of the state blue sky laws,
the bill did not empower government employees to screen proposed
sales for fraud ahead of time. Rather, the financial statements filed by
sellers of securities would be made available for public inspection (and
mailed to anyone willing to pay the postage). If the statements included
any false information, purchasers would be deemed to have relied on
the statements, and the seller would have to give them their money
back. “My earnest hope is to secure the enactment of some fair, sane,
and workable Federal ‘blue-​sky’ law,” Taylor declared. He had the sup-
port of nearly all the state blue sky commissions, who had difficulty
exercising authority over interstate transactions. But the IBA opposed
the bill, despite the organization’s professed support for a federal blue
sky law, because of the provision making the seller responsible for in-
correct information in the financial statements filed with the govern-
ment. “We have no objection to being made responsible for fraud or
any negligence,” explained Paul Keyser, testifying on behalf of the IBA.
“But we do not think it is fair to make us responsible for mere mistake
or accident where all good faith exists.”45 The bill never made it out of
the committee.
Congress considered several other federal blue sky bills over the next
several years, none of which was enacted. The idea that got the most at-
tention was embodied in bills repeatedly introduced by Representative
Edward Denison of Illinois, who proposed a federal ban on the inter-
state sale of securities whenever the sale would violate the blue sky law
of the state in which the purchaser was located. At the time, this was
a familiar way of supplementing the states’ limited power over inter-
state commerce. Before national Prohibition, Congress had enacted the
same kind of statute to help dry states keep out alcohol shipped from

45 Proposed Federal “Blue-​Sky” Law: Hearing Before the Committee on the Judiciary,


House of Representatives, 66th Cong., 1st Sess. (1919), 3–​10, 126.
  163

Selling Blue Sky 163

wet states. But Denison’s bills were consistently opposed by the IBA,
who wanted federal legislation precisely so that state officials would not
be able to exercise so much discretionary power over their business.
“It would be a ridiculous situation,” scoffed the New York investment
banker George Hodges, “for the Federal Government to delegate its
power to a blue-​sky commission in Kansas or wherever it might be.”46
By the late 1920s, despite the existence of blue sky laws in nearly
every state, and despite the interest of both bankers and regulators in
adopting a federal version, Congress’s efforts to enact a federal blue sky
law had foundered on disagreement over the details. Of course, the
stock market was booming. Speculators were making money. It was not
the best time to interest lawmakers or investors in imposing new limits
on speculation.
That would change.

46 Joel Seligman, The Transformation of Wall Street (Boston: Houghton Mifflin,


1982), 49; “Blue-​Sky” Bill:  Hearings Before the Committee on Interstate and
Foreign Commerce of the House of Representatives, 67th Cong., 2nd Sess. (1922),
53; Regulating Sale of Securities: Hearings Before a Subcommittee of the Committee
on Interstate Commerce, United States Senate, 67th Cong., 4th Sess. (1923).
5
Q
Aftershocks of the Crash

Between March 1925 and September 1929, the Dow Jones Industrial
Average rose from 115 to 381. By the middle of 1932, it had plummeted
to 41. The stocks listed on the New York Stock Exchange were worth
nearly $90 million in the fall of 1929, but by the summer of 1932, their
value was only $15.6 million.1 Never before in the history of the United
States had so many people lost so much money so quickly.
The stock market crash of 1929–​32 gave rise to several far-​reaching
reforms. The Securities Act of 1933 required sellers of securities to dis-
close a wide range of information. The Securities Exchange Act of 1934
imposed a variety of restrictions on the buying and selling of stock.2
The Glass-​Steagall Act (a part of the Banking Act of 1933) separated
commercial banks from investment banks. The two securities acts still
govern the stock market today; the Glass-​Steagall Act would continue
to structure the banking industry, despite gradual weakening, until it
was repealed in 1999.

1 Stock Exchange Practices: Report of the Committee on Banking and Currency


Pursuant to S. Res. 84 (Washington, D.C.: Government Printing Office, 1934), 7.
2 The best accounts of the drafting and enactment of the two securities acts
are Joel Seligman, The Transformation of Wall Street: A History of the Securities
and Exchange Commission and Modern Corporate Finance (Boston: Houghton
Mifflin, 1982), 1–​100, and Michael E. Parrish, Securities Regulation and the New
Deal (New Haven, Conn.: Yale University Press, 1970). On the Glass-​Steagall
Act, see Edwin J.  Perkins, “The Divorce of Commercial and Investment
Banking: A History,” Banking Law Journal 88 (1971): 483–​528.

164
╇ 165

Aftershocks of the Crash 165

If one takes a long-╉term perspective on the regulation of speculation,


the remarkable (and thus far underappreciated) thing about these New
Deal reforms is not what they did but what they didn’t do. Earlier market
downturns had yielded regulation that was intended to restrict specula-
tors in a much more serious way. After the crash of 1792, New York pro-
hibited stock futures. When the dollar fell against gold during the Civil
War, Congress banned gold futures. When agricultural prices fell in the
late nineteenth century, Congress came within a whisker of shutting
down the commodity exchanges, and some states imposed major limits
on commodity trading. Yet after the biggest crash of all, the regulatory
response was relatively mild. The crash was widely blamed on specula-
tors, and there were many calls for placing serious limits on speculation
to prevent such a thing from happening again, but the reforms that fol-
lowed the crash curtailed speculation only very modestly. Why?

A Mammoth Gambling Machine


As the stock market reached new heights in the late 1920s, many worried
that the rise was the product of speculation rather than genuine prosper-
ity. “Rarely, if ever, has the stock market appeared to get so far out of
touch with business conditions as it has in the last year or two,” the econ-
omist Lewis Haney cautioned in early 1928. “Stocks have risen in value
much more than twice as fast as our industrial activity.” “It is ridiculous
to term purchases of stocks at present prices investments,” lectured the
banker Alan Temple. “They are speculations.” But such warnings could
hardly deter investors from jumping into the market after they had seen
their neighbors grow rich. “Increasing thousands of first-╉time speculators
are watching their paper profits mount, and are concluding that anyone
who works for a living is a boob,” Time magazine reported. It became a
commonplace in the press that the nation was in the midst of a specula-
tive craze that, like similar episodes in the past, was sure to end badly.
Carter Glass, the Virginia senator and former Treasury secretary, reported
in early 1929 that “there is a firm determination in Congress to enact
some legislation to abate this infernal menace of stock gambling.”3

3 Lewis H.  Haney, “The Stock Market and Business,” North American Review
225 (1928):  175; Alan H.  Temple, “The Financial Outlook,” North American
166

166 Speculation

Others were more optimistic. “Agitation against speculation


in Wall Street is very largely a case of sour grapes,” the Wall Street
Journal insisted. “It is felt that some people are making money with
apparent ease and it is known that they are making it in Wall Street,
which is always an object of distrust to the demagogue.” One booster
concluded that stock prices were entirely justified in light of all the
spectacular technological changes of the 1920s. “Indeed,” he imag-
ined, “who shall say what the future profits of such new industries
as radio, television, electric power, chain store methods of merchan-
dising, and so on will be?” In 1929, J. P. Morgan’s partner Thomas
Lamont assured the new president Herbert Hoover that even if
stocks were overvalued they were only modestly so, and that the bull
market was attributable instead to two positive developments. One
was the new willingness of ordinary Americans to invest in stocks,
which attracted capital that had formerly lain dormant in small towns
throughout the country. The other cause was the genuine underlying
prosperity of the United States. (“This document is fairly amazing in
light of two years after,” Hoover later scrawled across the top.) The
investment advisor H. W. Moorhouse concluded that advanced stock
prices were the product of “a new economic era” in which constantly
growing demand for stock would yield constantly rising prices. If
prices seemed higher than economic principles warranted, perhaps
it was the principles that needed adjusting. In Moorhouse’s view, the
market was “vaulting easily over economic precepts formerly con-
sidered impregnable.” As the Yale economist Irving Fisher famously
put it in mid-​October 1929, just a week before the market began to
plummet, stock prices had reached “what looks like a permanently
high plateau.”4

Review 226 (1928): 641; “Stock Market Jamboree,” Time, 28 May 1928, 32; A. M.
Sakolski, “American Speculative Manias, Past and Present,” Current History 30
(1929): 861; Carter Glass to G. H. Gray, 11 Feb. 1929, CG, box 311.
4 Wall Street Journal, 15 May 1928, 1; Donald Rea Hanson, “In Defense of
Speculation,” Forum 81 (1929):  57; Thomas W.  Lamont to Herbert Hoover,
19 Oct. 1929, HHL, box 1049; H.  W. Moorhouse, “What’s Happening in
Wall Street,” North American Review 226 (1928): 678; New York Times, 16 Oct.
1929, 8.
  167

Aftershocks of the Crash 167

Some were worried that stock speculation had spiraled out of con-
trol, others thought it entirely justified, and there was a third position
in the middle, that even if there was too much speculation, nothing
constructive could be done about it. George Roberts, the vice presi-
dent of the National City Bank of New York, was willing to concede
that “there is a vast amount of uninformed speculation” that served no
economic purpose. But “it is impossible to require everybody to pass a
civil service examination before engaging in any kind of speculation,”
Roberts observed, “and it is exceedingly difficult if not impossible to
discriminate at least in law, between acts of the same kind which in
one case represent perfectly legitimate operations and in another case
may be of the undesirable class.” Legislation curbing speculation might
be a cure worse than the disease, because it risked driving off the good
transactions, as well as the bad. Without speculation, “the conduct of
railways and large-​scale industry would probably be found a hard job,”
reckoned the Bankers’ Magazine. “To carry on these enterprises requires
a lot of money, which cannot be had from those who are investors of
the purely conservative type.”5
Previous booms had seen similar debates about the dangers of specu-
lation, but the debate in the 1920s was more intense and more con-
sequential than in earlier episodes, because the establishment of the
Federal Reserve System in 1913 gave the government a new policy lever.
By raising interest rates, the Federal Reserve could dampen speculation,
by making it more difficult for speculators to earn a positive return
with borrowed money. Throughout the second half of the decade, crit-
ics repeatedly called upon the Federal Reserve to do just that. “I have
noted with growing anxiety the steady increase in speculation in the
country, more particularly centered in the New York Stock Market,
which has for some time been engaged in positive gambling in secu-
rities,” Wisconsin senator Irvine Lenroot wrote in 1925, in an urgent
plea to Federal Reserve Board governor Daniel Crissinger to raise in-
terest rates. “Does not the present cheap money induced by the low
discount rates indirectly stimulate speculation on the New York Stock

5 George E.  Roberts, “Speculation and Its Influence Upon Business,” Bankers’
Magazine 117 (1928): 967; “Fresh Outburst of Speculation,” Bankers’ Magazine
116 (1928): 625.
168

168 Speculation

Exchange[?]‌”6 The Bankers’ Magazine recalled that in earlier specula-


tive booms, the rise in demand for money to invest would cause inter-
est rates to rise sharply, which would check speculation. The Federal
Reserve System had the beneficial effect of preventing disabling short-
ages of money, the magazine concluded, but at the cost of removing
this former limit on speculation.7
In the Senate, the Michigan Republican James Couzens com-
plained in the summer of 1929 that “the Federal Reserve Board has
been dumb in not dealing with a situation that should have been
dealt with months ago.” If the board had only raised interest rates,
he charged, “this great orgy of speculation would not have occurred.”
The Wisconsin Republican Robert La Follette proposed barring the
Federal Reserve from lending any funds that would be put to specula-
tive purposes. And these were the Senate’s more temperate and urbane
critics of the Federal Reserve’s role regarding speculation. When the
topic was debated in the Senate, the result was, as Time magazine put
it, “a mingled outburst of oratory, ethics, provincialism and a little eco-
nomics.” The Alabama Democrat James Heflin thundered that Wall
Street was a gambling monster that was destroying American homes.
Smith Brookhart of Iowa suggested prohibiting all banks from lend-
ing to speculators. William King of Utah wanted to ban margin trad-
ing. Thaddeus Caraway of Arkansas suggested prison terms for dealing
in grain and cotton futures.8 If the Federal Reserve could cause the
amount of speculation to come down from a dangerous level simply
by raising interest rates, critics wondered, why would it hesitate?
The question was hotly debated both within and outside the Federal
Reserve. The difficulty was that the interest rate was a blunt instrument.
Raising interest rates would make borrowing more costly for everyone,
not just for speculators, so it would dampen all business activity, not

6 Irvine Lenroot to D.  R. Crissinger, 23 Nov. 1925, HPPF, box 122. Herbert
Hoover claimed in his memoirs to have written this letter for Lenroot to
sign, while Hoover was secretary of commerce. Herbert Hoover, Memoirs
(New York: Macmillan, 1951–​52), 3:9–​11.
7 “Federal Reserve System and Speculation,” Bankers’ Magazine 115 (1927): 129.
8 Washington Post, 6 June 1929, 6; New  York Times, 19 Jan. 1928, 10; “Federal
Reserve v. Speculation,” Time, 25 Feb. 1929, 49.
  169

Aftershocks of the Crash 169

just speculation. Was it worth it? “The policy of raising the rediscount
rate in order to stop speculation is … based on the theory that the mil-
lions of innocents should be punished for the sins of the few,” argued
the banker Edward C. Stokes, the former governor of New Jersey. “If a
young inexperienced patron of my bank is branching out too much in
the speculative field, I quietly call him to my desk and tell him that he
has reached his limit, but I don’t raise the rate of interest on every other
borrower of our institution.” The banker Russell Leffingwell cautioned
that it would be impossible for the Federal Reserve to distinguish bor-
rowers seeking to speculate from borrowers seeking to use money for
ends that seemed more productive. “We are all, good and bad, sheep
and goats, farmers and bankers and merchants and speculators and in-
dustrialists, engaged in one great joint venture,” Leffingwell mused. “If
the Federal Reserve authorities set out to ruin some, by denying credit
to those they don’t approve of, the speculators, they will most likely
injure us all.”9
Higher interest rates, moreover, would slow or even reverse the re-
markable rise in stock prices, a result that was sure to be unpopular.
Before taking such action, the Federal Reserve had to be confident that
curtailing speculation justified harming so many people, but how could
officials, or indeed anyone, be sure that prices were already too high?
And as many suggested, any move by the government, even by the
quasi-​independent Federal Reserve, that was perceived to have ended
the boom was likely to have disastrous consequences for the incumbent
Republican Party. As one editorialist noted of Treasury Secretary and
Federal Reserve Board member Andrew Mellon shortly after the 1928
presidential election, “it would be too much to expect of any incum-
bent of that distinguished office to take otherwise than a roseate view of
the situation in the face of a national political campaign.”10

9 Allan H. Meltzer, A History of the Federal Reserve: Volume I, 1913–1951 (Chicago:


University of Chicago Press, 2003), 224–​57; William J. Barber, From New Era
to New Deal: Herbert Hoover, the Economists, and American Economic Policy,
1921–1933 (Cambridge: Cambridge University Press, 1985), 71–​ 77; E. C.
Stokes, “Control Speculation? No!,” Bankers’ Magazine 118 (1929): 714; Russell
Leffingwell to Carter Glass, 29 Apr. 1919, CG, box 283.
10 Wall Street Journal, 24 Apr. 1928, 1; “Federal Reserve Policy and Speculation,”
Bankers’ Magazine 118 (1929): 541.
170

170 Speculation

As it turned out, the Federal Reserve did raise the discount rate three
times in 1928, from 3.5 percent to 4 percent, then to 4.5 percent, and
finally to 5  percent. These measures had no discernible effect on the
amount of speculation or on stock prices. In the spring of 1929, as
the market continued to rise, Hoover and Mellon urged the Federal
Reserve Board to raise the rate yet again. The Federal Reserve Bank of
New York asked the board several times to increase the rate to 6 per-
cent. Each time, a majority of the board declined to do so.11
“The stock boom was blowing great guns when I  came into the
White House,” Hoover recalled in his memoirs. “Being fully alive to
the danger inherent in this South Sea Bubble and its inevitable reaction,
my first interest was to get it under restraint.” Hoover’s memory was
colored by hindsight. Before the crash he had told the banker Martin
Egan that he was “dubious about doing anything for fear that more
harm than good might result.” And of course in his later years Hoover
had an interest in defending himself against charges that as president
he had done too little—​he pointedly titled this chapter of his memoirs
“We Attempt to Stop the Orgy of Speculation.” Even so, Hoover was
no doubt right that there was not much even a president could do
apart from badgering the Federal Reserve to lower interest rates. “To
ask Congress for powers to interfere in the stock market was futile,”
he recalled, “and, in any event, for the President to dictate the price of
stocks was an expansion of Presidential power without any established
constitutional basis.” So Hoover did what he could to talk the market
down. He asked the publishers of the major newspapers and magazines
to run editorials warning readers of the dangers of speculation. He gave
speeches to the same effect. He urged bankers to exercise restraint in
lending, and he urged Richard Whitney, the president of the New York
Stock Exchange, to do something to curb the manipulation of stock
prices. None of these actions had any effect. In August 1929, the Federal
Reserve Board finally did raise the discount rate to 6 percent. It made
no difference. “The real trouble,” Hoover lamented, “was that the bell-
boys, the waiters, and the host of unknowing people, as well as the

11 David Cannadine, Mellon:  An American Life (New  York:  Alfred A.  Knopf,
2006), 359, 388–​89.
  171

Aftershocks of the Crash 171

financial community, had become so obsessed with the constant press


reports of great winnings that the movement was uncontrollable.”12
When the crash finally came, speculators took much of the blame.
Alexander Noyes, the financial editor of the New York Times, castigated
“professional speculators of large private means but utter recklessness,”
as well as amateur “outside speculators, large and small, most of whom
were totally ignorant of intrinsic values and inspired by nothing but the
gambling purpose of bidding on a market for an expected further rise.”
The financial journalist John T.  Flynn complained that buying stock
had become just like betting on racehorses. “Trading in stocks and
bonds should be limited to buying and selling between persons who
use stocks and bonds as media of investment,” Flynn insisted. “Their
use as counters in a gigantic game of gambling ought to be brought to
an end.” The North American Review scoffed at those who, before the
crash, “incessantly preached that a new era had arrived,” when the sup-
posed new era turned out to be just another speculative bubble. Even
the Wall Street Journal, which had defended investing in stocks on the
way up, acknowledged after the crash that “many people all over the
country have learned the essential difference between investment and
speculation.”13
In late 1929 and the early 1930s, as the market slid ever lower and the
Depression deepened, there was considerable public demand for some
form of retribution against “Wall Street,” the shadowy force that had
plunged the nation into distress. “Wall Street speculation has ruined
hundreds of thousands of people in the United States,” complained one
correspondent to President Herbert Hoover. Another charged that “this
‘Stock Market’ is nothing more or less than a Mammoth Gambling
Machine, out to rob any and every one.” The mayor of Mobile, Alabama,
advised Hoover that “the most important work you have to do in the

12 Martin Egan to Thomas Lamont, 23 Oct. 1929, HHL, box 1049; Hoover,
Memoirs, 3:16–​19.
13 Alexander D.  Noyes, “The Stock Market Panic,” Current History 31
(1929): 619; John T. Flynn, “Speculation and Gambling,” Harper’s Magazine
160 (1930):  201; John T.  Flynn, “The Wall Street Debt Machine,” Harper’s
Magazine 167 (1933): 138; Max Winkler, “Paying the Piper,” North American
Review 229 (1930): 50; Wall Street Journal, 5 Nov. 1929, 1.
172

172 Speculation

interest of civilization is to abolish the gambling features in our stock


exchanges,” while a Minneapolis bond salesman shared his view that
the New York Stock Exchange “is like a boil or fester, eating away at
the vitality of the economic life of our country.” The Ohio representa-
tive Charles Truax expressed this sentiment on the House floor, when
he declared that “there are a lot of financial buzzards down on Wall
Street who ought to be shot.” Truax favored any legislation that “will do
something to the bloodiest band of racketeers and vampires that ever
sucked the blood of humanity, John ‘Pirate’ Morgan & Co.” He urged
the government to imprison “the bloody butchers of Wall Street who
have lived in high style on the blood of the common people.”14
Others contended that much of the blame belonged to the govern-
ment, for failing to put a stop to the speculative excess while there was
still a chance. “Who caused the Panic of 1929?” asked the economist
H. Parker Willis in early 1930. It was the Federal Reserve Board, he
contended, for keeping interest rates artificially low. “We had the con-
trol in 1925, in 1926, in 1927, in 1928, and in 1929, and we did not use
it,” lamented Percy Johnson, the chairman of the board of Chemical
Bank. “We had the control in the power of the Federal Reserve Board,
and if the board had not bent its policy to meet the wishes of the
Treasury, which wanted to borrow cheap money for the government,
but had raised these rates, as it should have done, we could have
stopped this speculation in its incipiency in 1927 and 1928 and avoided
this colossal crash that has come with such widespread disaster to every
one.” The columnist Walter Lippmann faulted Hoover for having been
too weak to put pressure on the Federal Reserve. The Federal Reserve
was even accused of having deliberately prolonged the boom for politi-
cal reasons, to ensure Hoover’s victory in the election of 1928.15

14 Washington Post, 27 Oct. 1930, 6; M. Rosenbaum to Herbert Hoover, 13 Nov.


1929, HHL, box 159, folder 5; L. I. Conrad to Herbert Hoover, 14 Nov. 1929,
HHL, box 159, folder 5; Harry T. Hartwell to Herbert Hoover, 17 Feb. 1930,
HHL, box 159, folder 6; Francis H. Gill to Walter H. Newton, 10 Apr. 1932,
HHL, box 159A, folder 2; Congressional Record 78 (1934): 7939, 7941.
15 H. Parker Willis, “Who Caused the Panic of 1929?,” North American Review 229
(1930): 174; New York Times, 18 Mar. 1934, E4; Walter Lippmann, “The Peculiar
╇ 173

Aftershocks of the Crash 173

Whether one faulted speculators or the government that had failed


to control them, it was clear that something had to be done to prevent
such a calamity from happening again. But what? “It was easy, after the
Market had broken, to denounce speculators as fools and speculation as
vicious,” Time magazine observed. But “where could the line be drawn
between farsightedness and folly?”16 How could the country protect
itself against the excesses of speculation without killing off investment?

The Ruin of€Thousands


Previous market downturns had yielded new regulatory proposals aimed
at curbing speculation, so it was no surprise that the stock market crash
of 1929–╉32 brought forth a wide range of calls to regulate the market.
“If there ever was a time when a bill restricting speculation would be ac-
claimed with favor, it is at the present time,” recognized the Cincinnati
broker J.  R. Edwards. At the Investment Bankers Association’s 1932
convention, the IBA’s Legislation Committee reported that it could
barely keep up with all the bills that had been introduced in Congress.17
Some of these proposals were quite severe. Many people urged
President Hoover to shut down the nation’s stock exchanges until
conditions had stabilized. The veteran Kansas senator Arthur Capper
informed Hoover that “this feeling is wide spread.” The Colorado
representative John Martin thought that if the question were put to
a referendum, most Americans would even vote in favor of abolishing
the stock exchanges permanently. “The exchanges are lucky,” Martin
remarked, “that Congress and not the country is regulating them.”
Others, including the Kentucky legislator Herman Handmaker, sug-
gested fixing stock prices so they could not fall. Henry Heimann, the
executive manager of the National Association of Credit Men, reported
“a very pronounced feeling” among the association’s members that

Weakness of Mr. Hoover,” Harper’s Magazine 161 (1930): 1; J. M. Daiger, “Did


the Federal Reserve Play Politics?,” Current History 37 (1932): 25.
16 “Market ‘Lesson,’â•›” Time, 18 Nov. 1929, 51.
1 7 J.  R. Edwards to Carter Glass, 9 Jan. 1932, CG, box 301; Proceedings of the
Twenty-╉
First Annual Convention of the Investment Bankers Association of
America (Chicago: Investment Bankers Association of America, 1932), 163.
174

174 Speculation

stock prices should be pegged at their values as of some date in the past
and then not permitted to decline. The Boston banker Allan Forbes had
the same idea, but his friend J. P. Morgan politely talked him out of it.
If “people did not like the prices when they were fixed,” Morgan told
Forbes, “it would simply mean they would go and set up an outside
market” and trade there instead of on the exchange.18 It is a good indi-
cator of the magnitude of the crisis that such desperate measures could
have been proposed by such knowledgeable people.
Previous downturns had given rise to proposals to ban short selling,
the practice of agreeing to sell in the future an item one did not yet own
to profit from a decline in its price. There were many more such calls
in the early 1930s. Letters poured into the White House decrying “the
destructive practice of short selling,” as one New Jersey entrepreneur
put it. An engineering professor at the University of Michigan insisted
in 1930 that “if there had not been avalanche after avalanche of short
selling last fall we would have had an orderly recession of only about
half the historic panic drop.” As always, something seemed unsavory
about a practice that made someone desire—​and possibly even act to
bring about—​a crash. “A man who sells and destroys the value of your
securities is no better than the man who burns down your house,” in-
sisted one of Herbert Hoover’s many correspondents. One of Arthur
Capper’s constituents urged him to “tell the Banking Committee that
if they have any doubt that short selling is the principal cause of the
depression to put a ban on short sales for ninety days and watch the
general market advance fifty percent.” It seemed unfair that short sell-
ers should profit from the despair of others. “How is it that a practice
so contrary to public policy can be permitted in this country?” asked
the oilman Joseph Pointer. “In whose possession are the twenty bil-
lions that disappeared from the pockets of the carpenter, the actor, the

18 George W.  McPherson to Herbert Hoover, 29 Oct. 1929, HHL, box 159,
folder 4; M. B. Friend to Lawrence Richey, 1 June 1932, HHL, box 159A, folder
3; Arthur Capper to Hebert Hoover, 27 Apr. 1932, HHL, box 159A, folder
2; Congressional Record 78 (1934): 8016; Herman H. Handmaker to Herbert
Hoover, 13 Nov. 1929, HHL, box 159, folder 5; Henry H. Heimann to Herbert
Hoover, 26 May 1932, HHL, box 159A, folder 3; J.  P. Morgan Jr. to Allan
Forbes, 19 Apr. 1932, JPM, box 25, letterpress book 43.
  175

Aftershocks of the Crash 175

small executive, and the chorus girl?” Short sellers were “people who
put profits before patriotism,” reasoned William Buck, a Philadelphia
salesman—​or else they were “foreigners who wish no good to the
United States.” The Kansas Republican Charles Scott advised Hoover
that “if you can succeed in suppressing short selling in stocks and bonds
and food products it will win you the vote of every Republican farmer
in the country.” Democrats too, suggested the humorist Will Rogers. If
the practice were banned, he predicted, “at least 115,000,000 out of the
120,000,000 would put on a celebration that would make Armistice
Day look like a wake.”19
Hoover himself was dubious about short selling. “Men are not justi-
fied in deliberately making a profit from the losses of other people,” he
wrote in early 1932. Hoover also shared the widespread suspicion that
short sellers manipulated prices by “pounding down” the market. He
worried that “these operations destroy public confidence and induce
a slowing down of business.” The Hoover administration accordingly
took some modest steps to limit short selling. George Akerson, Hoover’s
press secretary, prepared a list of prominent New York “bears” who “or-
ganize short selling at certain hours of the day, thereby creating a real
strain on the market.” Hoover met with the leaders of the New York
Stock Exchange and the Chicago Board of Trade, to urge them to adopt
rules to curb short selling. He told Richard Whitney, the president of
the stock exchange, that “whatever defense there may be for the prac-
tice of short selling during ordinary and normal times,” it was “most
injurious during periods of depression like the present.” Hoover issued
Whitney an ambiguous threat, that “unless the Stock Exchange itself
would promptly take steps to stop such injurious practices he would feel
it his duty to proceed along other lines.” At Hoover’s request, in January

19 Ansley H. Fox to Herbert Hoover, 17 Oct. 1930, HHL, box 159A, folder 7;
William S. Hazelton to Herbert Hoover, 16 Oct. 1930, HHL, box 159A, folder
7; E. C. Stokes to Herbert Hoover, 27 Sept. 1930, HHL, box 159A, folder 7;
H. L. Stout to Arthur Capper, 23 Apr. 1930, HHL, box 159A, folder 2; Joseph
Pointer to Herbert Hoover, 20 Nov. 1929, HHL, box 159A, folder 6; W. L.
Buck to Louis McHenry Howe, 23 Mar. 1933, FDR, POF 34, box 1; Charles
F. Scott to Herbert Hoover, 24 Feb. 1932, HHL, box 159A, folder 5; Rogers
quoted in Nathan T. Porter, “Business and Finance,” Overland Monthly and
Out West Magazine 90 (1932): 41.
176

176 Speculation

1932, the stock exchange began sending reports of short-​sale statistics to


the White House, at first daily and then weekly, a regimen that lasted for
the remainder of Hoover’s term in office.20
There was also substantial sentiment in Congress against short sell-
ing. “In the fall of 1929 I started a crusade against short selling,” the
Chicago congressman Adolph Sabath declared. “I have done everything
humanly possible to have the stock exchanges stop short selling.” His
Idaho colleague Compton White agreed that “businesses have been
ruined and various enterprises have been destroyed by piratical attacks”
by short sellers. “We know that something should be done to curb this
practice.” In early 1932, the House Judiciary Committee held four days
of hearings on several bills that would have curbed short selling or even
completely abolished it. The Senate issued subpoenas to nearly thirty
brokerage firms “to find out whether a large short interest in the market
does artificially depress the value of the securities,” as Senator Frederic
Walcott of Connecticut put it.21
Short selling had been defended against such attacks for more than a
century, on the grounds that short selling does not in fact cause prices
to decline (because there has to be a purchaser on the other side of every
short sale, and because every short seller has to become a buyer at some
later date), and indeed that short sellers benefit markets by steering them
away from the highest peaks and the lowest valleys. But these arguments
were not at their most plausible in the early 1930s, when the preceding
few years had seen the highest peaks and the lowest valleys in living
memory, and when hoping for lower prices seemed like cheering for
disease. Short selling was “losing caste,” the business writer J. George
Frederick observed; “the classic economic defenses for the practice are
all down.” Louis Guenther, publisher of the magazine Financial World,

20 Herbert Hoover to Thomas W.  Lamont, 2 Apr. 1932, HHL, box 1049;
“Memorandum,” 3 Oct. 1930, GEA, box 19; Silas H.  Strawn to Lawrence
Richey, 28 Oct. 1931, HHL, box 851; B. H. Meyer to Theodore Joslin, 19 Mar.
1932, HHL, box 159A, folder 10; short-​sale reports in HHL, box 159A, folder 13.
21 Congressional Record 78 (1934): 8028, 8027; Short Selling of Securities: Hearing
Before the Committee on the Judiciary, House of Representatives (Washington,
D.C.: Government Printing Office, 1932); Frederic C. Walcott to Walter H.
Newton, 20 Apr. 1932, HHL, box 159A, folder 2.
  177

Aftershocks of the Crash 177

perceived that “short selling is coming into bad repute with the public,”
and he thought it a good development in a depression, because “it is in
just such a time that everyone should endeavor to employ his abilities
and resources towards the maintenance of public confidence.”22
The leaders of the New York financial community nevertheless made
the case for short selling, both in private and in public. “We are deathly
afraid of Hoover,” the New York broker Charles Chambers told Carter
Glass. “The big men in the street do not know just what piece of eco-
nomic quackery he is going to propose next. It is plain that he is a
desperate man and is snatching at every little straw in order to attempt
to stave off the bad beating that is surely coming to him in the next
election.” Chambers and his colleagues were particularly worried about
a ban on short selling. “Hoover and the group of petty men who are
running the government will stoop to anything, even to ruining an
organized security market,” Chambers despaired. “For centuries we
have been striving for liquidity and now certain members of Congress
propose to ruin the liquidity of securities by abolishing short selling.
Senator, it would be a catastrophe.” Thomas Lamont assured Hoover
that the fall in the stock market was caused not by short sellers but by
declining company earnings, and that banning short selling would only
impede the recovery, because “bulls will not buy in a market where
bears cannot sell, for today’s bull wants to be able to change his foot and
sell tomorrow.” New York Stock Exchange president Richard Whitney
defended short selling in speeches all over the country, some of which
were broadcast live on the radio and printed as pamphlets for distri-
bution. His 1931 speech before the Hartford Chamber of Commerce,
entitled “Short Selling,” was heard on sixty radio stations in over forty
states and had a print run of more than 750,000 copies.23

22 J.  George Frederick, “Short Selling Loses Caste,” North American Review
234 (1932): 58; Louis Guenther, “Vicious Short Selling,” Financial World, 22
Oct. 1930.
23 Charles W. Chambers to Carter Glass, 14 Dec. 1931, CG, box 301; Thomas
W. Lamont to Herbert Hoover, 1 Apr. 1932, HHL, box 1049; Richard Whitney,
“Short Selling” (16 Oct. 1931), HHL, box 160, folder 8; Richard Whitney to
“All Members and Partners of Stock Exchange Firms,” 4 Nov. 1931, HHL, box
159A, folder 12. Other examples of Whitney speeches defending short selling
include “Speculation” (10 Oct. 1930), “Short Selling and Liquidation” (15 Dec.
178

178 Speculation

The major exchanges also took small steps of their own to limit short
selling, in the hope of fending off more serious legislative incursions.
In 1931, the New York Stock Exchange asked its members not to accept
orders for short sales except where customers were hedging against losses
from securities they already owned. The following year the stock ex-
change prohibited members from lending customers’ securities to other
brokers (short sellers borrowed securities before selling them) without
first obtaining the customers’ written authorization. These measures
were more cosmetic than substantive; the former carried no sanctions
for violators, and the latter was easy to comply with. The New  York
lawyer W. R. Perkins shared what seemed to be the unanimous view of
the local press that short selling would proceed unabated. The Chicago
Board of Trade prohibited what the board called “harmful speculative
short selling.” This was likely just as cosmetic, as it was clear that the
board of trade did not share the Hoover administration’s view as to
which kinds of transactions were harmful.24 Despite such efforts, there
was a real possibility in the years after the crash that the government
would do something to curb short selling, certainly a greater possibility
than at any time since the agrarian protests against commodity trading
in the 1890s.
Another common regulatory proposal was a ban on margin trad-
ing. Stockbrokers and commodity brokers normally did not require
their customers to pay the full price of whatever the brokers bought
on their behalf. The customer merely had to deposit a “margin,” or
a percentage of the purchase price, and the broker would lend him
the rest. If the customer’s losses reached the margin, the broker would
call upon the customer for more money, and if the customer did not
deposit more, the broker would close out the customer’s account. It

1931), and “The New York Stock Exchange” (27 Dec. 1932), all in HHL, box
160, folder 8.
24 A. S. Brown to Herbert Hoover, 25 Sept. 1931, HHL, box 159A, folder 8;
New York Stock Exchange Circular, 18 Feb. 1932, HHL, box 159A, folder
10; W.  R. Perkins to Herbert Hoover, 20 Feb. 1932, HHL, box 159A,
folder 9; Chicago Board of Trade Circular, 26 Oct. 1932, HHL, box 159A,
folder 5; Walter H. Newton to Arthur M. Hyde, 20 Oct. 1932, HHL, box
159A, folder 5; Chicago Board of Trade to Herbert Hoover, 2 Nov. 1931,
HHL, box 159A, folder 4.
  179

Aftershocks of the Crash 179

was widely understood that margin trading—​trading with borrowed


money—​had caused prices to fall more steeply than they otherwise
would have. When prices started to decline, investors had to sell some
of their stocks to keep up their accounts in their other stocks, and this
sell-​off caused prices to decline even more, in a vicious circle. The bro-
kers, in turn, often borrowed money from banks to cover their loans to
customers, with the loans secured by the stock. When the value of this
collateral dropped, banks would call in the loans, which likewise caused
even more selling as brokers all scrambled for money at the same time.
When the market fell by even a small amount, everyone had to sell
simultaneously, which could turn a small drop into a big one. “The real
major cause of the collapse,” explained Michael Cahill, president of the
Plaza Trust Company, “was not merely that securities were purchased
at levels far beyond their sound investment values, … but it was this
situation added to the fact that such purchases were made on borrowed
capital.”25
Many accordingly suggested that the real danger lay not in specula-
tion but in speculation on margin. Arthur Capper, the Kansas sena-
tor, wanted to prohibit it outright. So did Adolf Berle, the Columbia
law professor who advised Franklin Roosevelt on financial regulation
during and after Roosevelt’s 1932 presidential campaign. “Margin trad-
ing ought not to be permitted,” Berle declared. “There is no real reason
why credit should be involved in the stock market to the extent it is
now.” The problem was that “the free use of credit encourages the wild
fluctuations.” If speculators had to use their own money rather than
borrowing it from their brokers, their losses might cause harm to them-
selves, but their losses would not set off a chain reaction of defaults and
forced sales that harmed the broader public. “Gambling, as gambling,
we shall never eradicate by any process of law,” Berle concluded. It
would be better instead, he advised, to “restrict the class of gamblers to
a group which can afford to lose, and the total volume of gambling to a
point where it will not involve the whole financial structure.”26

2 5 Michael B. Cahill to Herbert Hoover, 3 May 1930, HHL, box 159, folder 6.
26 Arthur Capper to Herbert Hoover, 26 Jan. 1932, HHL, box 159A, folder 9;
“Memorandum to the Committee on Stock Exchange Regulation” (24 Oct.
1933), AAB, box 22.
180

180 Speculation

In the floor debates over the bills that would become the legislation
of 1933 and 1934, several members of Congress urged the prohibition
of margin trading. “Security speculation is practically the only gam-
bling game which can be carried on on credit,” argued Senator Robert
Bulkley of Ohio. “You cannot go to a race track and make a bet on
a horse race unless you have your money in your hand. There is no
money desk in the betting ring where loans can be negotiated for bet-
ting purposes. And there ought not to be a money desk on the floor
of the stock exchange.” Margin trading, Bulkley declared, “is funda-
mentally wrong.” Duncan Fletcher, who chaired the Senate Banking
Committee, believed that “margin trading in this country has led to a
great many abuses and to excessive speculation,” because it allowed too
many people to get in over their heads. “When a person is permitted
to put up a few dollars on margin to buy stocks and take chances of
winning or losing, he is tempted to do so,” Fletcher worried. “Margin
operation, in my judgment, is responsible for the ruin of thousands
and thousands of people.” In the House, Martin Smith of Washington
State described margin trading as “evil,” because it “acts as a snare or
a trap, and a great many people buy more stocks than they can afford
to buy, and when the call for additional margin comes their accounts
are wiped out. Undoubtedly more money was lost by the American
public because of marginal sales in the stock collapse of 1929 than
on account of insufficiency of value in the securities when they were
originally issued.” This heartfelt statement moved Smith’s colleagues
to applause.27
Another proposal aimed at the same target was to ban banks from
lending to brokers and speculators. The chain of credit from brokers
to their customers had banks at its origin, so if the funds could be cut
off at their source, speculators would be unable to buy stock with bor-
rowed money. Behind the banks lay the Federal Reserve, which meant
that the government was in effect the ultimate source of the loans. The
Senate Banking Committee held hearings on the subject as early as
1928, when the market was still on the way up. “I do not object to a

27 Congressional Record 78 (1934):  8387, 8386, 8175; Congressional Record 77


(1933): 2941.
  181

Aftershocks of the Crash 181

man throwing away his money,” declared Earle Mayfield of Texas, “but
what I object to is the Government lending about $3,000,000,000 to
help him do it.” A few months before the crash, William Henry King of
Utah again proposed an investigation into the propriety of bank loans
to brokers and speculators. The idea was revived after the crash. If his
proposal had been acted upon promptly, King argued, it “would have
prevented the debacle of 1929.”28
There were also several proposals for a federal tax on securities trans-
actions as a means of curbing speculation. The federal government al-
ready taxed stock transactions, as did New York, but these taxes were
very small because they were intended to raise revenue rather than
to deter trading. The new proposals were much more substantial.
Representative Fiorello La Guardia of New York introduced a bill im-
posing a quarter of a percent tax on all securities sales and a 25 percent
tax on short selling, the latter of which would have been tantamount
to a prohibition. Herbert Claiborne Pell, the former congressman (and
the father of Claiborne Pell, who would represent Rhode Island in the
Senate for decades), repeatedly suggested a 1 percent tax on all sales of
securities, to prevent speculative transactions without deterring genu-
ine investment. “Such a tax would practically do away with all ‘in and
out’ speculation as it would be too big a percentage for the gambler
to face,” Pell reasoned, but the tax would be small enough to “achieve
this result without in any way interfering with legitimate purchase and
sale.” As Pell told his friend “Frank” Roosevelt, “a tax of one percent
would so lessen the profits of those who have to rely on the mainte-
nance of a quick market for their profits that the business would be
wound up.” Carter Glass favored a tax on sales of stock held by the
seller for less than sixty days. “People do not ordinarily ‘invest’ their
funds for an hour or day or week or month,” Glass insisted. “Ninety
per cent of such transactions constitute unadulterated gambling on the
state of the market and the thing should be abated.” Representative Jed
Johnson of Oklahoma introduced legislation imposing a 1 percent tax

28 Brokers’ Loans: Hearings Before the Committee on Banking and Currency, United
States Senate (Washington, D.C.: Government Printing Office, 1928), 7; New
York Times, 11 July 1929, 33; 25 Oct. 1929, 3; 15 May 1930, 21; Congressional
Record 78 (1934): 1991.
182

182 Speculation

on securities sales, or as Johnson referred to it on the House floor, a tax


on “those white-​collared parasites.” Johnson explained that “I would go
further if I had my way. I would make the tax so much that it would tax
the damnable stock exchanges out of business.”29
The most common and well-​known laws governing stock transac-
tions in the years before the crash were the state blue sky laws, so it is
not surprising that after the crash, many suggested that Congress should
enact a federal blue sky law. Some used the term “blue sky” in a general
sense, to describe any laws that would govern the stock market. Adolf
Berle, for example, discussed “the purpose of national blue sky regu-
lation” without meaning that the federal government should actually
copy what the states had done. But others hoped Congress would enact
a law that would do exactly what the first generation of state blue sky
laws had done back in the 1910s—​set up a government body to evalu-
ate the soundness of proposed stock issues and bar the unsound from
ever being offered to the public. Representative Adolph Sabath intro-
duced a federal blue sky bill that would have established a three-​person
commission to pass on the merits of all stock issues. The idea was sup-
ported by state blue sky officials, who were all too aware of their own
impotence. Donald Pomeroy, who ran Minnesota’s Securities Division
and served as the president of the National Association of Securities
Commissioners, thought that a federal blue sky agency would be “of
invaluable assistance and benefit” to state regulators, because “the sale
and distribution of securities generally is a national rather than a state
problem.” One former state blue sky official, James Mott, now repre-
sented Oregon in the House of Representatives. He thought the securi-
ties bills that eventually became law in 1933 and 1934 were “amateurish”
because their drafters had no blue sky experience and had neglected to
speak with anyone who did. “The State blue-​sky laws of this country
constitute the entire corpus juris of securities regulation in the United
States,” Mott declared. “The authors of this bill would have done well

29 Pittsburgh Post-​Gazette, 8 Dec. 1931, 33; Congressional Record 78 (1934): 8097;


Herbert Claiborne Pell, “A Cure for Speculation,” North American Review
231 (1931):  313; Herbert Claiborne Pell to Franklin Roosevelt, 14 Apr. 1933,
FDR, POF 34, box 1; Carter Glass to L. C. Soule, 1 Oct. 1931, CG, box 278;
Congressional Record 78 (1934): 8102–​3.
  183

Aftershocks of the Crash 183

to consult some of them before trying to draft a Federal securities law.”


What was missing, Mott insisted, was a “securities commissioner with
authority to make an investigation of every security proposed to be
sold.”30
But the state blue sky laws, especially the early ones that empow-
ered government officials to judge the merits of stock offerings, also
served as a negative model, a reminder of a regulatory approach that
had proven unworkable. “Competition and the progress of inven-
tion make it inevitable that many enterprises will fail,” explained the
law professor William O.  Douglas. “This is reason enough why the
state should not pronounce investments sound or unsound.” It was
simply impossible to predict whether an investment would pay off.
The Washington lawyer Huston Thompson, a former member of the
Federal Trade Commission, recalled “some promotions started in the
Southwest which looked perfectly ridiculous, as wildcat as they could
be. Of course, they were advertised as great opportunities for investment
and profit,” so the FTC investigated to determine whether they were
frauds. “In one of those wildcat schemes,” Thompson continued, “they
drilled wells and one of them brought in a 10,000-​barrel well. Where
would we have been if we had said, ‘You cannot go ahead with that
speculation’?” And when state officials had tried to separate the sound
businesses from the unsound, their record had not been good. “I have
worked with Blue Sky Commissions in the various states for some years
and I have never found them anything but an annoyance,” remarked
the New York corporate lawyer Guido Pantaleoni. “I have never known
them to stop a fraud.” It was just too hard a job. As the lawyer William
Breed concluded, “it is impossible for a body of men to determine how
a business is going to come out.” “Speculative risk … will be present
in the future as it has been in the past,” acknowledged Joseph Kennedy,
the first chairman of the Securities and Exchange Commission, “for no
body of men, no government, no nation, is sufficiently wise to define

30 “The Purpose of National Blue Sky Regulation,” n.d., AAB, box 20; New York
Times, 6 Feb. 1932, 25; Donald L. Pomeroy to A. A. Berle Jr., 28 Nov. 1932,
AAB, box 20; Congressional Record 78 (1934):  8101; Congressional Record 77
(1933): 2947.
184

184 Speculation

the perfect investment, or to guarantee it, or to eliminate the risks of


speculation.”31
Corporate officers and directors were especially criticized in the wake
of the crash, for taking advantage of their inside knowledge to speculate
at shareholders’ expense. One such critic was the industrialist Alfred du
Pont, who had sued his own brother over such matters during his days
at the family chemical company. “How officers placed in positions of
trust, as all the officers of corporations are, can betray those trusts by
speculating on information which comes to them officially is beyond
my understanding,” du Pont despaired. For a cure, he favored “gov-
ernment control over the operation of stock exchanges to … prohibit
just this sort of thing.” The ability of insiders to enrich themselves at
the public’s expense was the main topic of high-​profile hearings in the
Senate Banking Committee in 1933 and 1934, led by the committee’s
chief counsel, Ferdinand Pecora.32
The stock market crash and the Depression thus gave rise to a wide
variety of proposals to curtail speculation. In the early 1930s, a well-​
informed observer might reasonably have expected Congress to pro-
hibit short selling, to prohibit margin trading, to bar banks from lend-
ing to brokers and speculators, to impose a substantial tax on securities
transactions, to ban insider trading, or perhaps even to enact a federal
blue sky law modeled on the state laws of the preceding two decades. As
it turned out, however, Congress did none of those things.

31 William O. Douglas and George E. Bates, “The Federal Securities Act of


1933,” Yale Law Journal 43 (1933): 172; Federal Securities Act: Hearing Before
the Committee on Interstate and Foreign Commerce, House of Representatives
(Washington, D.C.: Government Printing Office, 1933), 58; Guido Pantaleoni
to Adolf Berle, 9 May 1933, AAB, box 20; Securities Act: Hearings Before the
Committee on Banking and Currency, United States Senate (Washington, D.C.:
Government Printing Office, 1933), 177; Address of Hon. Joseph P. Kennedy,
Chairman of Securities and Exchange Commission, at National Press Club July
25, 1934 (Washington, D.C.: Government Printing Office, 1934), 4.
32 Alfred I. du Pont to Frederic C. Walcott, 10 June 1932, HHL, box 159A, folder 3;
Michael Perino, The Hellhound of Wall Street: How Ferdinand Pecora’s Investiga­
tion of the Great Crash Forever Changed American Finance (New York: Penguin
Press, 2010); Ferdinand Pecora, Wall Street Under Oath:  The Story of Our
Modern Money Changers (New York: Simon & Schuster, 1939).
╇ 185

Aftershocks of the Crash 185

The Best of€Disinfectants


“When Wall Street took that tail spin,” Will Rogers cracked, “you had
to stand in line to get a window to jump out of, and speculators were
selling space for bodies in the East River.” But his next line was no
joke, especially not to the Hoover administration. “You know there is
nothing that hollers as quick and as loud as a gambler,” Rogers noted.
“They even blame it on Hoover’s fedora hat.” The very next day, one of
Hoover’s advisors remarked that “when each individual is explaining to
his wife how he lost the family funds or to partners the disappearance of
the concern’s assets, the blame, of course, must be placed on someone
and, as Will Rogers said ‘It might be on Hoover’s fedora hat.’â•›”33
The administration was in something of a bind. On one hand,
Hoover viewed the boom-╉and-╉bust cycle as a natural phenomenon
that defied human intervention. “These vast contagions of speculative
emotion have hitherto throughout all history proved themselves un-
controllable by any device that the economist, the business man, or
the Government has been able to suggest,” he told the Chamber of
Commerce in 1930. Even the government’s power to raise interest rates,
he recalled, had “offered little real retardation to the speculative mania
of the country.” And a downturn was a particularly inopportune time
to attack the stock market or to propose new restrictions on trading,
because such moves might cause prices to fall even more. Even an in-
vestigation of traders would be risky, as one member of the administra-
tion pointed out. “If the principal bear raiders were put on the stand,”
he predicted, “they would attempt to show that stocks were much over
valued at the present time, … which would greatly disturb business.”
This tendency to avoid major change was reinforced by the fact that the
Republicans held majorities in both houses of Congress between 1929
and 1931 and still controlled the Senate between 1931 and 1933. As one of
Hoover’s supporters observed, “the Stock Exchange is one of the most
powerful and far reaching institutions in the country, and twenty out of

33 James M. Smallwood and Steven K. Gragert, eds., Will Rogers’ Daily Telegrams
(1978) (Claremore, Okla.:  Will Rogers Memorial Museums, 2008), 2:1929;
Anonymous memorandum, 25 Oct. 1929, HHL, box 154, folder 14.
186

186 Speculation

every twenty five of its members are republicans and contribute toward
republican campaigns.”34
On the other hand, not proposing new restrictions opened the ad-
ministration up to considerable criticism. Roosevelt’s strategy in the
1932 election “was to allege that I had made the depression and then
done nothing about it,” Hoover later complained—​“that I was person-
ally responsible for the stock market boom and the orgy of specula-
tion.” The Hoover administration had pursued none of the regulatory
options that were on the table. Roosevelt took the opportunity to cam-
paign on a vague promise to clean up Wall Street. “The money chang-
ers have fled from their high seats in the temple of our civilization,” he
declared in his inaugural address. “We may now restore that temple to
the ancient truths. The measure of the restoration lies in the extent to
which we apply social values more noble than mere monetary profit.”
Among the needed “safeguards against a return of the evils of the old
order,” Roosevelt announced, was “a strict supervision of all banking
and credits and investments, so that there will be an end to speculation
with other people’s money.”35
Roosevelt became president on March 4, 1933. By March 29, securities
bills had been introduced in both houses of Congress, and the Securities
Act of 1933 was enacted in late May, as part of the flurry of legislation
in the first hundred days of Roosevelt’s presidency. The act required a
company issuing securities to file with the Federal Trade Commission a
registration statement containing a great deal of information about the
company’s financial condition. (The FTC would be replaced in this role
a year later by a new agency, the Securities and Exchange Commission.)
Companies also had to disclose much of the same information to pro-
spective purchasers. In broad outline, corporate insiders would be liable
to purchasers of the stock for their losses if their disclosures included
any material falsehoods or omitted any important facts.36 The Securities

34 Address of President Hoover at the Annual Dinner of the Chamber of Commerce


of the United States (Washington, D.C.: Government Printing Office, 1930), 3,
6; Anonymous memorandum, 2 Mar. 1932, HHL, box 159A, folder 10; Frank
W. Blair to Robert D. Hatch, 23 Apr. 1932, HHL, box 159A, folder 3.
35 Hoover, Memoirs, 3:240; Roosevelt’s inaugural address at http://​www.presidency
.ucsb.edu/​ws/​index.php?pid=14473.
36 48 Stat. 74 (1933).
  187

Aftershocks of the Crash 187

Act did not impose any substantive limits on trading: it did not regulate
short selling or margin trading, it did not prohibit corporate insiders
from taking advantage of their positions, and it said nothing about what
kinds of transactions would be permitted or forbidden. Nor did the
Securities Act empower government officials to protect investors by re-
viewing the merits of a proposed offering. Instead, the act required only
the disclosure of information before selling stock.
Why this reliance solely on disclosure? Disclosure had been a favored
regulatory strategy of the progressive reformers earlier in the century.
“Sunlight is said to be the best of disinfectants,” Louis Brandeis had
famously written in 1913. Indeed, Brandeis coined this metaphor while
arguing that investment bankers should be forced to reveal to inves-
tors the commissions they had received for their services.37 Progressive
reformers, who had long looked across the Atlantic for inspiration,
could cite the example of the British Companies Act, which required
corporations issuing stock to publish a prospectus containing speci-
fied information.38 The second wave of state blue sky laws, from the
mid-​1910s forward, had required many new companies to make various
disclosures before selling stock. In the intervening years, others had
applied this principle to corporate information more generally. The
economist William Ripley contended in 1926 that shareholders even in
established companies had a “right to adequate information” and that
official corporate pamphlets fell woefully short of telling shareholders
what they needed to know. During the 1932 presidential campaign,
Ripley urged Herbert Hoover to make corporate disclosure part of his
platform and warned that if Hoover did not, Roosevelt would.39

37 Louis D. Brandeis, “What Publicity Can Do,” Harper’s Weekly, 20 Dec. 1913,
10–​12. This essay was later published as a chapter in Louis D. Brandeis, Other
People’s Money and How the Bankers Use It (New York: Frederick A. Stokes Co.,
1914), 92–​108.
38 On Progressive reformers’ use of European models, see Daniel T.  Rodgers,
Atlantic Crossings: Social Politics in a Progressive Age (Cambridge, Mass.: Harvard
University Press, 1998).
39 William Z. Ripley, “Stop, Look, Listen! The Shareholder’s Right to Adequate
Information,” Atlantic Monthly, Sept. 1926; William Z. Ripley, Wall Street
and Main Street (Boston: Little, Brown and Co., 1927), 156–​207; William Z.
Ripley to Herbert Hoover, 11 July 1932, HHL, box 159A, folder 3.
188

188 Speculation

And Roosevelt did. In his acceptance address at the Democratic con-


vention in July, Roosevelt sounded like Brandeis. He called for “the let-
ting in of the light of day on issues of securities, foreign and domestic,
which are offered for sale to the investing public.” Such a measure, he
declared, “would help to protect the savings of the country from the dis-
honesty of crooks and from the lack of honor of some men in high finan-
cial places. Publicity is the enemy of crookedness.” The Democratic Party
platform advocated “protection of the investing public by requiring to
be filed with the government and carried in advertisements of all foreign
and domestic stocks and bonds true information as to bonuses, commis-
sions, principal invested, and interests of the sellers.” As Joel Seligman has
observed, Roosevelt took a Brandeisian approach to the stock market: in
his speeches while campaigning and as president, Roosevelt often para-
phrased Brandeis in calling for greater disclosure by sellers of securities.40
The primary intellectual influence on the 1933 act was the Harvard
law professor Felix Frankfurter, a longtime Brandeis disciple and a fre-
quent advisor to Roosevelt dating back to Roosevelt’s days as gover-
nor of New York. Frankfurter was assisted in the actual drafting by his
Harvard colleague James Landis and by two former students, Thomas
Corcoran and Benjamin Cohen. At times they even received sugges-
tions from Justice Brandeis himself. This was a group predisposed to
favor disclosure as a regulatory tool. As Roosevelt’s advisor Raymond
Moley recalled, “there was a deep split in the early New Deal between
the so-​ called Brandeis school—​ consisting of Frankfurter, Samuel
Untermyer, James M. Landis, and Benjamin V. Cohen—​and believers
in national economic planning like Hugh Johnson, [Rexford] Tugwell,
and [Henry] Wallace.”41 By putting the former faction in charge of

40 http://​www.presidency.ucsb.edu/​ws/​?pid=75174; http://​www.presidency.ucsb
.edu/​ws/​index.php?pid=29595; Seligman, Transformation, 19–​20, 41–​42.
41 James M. Landis, “The Legislative History of the Securities Act of 1933,” George
Washington Law Review 28 (1959): 29; Donald A. Ritchie, James M. Landis: Dean
of the Regulators (Cambridge, Mass.: Harvard University Press, 1980), 43–​61;
William Lasser, Benjamin V.  Cohen:  Architect of the New Deal (New Haven,
Conn.: Yale University Press, 2002), 65–​85; Bruce Allen Murphy, The Brandeis/​
Frankfurter Connection:  The Secret Political Activities of Two Supreme Court
Justices (New York: Oxford University Press, 1982), 132–​38; Raymond Moley,
The First New Deal (New York: Harcourt, Brace & World, 1966), 306.
  189

Aftershocks of the Crash 189

drafting the Securities Act, Roosevelt effectively decided in favor of dis-


closure and against more substantial government involvement in the
securities market.
Frankfurter’s group deliberately shut other voices out of the drafting
process, especially voices from the New York legal and financial com-
munity. As the New Yorker Adolf Berle explained to William Woodin,
Roosevelt’s first Treasury secretary, “the new draft has been made up by
Felix Frankfurter, apparently without consulting any of the banking
or accounting people here. Conceding that all of the banking com-
munity are villains (they are not), it still would be useful to have their
experience.” Berle acknowledged that “Felix is a brilliant man with ex-
cellent ideas, but some experience sometimes helps in these matters.”
(Berle was less complimentary when he complained to Roosevelt that
Frankfurter “would like to do nothing unless he can have everything
he wants.”) Arthur Dean, one of the leaders of the New York corporate
bar, was apoplectic about Frankfurter’s utter lack of interest in the views
of securities lawyers. “Last Spring I offered to sever myself for a period
of months from the work in the office and place myself entirely at Felix
Frankfurter’s disposal in working on a draft of the Act,” he sputtered
to the banker Alexander Sachs. But Frankfurter “declined the offer and
I have since learned that he seriously questions whether any one who is
actively engaged in passing on security issues can have a sufficiently ob-
jective mind in drafting such legislation.” (“Unfortunately, Dean seems
to be incapable of being satisfied by any reasonable measures,” James
Landis explained to Sachs, who had left Lehman Brothers to work at
the National Recovery Administration. “He represents one point of
view; the Securities Act another.”42) Drafted very quickly and in private
by a small group of Harvard law graduates under the leadership of Felix
Frankfurter, the Securities Act of 1933 embodied a philosophy of disclo-
sure as a means to cure the stock market.
The purpose of disclosure was to protect the ordinary person, who
had no expertise in financial matters and who would otherwise lack

42 Adolf Berle to William H. Woodin, 12 Apr. 1933, AAB, box 20; Adolf Berle
to Franklin Roosevelt, 11 Jan. 1933, FDR, PPF 1306; Arthur H.  Dean to
Alexander Sachs, 16 Sept. 1933, AS, box 150; James Landis to Alexander Sachs,
5 Oct. 1933, JML-​HLS, box 9, folder 7.
190

190 Speculation

access to the information he needed to make an intelligent investment.


“Today the need to protect the public against itself and against specu-
lation in stocks is far more urgent than ever before, because the entire
country has become ticker-​minded,” the broker Edwin Lefèvre argued.
“In the old days the stock market was more like a battlefield where one
professional or group of professionals fought against other profession-
als,” but by the late 1920s, “the man or woman who wasn’t trading in
the stock market was a freak, a coward, a pauper or a great but unrec-
ognized artist.” With this greatly expanded participation in the market,
argued Joseph Kennedy, “the average American stockholder obviously
is a man of such small means that he needs governmental protection.”
With “every widow or doctor in Podunk or Oshkosh” buying stock,
noted Jerome Frank, one of Kennedy’s successors, “it is precisely that
kind of poor sucker that the statute is designed to protect.” But rather
than protect investors by guiding their decisions, as the initial blue
sky laws had tried to do, the Securities Act would protect investors by
supplying them with enough knowledge to make their own decisions.
“Government can’t provide any substitute for investors’ judgment,” in-
sisted William Douglas. “We can demand full disclosure of the facts,
but we cannot provide sound business judgment, nor can we save a fool
from his folly.” Ordinary stockholders should “get the same kind of
fair treatment they would get if they were big partners instead of little
partners in industry.”43
Mandating disclosure on the part of issuers was hardly an obvious
response to the crash. For one thing, the crash could not plausibly be
blamed on a lack of disclosure, because corporate insiders lost just as
much as anyone. “The great losses suffered through securities pur-
chased in the boom years have been due primarily to the effect of the
world-​wide depression, which has caused values and prices not alone
of securities but of all commodities to fall precipitously,” noted the

43 Edwin Lefèvre, “Blame the Broker,” Saturday Evening Post, 9 Apr. 1932,
26; Address of Hon. Joseph P. Kennedy, Chairman of Securities and Exchange
Commission, at Meeting of the Boston Chamber of Commerce (Washington,
D.C.: Government Printing Office, 1934), 2; Jerome Frank to Emanuel Celler,
28 Aug. 1939, JNF, box 23, folder 90; Press conference, 22 Sept. 1937, 2–​3,
WOD, box 24, folder 3.
  191

Aftershocks of the Crash 191

New York lawyer Eustace Seligman. “With comparatively few excep-


tions, the losses suffered by investors in securities have been due to the
fact that the investors and the investment bankers, like everyone else
in the country, mistakenly believed that the New Era was here to stay.”
Seligman thought disclosure was a good idea, and indeed, as early as
1925, he had favored requiring corporations to publish regular state-
ments of their financial condition. But he had no illusions that a disclo-
sure requirement would have saved anyone from losing much money.44
Moreover, it was not clear that the ordinary investors the Securities
Act sought to protect would actually read the financial information that
companies had to disclose, or that they would understand it if they did
read it. “Investors by and large are woefully ignorant,” lamented the
investment banker E. G. Parsly. “While there are a few exceptions in-
vestors in the main depend on someone for advice and are for the most
part incapable of forming any opinion of their own as to the merits of
a security on the basis of cold facts submitted to them.” What would
investors make of the prospectus required by the Securities Act? It “will
be beyond the comprehension of the vast majority,” Parsly concluded,
“and I am satisfied that a very large percent of investors will not even at-
tempt to read such prospectuses. I know from experience that it is hard
to get them to read a circular of even the briefest character.” The invest-
ment banker Alexander Sachs was an early New Dealer—​he ran the
economic research division at the National Recovery Administration—​
but he was just as skeptical about the value of disclosure. The prospec-
tus was bound to be “as elaborate as an insurance policy or a real estate
lease,” Sachs worried, “which have become so complicated that the av-
erage man signs them without reading.”45 If investors would not read
a prospectus, or if they could not understand what they were reading,
what good would disclosure do?
Bankers and corporate insiders had a very different concern—​not
that the Securities Act would be ineffective but that it would be too effec-
tive. They “are afraid,” Russell Leffingwell of the Morgan bank told his
old friend Franklin Roosevelt. “They are really afraid.” The act imposed

44 Eustace Seligman, “Amend the Securities Act,” Atlantic Monthly, March 1934.


45 E. G. Parsly to Baldwin B. Bane, 15 Sept. 1933, JML-​LC, box 149; Alexander
Sachs to James Landis, 6 Feb. 1934, AS, box 150.
192

192 Speculation

liability for “an untrue statement of a material fact” or the omission “to
state a material fact.” But how could anyone be 100 percent sure that
every single sentence in a long document was exactly correct? “Officers
and directors of a company must rely upon engineers, accountants and
attorneys for information,” D. F. Kelly of Chicago protested. “No man
who values his good name would be willing … to jeopardize his repu-
tation and accept the responsibility for the errors of his professional
advisers.” And of the infinite number of facts that were not included
in a registration statement or a prospectus, how could anyone predict
which ones a court, years later, might deem “material”? The banker
W. B. Neergaard worried that “if we sell to an investor, bonds, even of
such quality of those of the American Telephone & Telegraph Company
or first mortgage bonds of the Union Pacific Railway, and the bond
market, as a whole, should decline ten points, that investor could bring
an action against us to recover his loss marketwise on the grounds that
we did not disclose to him all of the material facts.” The law “would
effectively cripple the financing of all corporations,” complained the
Merchants’ Association of New York, “because of the excessively drastic
responsibilities imposed … upon directors and professional advisors.”
The Boston investment banker Waldo Kendall concluded that “if the
intent of the Act had been to put all reputable investment bankers out
of business it could not have been better framed.”46
Not long after the law’s enactment, it was already a common view
among the New York investment banking and legal community that the
liability provisions in the Securities Act were deterring companies from
issuing securities. The banker W. C. van Antwerp told Felix Frankfurter
in June that the act had already driven one investment bank out of
business. “In my own certain knowledge,” thundered the lawyer Henry
Stimson, “that Act today, by reason of the dangerous liability which it
imposes upon those who must sign the registration statements for all

46 Russell Leffingwell to Franklin Roosevelt, 4 Jan. 1934, FDR, PPF 866; D. F.
Kelly to Franklin Roosevelt, 11 May 1933, FDR, POF 242, box 1; Barrett
Wendell Jr. to William Phillips, 3 June 1933, FDR, POF 242, box 1; W.  B.
Neergaard to Duncan U. Fletcher, 15 Dec. 1933, JML-​LC, box 149; S. C. Mead
to Franklin Roosevelt, 12 May 1933, FDR, POF 242, box 1; Waldo Kendall to
James Landis, 18 Jan. 1934, JML-​HLS, box 7, folder 12.
  193

Aftershocks of the Crash 193

new flotations of capital bonds, is preventing not only the embarka-


tion of new capital but the refunding of short term indebtedness to an
extent which is most seriously preventing the revival of business.” John
W. Prentiss’s firm “are both stock brokers and investment bankers, or
perhaps I  might more correctly say that we were investment bankers
until the Securities Act,” he chided Roosevelt. The problem, as Prentiss
saw it, was “the penalty clauses in the Securities Act of 1933 which are
preventing us all from attempting to do any business.” The banker
Otto Kahn likened the act to Prohibition, as an example of legislation
with a laudable goal that nevertheless “overshoots the mark and is not
guided by practical knowledge and by a sound and informed appraisal
of the consequences.” In early 1934, the Chamber of Commerce pub-
lished a pamphlet insisting that the act had caused “a virtual cessation
of high-​grade and sizable corporate issues,” because “the civil liabilities
are so extreme.” Even Daniel Roper, Roosevelt’s own secretary of com-
merce, concluded that the act had brought financing to a standstill. The
Washington businessman E. L. Rice thought the economic slump that
would ensue would be enough “to bury the Democratic party at the
next and succeeding elections for another twelve years.”47
Among the Roosevelt administration and its supporters, such con-
cerns were dismissed as nonsense. “A banker or corporation director
expects his Rolls-​Royce dealer to take back from him a limousine not
of the kind or quality it was represented to be,” needled the lawyer
Bernard Flexner, “but for his own customer to ask him to take back
sour securities is sheer cheek!” When Eustace Seligman informed
Frankfurter that there was no reason to punish bankers for failing to

47 Arthur F. Dean, “Economic and Legal Aspects of the Federal Securities Act of
1933” (19 Sept. 1933), 2–​3, AS, box 150; W. W. Dulles to Alexander Sachs, 12
Sept. 1933, JPL, box 68; Henry Bruère to Franklin Roosevelt, 1933, JPL, box
68; W. C. van Antwerp to Felix Frankfurter, 9 June 1933, FF, reel 84; Henry
Stimson to Felix Frankfurter, 5 Dec. 1933, FF, reel 84; John W. Prentiss to
Franklin Roosevelt, 15 Feb. 1934, FDR, POF 242, box 1 (emphasis added);
Otto H. Kahn to Franklin Roosevelt, 12 May 1933, FDR, POF 242, box 1;
The Federal Securities Act of 1933 (Washington, D.C.: Chamber of Commerce,
1934), 5, 9, in JML-​HLS, box 9, folder 2; Daniel Roper to Franklin Roosevelt,
2 Feb. 1934, FDR, POF 34, box 1; E. L. Rice to Franklin Roosevelt, 24 May
1933, FDR, POF 242, box 1.
194

194 Speculation

disclose information, because “bankers of standing and of financial re-


sponsibility have in every case that I  have had any connection with,
gone to the extreme caution to insure that the circular was correct and
complete,” Frankfurter rather snidely retorted: “What a pity that you
have not been connected with all security issues, for then there would
be no need for any corrective legislation on the subject.”48
The New Dealers interpreted the slowdown in securities issues as
a deliberate effort by the bankers and securities lawyers to blackmail
Congress into repealing the Securities Act. Frankfurter told Roosevelt
that he had been told that the “leading bankers and the big law firms
are trying to create a bankers’ strike.” As Roosevelt complained, “some
of these so-​called bankers would be perfectly willing to let ten million
people starve if they could continue to sell securities for a nice com-
mission.” Frankfurter thought “the real culprit of the drive against the
Securities Act are some of the leading law firms who made such a fat
killing out of the abuses which brought the Securities Act into exis-
tence. They really want to do business at the old stand.”49
A year later, the Securities Exchange Act of 1934 took more direct
aim at the speculation commonly said to have caused the crash, but
even the 1934 act imposed limits that were mild relative to the propos-
als that had circulated in the preceding years. Rather than prohibit-
ing margin trading altogether, or even requiring customers to deposit
a certain percentage of the purchase price with the broker, the act del-
egated to the Federal Reserve the power to prescribe rules to govern
margin requirements. Rather than barring or limiting short selling, the
act delegated to the new Securities and Exchange Commission the au-
thority to prescribe rules to govern short selling. And rather than ban-
ning corporate insiders from trading on their superior information, the

48 Franklin Roosevelt to Felix Frankfurter, 2 Nov. 1934, FDR, PPF 866; Franklin
Roosevelt to Otto H.  Kahn, 22 May 1933, FDR, POF 242, box 1; Bernard
Flexner, “The Fight on the Securities Act,” Atlantic Monthly, Feb. 1934; Eustace
Seligman to Felix Frankfurter, 24 Apr. 1933, and Felix Frankfurter to Eustace
Seligman, 25 Apr. 1933, FF, reel 84.
49 Max Freedman, ed., Roosevelt and Frankfurter: Their Correspondence 1928–​
1945 (Boston: Little, Brown and Co., 1967), 158; Franklin Roosevelt to Walter
Nesbit, 11 Dec. 1933, FDR, PPF 1057; Felix Frankfurter to Louis Howe, 13
Sept. 1933, FF, reel 84.
  195

Aftershocks of the Crash 195

act merely required insiders to disclose trades in their own companies’


stock and prohibited insiders from profiting from purchases and sales
(or sales and repurchases) within a six-​month period.50
The Securities Exchange Act also prohibited various kinds of fraud,
but fraud was already illegal and had no defenders, so the act’s main
target was speculation. The object, as Roosevelt put it, was to make
stock prices “represent what is going on in the business and in our eco-
nomic life rather than mere speculative or ‘technical’ conditions in the
market.” All the act would do, Sabath insisted, was “restrict the gam-
bling activities of a small group of men who have no interest in the wel-
fare of the Nation.” Representative Carl Mapes of Michigan explained
that the act made no “attempt to draw any distinction between specula-
tion and investment, or to tell when speculation ends and investment
begins. It aims only to prevent excessive speculation.” Had such limits
been in effect a decade earlier, declared John Cochran of Missouri, “the
crash of 1929 would never have happened.”51
Critics doubted that such modest measures would have much effect
on the quantity of speculation. “It is quite ingenuous to suppose that
people can be wholly freed from the speculative urge merely by giving a
bureau in Washington regulatory power over stock exchanges,” scoffed
the Saturday Evening Post. Elmer Studley, whose congressional dis-
trict included Wall Street, cautioned that speculation “is as old as the
human race itself. To control it is quite beyond the powers of an act of
Congress.” Even Samuel Rayburn, the House sponsor of the act, admit-
ted that regulating the stock exchanges was at most a partial solution to
the problem of overspeculation. “I also recognize another man who is
very largely responsible for the misfortunes of the country and the ex-
cessive stock speculation and debacle,” Rayburn conceded. “That is Mr.
American Citizen who wants to get something for nothing.”52 This kind

5 0 48 Stat. 881 (1934), sections 7, 10, 16.


51 Stock Exchange Regulation: Letter from the President of the United States to the
Chairman of the Committee on Banking and Currency (Washington, D.C.:
Government Printing Office, 1934), 5; Congressional Record 78 (1934): 7689,
7921, 7693.
52 “More Regulation,” Saturday Evening Post, 17 Mar. 1934, 22; Congressional
Record 78 (1934): 7942, 7862.
196

196 Speculation

of criticism came from two camps—​those who favored no additional


regulation of the market (because such regulation would be pointless)
and those who favored much stronger regulation (to curtail speculation
in a significant way). From either perspective, the 1934 act was weak
medicine indeed.
Much of the financial community had the opposite criticism—​that
the act would be too strong. “Had not conceived that any bill could
be drawn so adverse to our business,” a New York investment banker
cabled Roosevelt. “In eliminating or punishing crookedness it ham-
strings every honest banker corporation officer and director.” “This is
going from no control at all to over control,” complained the president
of a Virginia bank. “This bill places in the hands of a small commis-
sion dictatorial powers similar to those enjoyed by Mr. Hitler and Mr.
Mussolini.” Another banker predicted that the 1934 act “will bring on
a panic, ruin the country, and be the death of the Democratic party.”53
Brokers worried about the imminent margin requirements, which
would make it harder to make loans on securities than on any other
kind of asset. “The bill is an attempt to treat what is really a credit
matter as a problem of morality,” one broker argued. “In attempting
to prohibit gambling, Congress is really tinkering with the credit prob-
lem of the country.” By reducing the quantity of trading, warned the
owners of an Ohio company, the margin requirements would widen
the spread between prices bid and asked on the stock exchanges, which
would only cause stock prices to fluctuate even more.54
Corporate officers and directors worried about the ban on profiting
from purchases and sales of their own company’s stock within a six-​
month period. “It has always been regarded as important that those
running a company have a large financial interest at stake,” declared the
banker and director Waddill Catchings. “How can a man afford to buy
stock in a company if he carries the risk of loss for six months but must

53 John M. Hancock to Franklin Roosevelt, 1 Feb. 1934, AS, box 150; L. M. von
Schilling to Carter Glass, 26 Feb. 1934, CG, box 316; Henry A.  Miller to
Carter Glass, 28 Feb. 1934, CG, box 316.
54 Frazier Jelke & Co., “Morning Stock Letter” (22 Mar. 1934), FDR, POF 34,
box 1; Robert Patterson Jr. and M. M. Harrod to Carter Glass, 27 Feb. 1934,
CG, box 316.
  197

Aftershocks of the Crash 197

pay any gain to the Corporation?” Catchings predicted that “men of af-
fairs will be reluctant to act as directors.” His concern was shared by the
vice president of a cement company, who feared that “responsible and
able corporate directors and executives throughout the country will feel
constrained to tender their resignations.”55
The New York Stock Exchange complained about the regular reports
that listed corporations would have to file, a burden the exchange con-
sidered so onerous that the government “might dominate and actually
control the management of each listed corporation.” And many wor-
ried that the 1934 act would discourage so much capital from entering
the market that the value of all stocks would plummet. “Is it fair to
the people who own securities?” asked one of Carter Glass’s correspon-
dents. “It may be popular at the minute to punish Wall Street, but
I am wondering how popular this bill will be with investors all over the
country, when they see the marketability and prices of their securities
perhaps seriously affected.”56
In practice, the 1934 act’s bite would depend on how much of a
margin the Federal Reserve required and on the precise nature of the
short-​selling regulations that would be promulgated by the Securities
and Exchange Commission. The Federal Reserve initially required a
minimum margin of only 25 percent—​that is, brokers could still lend
their customers 75  percent of the value of what the customers were
buying. The margin requirement fluctuated thereafter, to 55 percent in
1936, then back down to 40  percent in 1937. Since World War II it
has been at 50 percent or higher, and it has stayed right at 50 percent
since 1974.57 The purpose of giving the Federal Reserve the authority
to set a margin requirement was to reduce the volatility of stock prices,
which was widely thought to be a product of speculative trading on
credit. In fact, however, the effect of the margin requirement has been

55 Waddill Catchings to Marvin H. McIntyre, 23 Apr. 1934, FDR, POF 34, box
2; Dwight Morgan to Carter Glass, 5 Mar. 1934, CG, box 316.
56 Richard Whitney to the Presidents of All Listed Corporations, 14 Feb. 1934,
AAB, box 22; Frank McNulty Ransom to Carter Glass, 29 Mar. 1934, CG,
box 316.
57 Susan B. Carter et al., eds., Historical Statistics of the United States (Cambridge
University Press Online Edition, 2006), table Cj863.
198

198 Speculation

studied extensively over the years by economists, who have been nearly
unanimous in concluding that the margin requirement has had little
or no effect on the volatility of stock prices.58 Even without a margin
requirement, brokers may well be careful about lending their money to
customers engaging in risky purchases. And of course borrowing from
a broker is not the only way to speculate. The margin requirements au-
thorized by the 1934 act thus appear not to have done much to curtail
speculation.
The same is true of the short-​selling regulations authorized by the
1934 act. The Securities and Exchange Commission did not regulate
short selling until 1938, and even then, all it did was promulgate the
“uptick rule,” which prohibited short sales only at prices lower than
the last sale price.59 The consensus among economists is that the uptick
rule imposed a cost on traders without producing any corresponding
benefit: it made short selling slightly more difficult, but it had no effect
whatsoever on the speed with which prices declined in a downturn.60
For this reason, the SEC repealed the uptick rule in 2007.

58 Thomas Gale Moore, “Stock Market Margin Requirements,” Journal of


Political Economy 74 (1966): 158–​67; Michael A. Goldberg, “The Relevance
of Margin Regulations,” Journal of Money, Credit and Banking 17 (1985):
521–​27; Michael L. Hartzmark, “The Effects of Changing Margin Levels on
Futures Market Activity, the Composition of Traders in the Market, and Price
Performance,” Journal of Business 59 (1986): S147–​S180; David A. Hsieh and
Merton H. Miller, “Margin Regulation and Stock Market Volatility,” Journal
of Finance 45 (1990): 3–​29; Paul J. Seguin and Gregg A. Jarrell, “The Irrelevance
of Margin: Evidence from the Crash of ’87,” Journal of Finance 48 (1993):
1457–​73; Christian E. Weller, “Policy on the Margin: Evaluating the Impact
of Margin Debt Requirements on Stock Valuations,” Journal of Economics and
Finance 26 (2002): 1–​15.
59 Jonathan R.  Macey et  al., “Restrictions on Short Sales:  An Analysis of the
Uptick Rule and Its Role in View of the October 1987 Stock Market Crash,”
Cornell Law Review 74 (1989): 799–​835.
60 Lynn Bai, “The Uptick Rule of Short Sale Regulation: Can It Alleviate Downward
Price Pressure From Negative Earnings Shocks?,” Rutgers Business Law Journal
5 (2008): 1–​63; Karl B. Diether et al., “It’s SHO Time! Short-​Sale Price Tests
and Market Quality,” Journal of Finance 64 (2009): 37–​73; Gordon J. Alexander
and Mark A. Peterson, “Short Selling on the New York Stock Exchange and the
Effects of the Uptick Rule,” Journal of Financial Intermediation 8 (1999): 90–​116.
╇ 199

Aftershocks of the Crash 199

The New Deal securities acts thus imposed only relatively mild re-
strictions on speculation, certainly much milder than might have been
expected given the severity of the Depression and the extent to which
speculation was considered responsible for it.

Capitalistic Reformers
Part of the reason for this mildness was of course the countervailing
pressure from the New York financial community, a powerful lobby
with a strong and focused interest in fending off regulation.61 But this is
hardly a complete explanation. The big New York banks were unable to
prevent Congress from erecting a wall between investment banks and
commercial banks. The banks and the stock exchange were unable to
exert much influence on the disclosure requirements of the Securities
Act of 1933, which they opposed just as strongly. And while they
succeeded somewhat in weakening the initial draft of the Securities
Exchange Act of 1934, they could not stop that legislation either. There
was considerable sentiment in Congress to strike hard at Wall Street. If
the New Dealers had wanted to impose stricter limits on speculation,
they probably could have.
But they did not want to. The primary reason for the moderation
of the New Deal securities acts lies in the backgrounds and interests
of the New Dealers themselves. They were, by and large, urban in-
tellectuals who accepted the existence of big financial institutions
as necessary. They had no desire to change the structure of the na-
tion’s financial system in any significant way. They recognized that
they differed in this regard from the earlier generation of Progressive
reformers, epitomized for the New Dealers by the octogenarian
Louis Brandeis, who was still active on the Supreme Court. Brandeis
“wants drastic taxation of big business units, accompanied by leav-
ing small business … strictly alone,” Berle informed Roosevelt. “His
view, if ever stated, would command wide popular support.” But
Berle made clear that he disagreed. “As long as people want Ford

61 Ron Chernow, The House of Morgan: An American Banking Dynasty and the
Rise of Modern Finance (New York: Simon & Schuster, 1990), 379–╉81.
200

200 Speculation

cars,” he told Roosevelt, “they are likely to have Ford factories and
finance to match.”62
From the New Dealers’ perspective, the goal of regulating the securi-
ties markets was not to transform them but to preserve them, by ward-
ing off more serious changes. The New Deal took place while European
countries were turning to communism and fascism. Much of the mo-
tivation for the securities acts, as with the rest of the New Deal, was
to prevent such threats to capitalism from taking root in the United
States. The 1934 act “is conceived in a spirit of the truest conservatism,”
declared the House report that brought it to the floor of Congress.
“This bill seeks to save, not destroy, stock markets and business, by
making necessary changes in time.” As one investment bank approv-
ingly observed, the New Dealers “who drafted the bill in its original
form are not Bolsheviks or Communists, but Capitalistic Reformers.”
Or as Berle explained, the SEC “is in existence primarily to preserve the
capitalist form.”63
This was a major theme in the correspondence of the veteran
New Dealer Jerome Frank, who chaired the SEC from 1939 to 1941
after serving in the Agricultural Adjustment Administration and the
Reconstruction Finance Corporation. “Unfairness certainly arouses the
ire of investors,” he declared. “If they are seriously mulcted, they will
grow impatient with a government that allows such things to happen.
Such an attitude will pave the way for Fascism—​and a Fascism which
will lead to a deadly attack on the very investment bankers and security
traders who are today loudest in their protests against regulation.” As he
explained on another occasion, “loss of public confidence in the integ-
rity of our securities markets might be fatal to the existence of such mar-
kets.” For this reason, he proudly informed the New York Times reporter
Arthur Krock, Frank considered the SEC “a conservative institution,

62 Adolf Berle to Franklin Roosevelt, 23 Apr. 1934, FDR, PPF 1306. See also
Jerome Frank to Morris Ernst, 30 Sept. 1939, JNF, box 21, folder 15; Jerome
Frank to Randolph Paul, 24 Sept. 1938, JNF, box 36, folder 554.
63 Ira Katznelson, Fear Itself: The New Deal and the Origins of Our Time (New
York: W. W. Norton, 2013); H.R. Rep. No. 1383, 73rd Cong., 2nd Sess., 3
(1934); Grant & Co., “Weekly Investment Letter,” 1934, FDR, POF 34, box 2;
Adolf Berle to Jerome Frank, 8 Feb. 1939, JNF, box 22, folder 41.
  201

Aftershocks of the Crash 201

striving in its own legitimate field, patiently, sensibly, but persistently,


to assist in the work of conserving, by improving, our profit system.”64
Reinforcing this mindset was the fact that many of the New Dealers
came from the same social circles as the financiers and lawyers they
sought to regulate. At the top, Franklin Roosevelt was born into a
wealthy family and attended Groton and Harvard, just like Richard
Whitney, the president of the New York Stock Exchange. (Roosevelt
was a few years older.) In 1934, in the midst of the controversy over
the new SEC, Whitney and Roosevelt attended the same sailing race.
When Whitney wanted to speak to Roosevelt, he simply called on the
telephone. Roosevelt maintained many other friendships with promi-
nent bankers and stockbrokers. Down in the trenches, many of the
lawyers who drafted the legislation and staffed the SEC had gone to
the same law schools and worked at the same law firms as the Wall
Street lawyers they were regulating. Thomas Corcoran graduated
from Harvard Law School, clerked for Oliver Wendell Holmes on the
Supreme Court, and spent five years as a securities lawyer in New York
before joining the Roosevelt administration. Benjamin Cohen, also a
Harvard graduate and a New York lawyer, had been a successful stock
speculator before he lost it all in the crash. William Douglas taught
corporate law at Yale, while Jerome Frank was a corporate reorganiza-
tion lawyer in New  York. Daniel Ernst has discovered that the first
twenty-​four lawyers hired at the SEC included twelve with Harvard law
degrees and four others from Columbia or Yale, that the agency’s first
three general counsels were all Harvard graduates, and that several early
SEC lawyers had worked at leading securities law firms in New York.
As Milton Freeman, one of these early lawyers, recalled, “the general
counsel’s office soon became the roosting place of the Class of 1925 of
Harvard Law School.” In 1932, when Hoover was still president, James
Landis—​then a young Harvard professor—​had even been paid by the
venerable New York law firm of Carter, Ledyard & Milburn to opine
that Congress lacked the constitutional authority to regulate short sell-
ing, an opinion the firm happily transmitted to its client, the New York

6 4 Attachment to Jerome Frank to William O. Douglas, 13 Oct. 1938, WOD, box


25, folder 166; Jerome Frank to Cyrus S. Eaton, 29 Apr. 1938, JNF, box 27, folder
247; Jerome Frank to Arthur Krock, 10 June 1939, JNF, box 31, folder 384.
202

202 Speculation

Stock Exchange.65 These regulators were not anti–​Wall Street; they


were Wall Streeters themselves, the same sorts of people as the ones
they were regulating. Things might have been very different if the new
president taking office in 1933 had been from a different part of the
country and had staffed the executive branch with people from a dif-
ferent background.
An additional factor tending toward mildness in regulation was that
the New Dealers all had strong personal and ideological motives to sup-
port Roosevelt’s re-​election. They knew that a stronger economy would
help Roosevelt and that overregulation would weaken the economy.
“Since we are all diligently trying to help the President in his program
of recovery,” SEC Commissioner John W. Hanes remarked, “the ma-
chinery must be well oiled in order to bring forward a continuous flow
of private capital to the market.” William Douglas declared that “it is
highly imperative that the jam now obstructing the flow of private sav-
ings to industry for capital investments should be broken.”66 Preserving
the mechanisms of finance was not just good for the country; it would
be good for the New Deal and its administrators too.
Up to a point, regulation might have facilitated the flow of capital,
if it made investors more confident about stepping into the market
after the disaster of the previous few years. But the New Dealers’ pri-
vate correspondence makes clear that they thought they were past that
point. They understood that there was a tradeoff between regulation
and market activity—​that placing new burdens on sellers of securities
meant that fewer securities were likely to be sold. Jerome Frank declared

65 Richard Whitney to Franklin Roosevelt, 10 Oct. 1934, FDR, PPF 1275; Marvin
H.  McIntyre to Franklin Roosevelt, 7 Feb. 1934, FDR, PPF 1275; Richard
Whitney to FDR, 8 Feb. 1934, FDR, PPF 1275; James L.  Houghteling to
Franklin Roosevelt, 19 Apr. 1934, FDR, POF 242, box 1; Henry McVickar
to Franklin Roosevelt, 27 Feb. 1934, FDR, POF 34, box 1; Daniel R. Ernst,
“Lawyers, Bureaucratic Autonomy, and Securities Regulation During the New
Deal” (unpublished paper, 2009), 7; Katie Louchheim, ed., The Making of the
New Deal:  The Insiders Speak (Cambridge, Mass.:  Harvard University Press,
1983), 141; James Landis to William Harding Jackson, 22 Feb. 1932, Jackson to
Landis, 25 Feb. 1932, and Jackson to Landis, 14 Apr. 1932, all in JML-​LC, box 6.
66 John W. Hanes to Jerome Frank, 15 Apr. 1938, JNF, box 28, folder 290; William
O. Douglas to Henry A. Wallace, 11 Apr. 1936, WOD, box 24, folder 1.
  203

Aftershocks of the Crash 203

that the government’s goals should be “(a) to protect investors, and also,
(b) always to avoid substantial impediments to expansion of capital,”
but he acknowledged that “those two objectives frequently cannot be
reconciled.” The New Dealers were willing to sacrifice a measure of eco-
nomic activity to protect investors. “Speculative trading must be very
greatly curtailed,” Franklin Roosevelt remarked to Henry Morgenthau
in early 1934. “That means, inevitably, a much smaller volume of trad-
ing on the stock exchanges.” Not long after, at a question-​and-​answer
session at the Economic Club of Chicago, James Landis likewise con-
ceded that it was “inevitable that in the case of some stocks the spread
between bid and asked prices will be somewhat increased due to less-
ened volume.”67
Roosevelt and Landis were right:  trading volume did decline as a
result of the securities acts. As the SEC staffer (and future chairman)
Ganson Purcell recognized in 1938, “volume has been considerably re-
duced in recent years, and other elements of what has traditionally been
known as liquidity are presently absent from our market.” But Purcell
thought this reduction in volume a worthwhile price to pay “for the
elimination of undesirable manipulative activity” and “restriction on
the irresponsible in and out speculation.” Frank conceded that “there
can be no doubt that prior to the 1929 crash we had a thicker market,”
but he too preferred a thinner market if it meant avoiding a return to
the practices of the 1920s. “If those practices are sufficiently unfair to
investors,” Frank warned, “they will revolt not only against investing,
but against the economic and political system which permits them to
be injured.”68
Opposition from brokers waned over time. “I think I am in pretty
close touch with the brokerage and investment banking community

67 Jerome Frank to “the Commission,” 11 June 1938, JNF, box 22, folder 62;
Henry Morgenthau Jr. to Franklin Roosevelt, 22 Mar. 1934, FDR, POF 34,
box 1; James Landis, untitled document with answers to questions from the
audience after a speech at the Economic Club of Chicago, 5 June 1934, JML-​
HLS, box 6, folder 2.
68 James C. Dolley, “The Effect of Government Regulation on the Stock-​Trading
Volume of the New  York Stock Exchange,” American Economic Review 28
(1938):  8–​26; Ganson Purcell to Jerome Frank, 18 Nov. 1938, JNF, box 37,
folder 580; Jerome Frank to Elisha M. Friedman, n.d., JNF, box 27, folder 233.
204

204 Speculation

in Boston,” the broker Lester Watson recalled as early as 1935, “and I


am not stretching it a bit when I say that we were very fearful when
the Securities & Exchange Act became a law that its administration
would definitely and severely interfere with if not indeed cripple our
business. Of course, the actual experience has been all the other way.”
After a meeting of the board of governors of the Boston Association of
Stock Exchange Firms, Watson told SEC chairman Joseph Kennedy
that “it was their unanimous opinion that you and your Commission
are doing one hell of a good job and they wanted me to pass that word
along to you.” Opposition from the New York brokers declined sharply
after 1938, when Richard Whitney, the former president of the New
York Stock Exchange, was convicted of embezzlement. The Whitney
scandal prompted something of a generational turnover at the stock
exchange. William McChesney Martin, the exchange’s new president,
was only thirty-​one years old, and he was far more cooperative with
the government than Whitney and his colleagues had been. “The New
York Stock Exchange, under the domination of a backward-​looking
management, continued to oppose” the New Deal legislation, William
Douglas recalled in 1938. “That former management was finally dis-
credited by the Whitney episode. Now there is a new management in
the Exchange which accepts the philosophy of the legislation; which
understands that it is better business to work with established law than
against it.” Jerome Frank paid Martin the ultimate compliment; he said
that Martin “represents a sort of New Deal in the Exchange.”69
The leading New  York investment bankers, who had vociferously
protested against the requirements imposed by the 1933 act, turned out
to be unintended beneficiaries of those requirements. By making the
sale of stock more complicated, the act gave an advantage to the more
prestigious New York banks over their smaller out-​of-​town competitors.
The detailed procedures required by the act suppressed competition

69 Lester Watson to Joseph P. Kennedy, 20 Sept. 1935, JPK, box 83; Lester Watson
to Joseph P. Kennedy, 7 Feb. 1935, JPK, box 83; Vincent P. Carosso, Investment
Banking in America: A History (Cambridge, Mass.: Harvard University Press,
1970), 380–​81; William O. Douglas to Stephen A. Early, 31 Oct. 1938, FDR,
POF 1060, box 2; Jerome Frank to Frank Murphy, 15 Mar. 1939, JNF, box 34,
folder 508.
╇ 205

Aftershocks of the Crash 205

among investment banks, which likewise benefitted the established


banks.70 In 1933, the leading investment bankers had gone on strike
to fight the Securities Act (at least that’s how it looked from inside the
Roosevelt administration), but within a few years that opposition had
evaporated.
The New York securities lawyers, who had led the fight against the
1933 and 1934 acts, also quickly adapted to the new laws. After all, these
were rules that issuers and investment banks could scarcely follow
without the paid assistance of securities lawyers at each step of the
way. Arthur Dean, who in 1933 had fumed about being shut out of
the drafting process, concluded by 1937 that he was “in full sympathy
with the underlying purposes of both the Securities Act of 1933 and the
Securities Exchange Act of 1934 and with the administrative problems
facing the Commission.” After three years of practicing under the two
laws, Dean had come to believe that “the financial community, purely
from a selfish standpoint, cannot afford to be without them.”71 The
new regulatory requirements had become the bread and butter of the
New York corporate bar.72

Speculating€Banks
For all the discussion in the early 1930s about prohibiting various spec-
ulative transactions, the most significant limit on speculation to arise
from the crash of 1929–╉33 concerned not types of speculation but types
of speculators. In four scattered sections of the Banking Act of 1933 that
have been known ever since as the Glass-╉Steagall Act, Congress prohib-
ited deposit-╉taking banks from selling or underwriting securities.73 For
the rest of the twentieth century, commercial banks (which take de-
posits and make loans) would be separate enterprises from investment

70 Paul G.  Mahoney, “The Political Economy of the Securities Act of 1933,”
Journal of Legal Studies 30 (2001): 1–╉31.
71 Arthur H. Dean, “The Lawyer’s Problems in the Registration of Securities,”
Law & Contemporary Problems 4 (1937): 155.
72 Thomas McCraw makes a similar point about accountants in Prophets of
Regulation (Cambridge, Mass.: Harvard University Press, 1984), 188–╉92.
73 48 Stat. 162, §§ 16, 20, 21, 32 (1933).
206

206 Speculation

banks (which assist firms in selling securities to investors). This separa-


tion was bitterly opposed by many bankers at the time.
The economy has changed so much since 1933 that to understand
the rationale behind the Glass-​Steagall Act, one has to reconstruct the
world of thought in which it was embedded. Why did people in the
early 1930s blame the Depression on what seems like a technical issue of
banking regulation? And why did they think that separating commer-
cial banks from investment banks was a likely way of preventing such a
disaster from happening again?
One of the most visible and troubling features of the Depression
was the unprecedentedly large number of bank failures. More than
nine thousand banks went under between 1930 and 1933. By the end of
1933, nearly half of the banks that had existed in 1929 were gone.74 The
human cost of a bank failure was far greater than it would be today.
The government did not insure bank deposits, so when a bank failed,
the depositors lost all they had.
A bank could fail for any number of reasons. Too many borrowers
might not be able to repay their loans, especially during a depression.
Too many depositors might try to take their money out of the bank at
once, again especially during a depression, when rumors that a bank
was teetering toward insolvency could be enough to cause a run on the
bank. These were the normal hazards of commercial banking. But some
of the bank failures of 1929–​33 were attributable to a different cause. In
some cases, the bank had made speculative investments with the depos-
itors’ money, investments that went sour during the stock market crash.
The ability of banks to speculate in this way was a product of a slow
evolution in the banking business over the previous thirty years. The
major urban banks began establishing or acquiring securities affiliates
in the first two decades of the twentieth century. These affiliates grew
steadily larger. In 1916, for example, the National City Company—​the
securities affiliate of the National City Bank of New York—​purchased a

74 H.  Parker Willis and John M.  Chapman, The Banking Situation:  American
Post-​
War Problems and Developments (New  York:  Columbia University
Press, 1934), 3–​20; Elmus Wicker, The Banking Panics of the Great Depression
(New York: Cambridge University Press, 1996), xv; Ben S. Bernanke, Essays on
the Great Depression (Princeton, N.J.: Princeton University Press, 2000), 44.
  207

Aftershocks of the Crash 207

well-​established investment bank, N. W. Halsey and Company, which


had placed more than $100 million in securities the previous year. All
of N.  W. Halsey’s two hundred employees except the president and
vice president became employees of the National City Company, which
instantly became one of the largest investment banks in the country.
Other banks followed suit. Throughout the 1920s, the securities af-
filiates of major commercial banks gradually took over much of the
underwriting business from private investment banks. Income from
securities began to rival interest on loans as a source of profits for com-
mercial banks.75
Well before the crash, some expressed concern about these securities
affiliates, because they allowed banks to take risks with their deposi-
tors’ money. The Annual Report of the Comptroller of the Currency
was normally a mild document, but in 1920, Comptroller John Skelton
Williams took the opportunity to excoriate the affiliates as “instruments
of speculation” that were endangering the soundness of the banks under
his supervision. “A national bank lends not only its own capital but the
money of its depositors, and in doing this is not expected to tie up its
funds in long-​time and unliquid loans in doubtful ventures,” Williams
argued. “The operations and practices of these ‘securities companies’ as
now conducted are often directly opposed to the elementary principles
of sound banking.” Williams urged Congress to bar commercial banks
from underwriting new issues of securities and from making specu-
lative investments. But the only relevant legislation in the 1920s, the
McFadden Act of 1927, strengthened the ability of commercial banks
to enter the securities market.76 When stock prices were on the way up,
there was little sentiment for keeping banks out of a line of business
that was growing ever more profitable.
The bank failures that followed the stock market crash brought the
issue into the spotlight. “Bankers should be forbidden by statute to
speculate, directly or indirectly, with the money of depositors,” urged
William Deming, who had recently stepped down as president of the

75 W. Nelson Peach, The Security Affiliates of National Banks (Baltimore: Johns


Hopkins Press, 1941).
76 Annual Report of the Comptroller of the Currency (Washington, D.C.:
Government Printing Office, 1921), 1:55–​57; 44 Stat. 1224 (1927).
208

208 Speculation

Civil Service Commission. The Columbia economist Henry Parker


Willis, who had been an advisor to the congressional committee that
created the Federal Reserve System and then the first secretary of the
Federal Reserve Board, emphasized the same point in the popular press.
“The one change which is most to be deplored in our entire banking
system of recent years has been the introduction of ‘department-​store
banking,’ or multiple-​function banking, in lieu of the older plan of
sharp division between financial types,” Willis argued. Safe commercial
banks had “become stock market banks,” a fact Willis believed was
“primarily responsible for the unliquid conditions which have come to
exist in many banks and which today constitute the greatest obstacle
perhaps to the restoration of financial soundness.”77
Of all the people newly interested in keeping banks out of the securi-
ties business, the most important was the Virginia senator Carter Glass.
Although Glass had never been a banker himself, he probably knew
more about banking than anyone else in Congress. Glass had been the
chair of the House Banking Committee in 1913, when Congress passed
the Federal Reserve Act. He had been secretary of the Treasury under
Woodrow Wilson. As a senator from 1920 until his death in 1946, Glass
took a special interest in banking issues. In 1931, he chaired a subcom-
mittee charged with surveying the national banking system, and from
1931 to 1933, he sponsored legislation to separate commercial banks from
investment banks. Although his name today, if it is remembered at all,
is virtually always paired with that of the Alabama representative Henry
Bascom Steagall, who was chair of the House Banking Committee all
through the 1930s and early 1940s, Steagall was primarily interested in
deposit insurance, which would become another important feature of
the Banking Act of 1933. The divorce of commercial banking and invest-
ment banking was Glass’s project, not Steagall’s. In the contemporary

77 William C. Deming to Herbert Hoover, 19 June 1931, HHL, box 154, folder
15; H. Parker Willis, “Reforms Due in Investment Banking—​and in Bank
Relations to Affiliates,” The Annalist, 17 Jan. 1930, 44; H. Parker Willis, “Who
Caused the Panic of 1929?,” North American Review 229 (1930): 179. Willis made
the same argument in academic journals, e.g., H. Parker Willis, “The Banking
Act of 1933—​An Appraisal,” American Economic Review 24 (1934): 108–​9.
  209

Aftershocks of the Crash 209

press, congressional efforts to accomplish that goal were always referred


to as “the Glass bill.”
Glass’s 1931 subcommittee produced a lengthy report on the bank-
ing system. One of its conclusions was that the securities affiliates had
made banks riskier enterprises than before. “There can be no objection
to the stockholders of a bank engaging in any other business they prefer
with their own funds,” the report observed. “However, if such activities
tend to affect directly the position and soundness of the bank itself,
they become of prime importance in the regulation of banking.” In
early 1932, Glass produced another report recommending that banks
be separated from their securities affiliates as much as possible. Glass
deplored “the dangerous use of the resources of bank depositors for
the purpose of making speculative profits.”78 At Glass’s urging, the
Democratic Party platform of 1932 included a declaration of support
for “the severance of affiliated securities companies from, and the di-
vorce of the investment banking business from, commercial banks.”
The hearings conducted by Glass’s subcommittee elicited widespread
concern about the dangers of speculating banks. William Randolph
Hearst published an editorial in his newspapers supporting the sepa-
ration of commercial banking from investment banking. “You know
that the crash and subsequent depression were very largely caused by
the overcapitalization and speculation encouraged by many impor-
tant banks and their affiliates,” Hearst told his readers. “Perhaps you
do not know that the high financiers do not want their highway rob-
bery interfered with. So they went to Congress to kill the Glass bill.”
The Democratic lawyer Samuel Untermyer, who had been counsel to
the House Banking Committee during the “Money Trust” investiga-
tion of 1912, declared that “nothing could more forcibly demonstrate
the extent to which the country is being ruled by financial interests
than the long years of toleration of this vicious partnership between
the banks and these speculative affiliates without a word of warning

78 Operation of the National and Federal Reserve Banking Systems: Hearings Before a
Subcommittee of the Committee on Banking and Currency, United States Senate,
Seventy-​First Congress, Third Session, Pursuant to S. Res. 71 (Washington, D.C.:
Government Printing Office, 1931), 1063; S. Rep. No. 584, 72nd Cong., 1st
Sess. (1932), 9–​10.
210

210 Speculation

or protest from the government authorities.” Glass received letters of


support from all over the country cheering on his efforts “to protect
the savings of millions of people,” as one Boston man put it, from the
“speculative tendencies” of securities affiliates.79
Glass’s correspondents, like the witnesses at his hearings, provided
numerous stories of banks that had been undone by speculation. Robert
Carson, the former president of the Iowa Chamber of Commerce, re-
ported that all four banks in Iowa City had failed, one because it had
made heavy investments in South American bonds that were now
worthless. Depositors had received only 10 percent of what had once
been their money. From Los Angeles, the lawyer Ralph Lindstrom re-
lated how one defunct local bank, the United States National Bank,
had joined in “the drunken orgy of our recently closed speculative
era.” Thousands of depositors in New York’s National City Bank were
ruined, explained a local insurance broker, because of the bank’s invest-
ments in the new aircraft industry. “If you will look up how many bank
failures there have been,” suggested a Boston broker, “you will find, I
believe, that eight out of ten are the ones that have been putting out
Investment Trusts and securities, which have gone about as bad as an-
ything could.”80
The problem, as these critics saw it, was that bankers had become
speculators—​with their depositors’ money. A Chicago engineer named
Edgar Rossiter recalled that “when the stock gambling started late in
1928 it was impossible to obtain a loan in Chicago.” Rossiter went from
bank to bank offering $125,000 in securities as collateral for a loan of
$16,500. At each bank the loan officer refused to make a loan but of-
fered to buy the securities instead. “At last one of the bankers said—​
why should we loan you $16,500 at six per cent when we are getting

79 Washington Herald, 18 Jan. 1933, CG, box 305; “Untermyer on the Glass Bill,”
Bankers’ Magazine 124 (1932): 510; Richard Feakes to David I. Walsh, 5 Jan.
1932, CG, box 284; see also Raymond E. Herman to Carter Glass, 13 Feb. 1932,
CG, box 293; Carter Glass to Garland A. Tunstill, 20 Jan. 1932, CG, box 285.
80 Robert N. Carson to Carter Glass, 10 May 1932, CG, box 296; Ralph G.
Lindstrom to Carter Glass, 7 Mar. 1933, CG, box 308; Samuel Berman to
Carter Glass, 13 May 1932, CG, box 296; L. R. Packard to Carter Glass, 20
May 1932, CG, box 297.
  211

Aftershocks of the Crash 211

from 15 to 20 percent on the stock exchange in New York?” Frederick


Wells spent thirty years as the director of a Minneapolis bank and had
come to realize that “bank officials frequently feel a greater obligation
to stockholders than to depositors, and that many of the unfortunate
conditions which have developed are the result of attempts to increase
earnings at the expense of conservative operation.”81
On this view, the solution was to force bankers to be conservative
with their depositors’ funds by restricting deposit-​taking banks to the
commercial banking business. “It seems to me that the country has a
right to expect of its Bankers high devotion to conservative principles
of banking,” insisted a man in Cleveland. The time had come “to force
the banks out of the security business.”82
This concern about bankers taking risks with their depositors’ money
was by far the most often cited reason for separating commercial bank-
ing from investment banking, but once Glass started holding hearings,
proponents of the bill had several other arguments in its favor. How
had stock prices become so inflated in the late 1920s? Senator Frederic
Walcott of Connecticut, a former banker himself, blamed the entry of
commercial banks into the stock market, which he argued had trig-
gered the speculative boom by flooding the market with money. Glass
agreed that one benefit of keeping the two kinds of banks separate was
to avoid another stock bubble “by forbidding people to use the fa-
cilities of Federal reserve banks to inflate prices on the stock exchange
to almost an inconceivable degree.” Depressions always followed over-
speculation, reasoned Robert Hopkins, a furnace maker in Ohio, and
it was money from the big banks that had fueled the New York Stock
Exchange in the 1920s. The country needed “legislation which will put
banks on a business basis instead of the main support of the biggest
gambling institution in the country.”83

81 Edgar A.  Rossiter to Carter Glass, 12 May 1932, CG, box 297; Frederick
B. Wells to Carter Glass, 21 Mar. 1933, CG, box 310.
82 Julian W. Tyler to Carter Glass, 5 Apr. 1932, CG, box 285.
83 Congressional Record 75 (1932):  9904; Operation of the National and Federal
Reserve Banking Systems, 52; Robert C. Hopkins to Carter Glass, 31 Mar. 1932,
CG, box 284. See also F. W. Buell to Carter Glass, 3 Jan. 1933, CG, box 296;
212

212 Speculation

Other critics alleged that a commercial bank faced a conflict of in-


terest when it ventured into the securities business, because while the
bank was entrusted with the care of its depositors’ money, it was also
selling securities to those very same depositors. “The trusting public
is being misled,” complained L. C. Irvine of Mobile, Alabama. “After
the age-​old education of the public to trust the banker for impartial
advice in financial matters it is little less than particeps criminis for
the Government to permit this confidence to be betrayed.” Critics sus-
pected that the banks were “more interested in the profit in the sale of
the securities, than they were in protecting the funds” of their deposi-
tors, as one banker from East St. Louis, Illinois, put it. A. H. Eckles,
the president of the Planters Bank & Trust Company in Hopkinsville,
Kentucky, lamented that his neighbors placed undue trust in the “pres-
tige and influence” of the big banks, which had induced them to buy
“securities that have proved in many cases to be practically worthless.”
Worse, the banks knew exactly how much was in their customers’ ac-
counts, which gave the banks an extra edge in pressuring customers into
buying whatever they had to sell. “The sales counter across which they
are hawking bonds and stocks is the curse of the banking system of this
country,” declared the Rev. W. A. Matthews, of the First Presbyterian
Church of Seattle. “Make it impossible for banks to be anything except
pure, simple banks.”84
Banks that sold securities also faced a conflict of interest with respect
to borrowers, many argued, because when a firm was indebted to a
bank, it was in the bank’s interest for the firm to sell securities to the
public (which would raise capital to repay the loan), even if an offering
of securities would benefit neither the firm nor the purchasers of the se-
curities. The very first issue of Fortune magazine, published in February
1930, worried for this reason that the sale of securities was “getting the

George M.  Coffin to Carter Glass, 22 Mar. 1932, CG, box 284; Walter
Schaffner to Carter Glass, 10 Oct. 1931, CG, box 302.
84 L. C. Irvine to Henry B. Steagall and Carter Glass, 1932, CG, box 293; J. F.
Galvin to Carter Glass, 12 May 1932, CG, box 296; A. H. Eckles, “A Country
Banker Looks at the Glass Bill,” Bankers’ Magazine 125 (1932): 136; Winthrop
A. Mandell to Carter Glass, 1 Apr. 1932, CG, box 285; W. A. Matthews to
Carter Glass, 12 Feb. 1932, CG, box 285.
  213

Aftershocks of the Crash 213

bank into promotional activities of which it should be judge rather


than advocate.” William Stark Smith of Milwaukee reported that “these
investment departments are frequently used to transfer poor bank loans
to the public through the issue of bonds or preferred stock.” In 1931,
Carter Glass received an anonymous letter from an employee of the
securities affiliate of a national bank, who revealed that the bank was
doing just that. “The sales talk was just dope to lull the unwary into
buying,” he disclosed. “One issue was put out to liquidate a bad loan of
the bank.” Until investment banks were kept separate from commercial
banks, warned the journalist John Flynn, “the function of the guardian
is submerged in that of adventurer and speculator.”85
These critiques of banks had a particularly sharp edge in the early
1930s. As banks failed left and right, it was a common concern that
the public was losing confidence in the banks that were still standing,
which would cause people to pull their money out of banks and keep it
at home, which would cause even more banks to fail. If shady practices
on the part of securities affiliates were making the public suspicious of
banks, some argued, that was reason enough to get banks out of the se-
curities business. “Confidence is the very life blood of a bank,” declared
a banker in Fort Worth, Texas. “Unless the security sales of our banks
are curbed, and that at once, it will break down the confidence that
the people have in our banking institutions, and will be the cause of
untold mischief.” With all the reports of banks selling bad investments,
wondered a Shriner from Memphis, “can one be blamed for putting his
funds in a lock box and hoarding?”86
The idea of banning commercial banks from selling securities thus
seems to have enjoyed widespread support for a few different reasons,
all of which had to do with the fear that the ability to sell securities was
encouraging banks to look after their own profits at the expense of their
customers. Of course, there was one interest group with the most to

85 “Banking, Group and Branch,” Fortune, Feb. 1930, 140; William Stark Smith
to Carter Glass, 27 Apr. 1932, CG, box 291; Anonymous to Carter Glass, 3
Dec. 1931, CG, box 301; John T. Flynn, “The Dangers of Branch Banking,”
Forum and Century 89 (1933), 261.
86 Garland A. Tunstill to Carter Glass, 26 Jan. 1932, CG, box 285; Leon Franks
to Carter Glass, 6 Feb. 1932, CG, box 284.
214

214 Speculation

gain from the measure—​the investment banks that were not affiliated
with a commercial bank. These banks, which were often called “private”
banks to distinguish them from the “national” commercial banks that
were under the jurisdiction of the Federal Reserve, had lost a big part
of their market share when the commercial banks entered the securities
business, and the separation of commercial banking from investment
banking promised to bring that market share back. Carter Glass could
hardly have been surprised to discover that among the heartiest sup-
porters of his banking bill were the investment bankers. “For years it
has been a great source of worry to the private banking houses that the
banks had, through their affiliates, gobbled up so large a part of the
issues of new securities to such a degree that the small houses were being
practically starved out of such business,” explained Waldo Kendall, an
investment banker in Boston. But Kendall could see better days around
the corner. “The fact is that all of this work has been done throughout
history up to within a few years ago by the private investment houses,
who stand ready and willing to do the necessary work,” he promised.87
The only significant opponents of the bill were, again not surpris-
ingly, the commercial banks. Their primary argument was that they
were needed, because the investment banks could not provide enough
capital to finance American business. “The long term capital market in
the United States has been developed in large part by member banks
and their securities affiliates,” insisted Charles McCain, the chairman
of the Chase National Bank, and William Potter, the president of the
Guaranty Trust Company of New York. They warned that forcing com-
mercial banks out of the securities business “will seriously impair the
ability of business in the future to get long term capital.” At its 1932
annual meeting, the American Bankers Association adopted a resolu-
tion emphasizing that “for many years national banks and state member
banks through their powers of underwriting and investing in securities
have supplied long-​term capital to many important industries,” and

87 Jonathan R.  Macey, “Special Interest Groups Legislation and the Judicial
Function: The Dilemma of Glass-​Steagall,” Emory Law Journal 33 (1984): 20;
E. B. Parsly to Carter Glass, 10 Feb. 1932, CG, box 285; J. F. B. Mitchell to
Carter Glass, 22 Nov. 1933, CG, box 311; Waldo S. Kendall to Carter Glass, 4
May 1932, CG, box 291.
  215

Aftershocks of the Crash 215

that “it has been the history of past depressions that recovery has largely
come about through the refinancing of industry and commerce in
which the member banks have played a material and important part.”
Fred Kent, a director of the Bankers Trust Company, testified that the
stock market would collapse even more than it already had if banks
like his were no longer allowed to sell securities. “It is not conceivable
that such a market could exist,” Kent reasoned, “if every natural chan-
nel through which funds are applied to maintain it was stopped up or
curtailed through legislation.”88
The weakness of this argument, as the investment bankers pointed
out, was that the commercial banks had been selling securities in large
quantities only for the previous few years. Before that, the business had
been in the hands of the investment banks, and no one had thought the
investment banks lacked access to sufficient capital. The commercial
banks had not added new capital to the system so much as they substi-
tuted themselves for the investment banks.
The commercial banks had a second argument against the Glass bill.
Commercial banks were regulated by the federal government, but the
private investment banks were not. “By divorcing the security busi-
ness from commercial banking,” the commercial banks contended,
“the former is turned over either to unregulated dealers, or to those
regulated only by such varying laws as the forty-​eight states may pro-
vide.” They suggested that a better plan would be to keep the securities
business with the federally regulated commercial banks, perhaps with a
new layer of regulation to prevent a recurrence of the abuses of the past
few years. “I think you will agree with me,” the Detroit banker E. F.
Connely wrote to Glass, “that it would be far better to have this busi-
ness conducted in connection with the banking business where it might
be rigidly administered than to have it revert to the hands of private
bankers answerable only to state governments subject to no examina-
tion and free to do about as they please.” But Glass did not agree, no

88 H.  H. Preston and Allan R.  Finlay, “Era Favors Investment Affiliates,”
American Bankers Association Journal 22 (1930): 1153; Charles S. McCain and
William C. Potter to Carter Glass, 29 Apr. 1932, CG, box 291; “Resolution of
American Bankers Association Upon Glass Banking Bill, S. 4412,” CG, box
296; Operation of the National and Federal Reserve Banking Systems, 512.
216

216 Speculation

doubt because the self-​interest of the commercial banks was so trans-


parent. “What has the American Bankers’ Association ever done in the
public interest?” scoffed another of Glass’s correspondents. “I have fol-
lowed the proceedings of their annual conventions for many years. The
trend of thought and action is in the interest of those managing the
banks, not the depositors or the public.”89
After the election of 1932, there was no longer any serious political
opposition to the separation of commercial from investment banking.
Carter Glass’s banking bill was incorporated into the Banking Act of
1933. “New Epoch in Banking Opens Under Glass-​Steagall Act” ran the
headline in the New York Times. Some of the biggest banks, seeing the
writing on the wall, had already divested themselves of their securities
affiliates, which after all had not been particularly profitable during the
Depression. Bankers Trust and the Manhattan Company had dropped
their affiliates in 1931. But most of the banks waited until near the June
1934 deadline. The First National Bank of Boston let go its affiliate,
the First Boston Corporation, which acquired Chase Manhattan’s old
affiliate. The Guaranty Trust Company, the National City Bank, and
the First National Bank all dissolved their affiliates. Most of the private
banks that had begun taking deposits stopped doing so and confined
themselves to investment banking. These included some of the firms
that would be among the most prominent investment banks for most
of the remainder of the century, including Kuhn, Loeb and Company;
Kidder, Peabody and Company; Lehman Brothers; and Goldman,
Sachs and Company. A  few of the private banks made the opposite
decision to become pure commercial banks, most notably J. P. Morgan
and Company, which spun off its securities business into a separate
firm that would be called Morgan Stanley.90 Investment banks and
commercial banks would remain separate for decades.

89 “Glass Banking Bill Passes the Senate,” Bankers’ Magazine 126 (1933): 210; E. F.
Connely to Carter Glass, 29 Mar. 1932, CG, box 293; T. J. Morrison to Carter
Glass, 24 Dec. 1932, CG, box 297.
90 New York Times, 25 June 1933, 3; “Wall Street Accepts New Deal in Finance,”
Literary Digest, 23 June 1934, 36; Chernow, House of Morgan, 384–​88; Carosso,
Investment Banking in America, 372–​75.
  217

Aftershocks of the Crash 217

The Glass-​Steagall Act was a major change in the structure of the


financial system, but it was small indeed compared with some of the
proposals that were seriously considered after the crash. The biggest
catastrophe in American economic history had largely been blamed on
speculation, but when the dust had settled, the law governing specula-
tion had scarcely changed.
6
Q
Land and Onions

One of the Broadway hits of 1925 was the Marx Brothers’ musical
The Cocoanuts. It opened during the Florida land boom, which served
as the show’s setting. Real estate prices in Florida were rising quickly.
Speculators were flocking south to buy and sell parcels of land. Irving
Berlin wrote the songs for The Cocoanuts, one of which, “Florida by the
Sea,” gently parodied the promotional literature for Florida that was
blanketing the rest of the country:

Down in the land where the trees are tall,


we’re asking you to pay a call
and live where it is summer all the year round.
Buy a lot,
any piece that we’ve got
will increase every season;
ask us why
everybody wants to lie
in the sun, there’s a reason.

By the time the film version of The Cocoanuts appeared four years
later, the bubble had burst. Florida land was selling for a small frac-
tion of 1925 prices. “Florida by the Sea” was moved from the middle
of the Broadway show to the opening credits of the movie, where it
took on a sharper satirical edge, as a reminder of the madness of a few
years before. Also acquiring a sharper edge were the words of Groucho
Marx, who plays a huckster trying to sell off land he knows is worthless.

218
╇ 219

Land and Onions 219

“All ye suckers who are gonna get trimmed, step this way for the big
swindle!” Groucho announces. “Friends, you are now in Cocoanut
Manor, one of the finest cities in Florida. Of course, we still need a few
finishing touches, but who doesn’t? This is the heart of the residential
district. Every lot is a stone’s throw from the station. As soon as they
throw enough stones, we’re going to build a station. Eight hundred
wonderful residences will be built right here. Why, they’re as good as
up. Better. You can have any kind of home you want to. You can even
get stucco—╉Oh, how you can get stuck-╉o!”
Everyone remembers the stock market boom and bust of the 1920s,
but only historians and film buffs remember that there was a nearly
simultaneous boom and bust in the real estate market, most vividly
in Florida, but not only there. The rise and fall of the stock market
gave rise to a national debate about limiting speculation in financial
assets, a debate that produced the New Deal financial regulation that
continues to structure the market today. But the rise and fall of the real
estate market did not give rise to an analogous debate or any analogous
regulation. And of course the real estate bubble of the 1920s was not
the first in American history, and it would not be the last. A similar
episode took place in the first decade of the twenty-╉first century, as real
estate prices soared upward, amidst considerable speculation, only to
plummet back down to earth. Once again, there would be considerable
discussion about speculation in financial assets, especially securities
backed by loans on real estate, but scarcely any debate about specula-
tion in real estate itself. Why the difference? Why have Americans been
so much more concerned about financial speculation than about real
estate speculation?

Florida€Frenzy
Land speculation has been a constant feature of American life since the
seventeenth century.1 So have periodic booms and busts, in both urban

1 John Frederick Martin, Profits in the Wilderness: Entrepreneurship and the Founding


of New England Towns in the Seventeenth Century (Chapel Hill:  University of
North Carolina Press, 1991); Alan Taylor, William Cooper’s Town:  Power and
Persuasion on the Frontier of the Early American Republic (New  York:  Vintage
Books, 1995); A. M. Sakolski, The Great American Land Bubble: The Amazing
220

220 Speculation

and rural areas. Recent studies suggest a common dynamic: as the pop-
ulation of an area grows, it takes some time to develop a corresponding
quantity of land. Land grows scarcer in the short run. Investors bid up
its price, but they underestimate the elasticity of the supply of land in
the long run. Eventually more land is developed and its price tumbles
back down.2
In the 1920s, older investors could still remember the real estate bub-
bles of the previous generation. Retail space in central Chicago that sold
for $1,500 per front foot in 1883 was worth $4,000 in 1889 and reached
$7,000 to $10,000 in 1891, only to drop back down in the 1890s. In
towns all over Nebraska, speculators poured millions into undeveloped
lots in 1886 and 1887, lured by easy credit and the hope that new rail-
roads would turn remote agricultural areas into bustling cities. A 640-​
acre farm near Omaha that had been bought several years earlier for
$2.50 an acre sold in 1887 for $1,000 an acre, and then a 120-​acre por-
tion of the farm sold a few months later for $2,500 an acre. Soon after,
prices were back where they had started. Similar bubbles blew up and
popped up all over the Midwest at around the same time, in Minnesota
and the Dakotas, in Kansas, in Texas, and in parts of Iowa, Wisconsin,
and Missouri. “The East invested vast sums in Western property,” one
observer recalled. “The rapid development of the resources of the West
lent plausibility to every reckless prophecy of higher prices; the con-
tinued inundation of Eastern money seeking chances of speculation
falsified the predictions of the foreboding.”3

Story of Land-​Grabbing, Speculations, and Booms From Colonial Days to the


Present Time (New York: Harper & Brothers, 1932).
2 Edward L.  Glaeser, “A Nation of Gamblers:  Real Estate Speculation and
American History,” American Economic Review: Papers & Proceedings 103, no. 3
(2013): 1–​42; Robert J. Shiller, “Historic Turning Points in Real Estate,” Eastern
Economic Journal 34 (2008): 1–​13.
3 Homer Hoyt, One Hundred Years of Land Values in Chicago: The Relationship of
the Growth of Chicago to the Rise of Its Land Values, 1830–1933 (1933) (Washington,
D.C.: Beard Books, 2000), 175–​81; Herbert L. Glynn, “The Urban Real Estate
Boom in Nebraska During the Eighties,” Nebraska Law Bulletin 6 (1928):
473–​76; Henry J. Fletcher, “Western Real Estate Booms, and After,” Atlantic
Monthly 81 (1898): 689.
  221

Land and Onions 221

Perhaps the sharpest rise and fall in real estate prices during the
late nineteenth century took place in Los Angeles during the boom
of 1887. When artesian well water began to be used for irrigation, and
the coming of the Southern Pacific Railroad introduced competition
between railroad lines and thus lower fares, emigrants and speculators
swarmed into southern California. Sixty new townsites were put on
the market in 1887 alone, most of which consisted of undeveloped land
in the middle of nowhere, bordered only by other equally imaginary
townsites. Some, including Claremont, actually got built, but most
would never exist except on paper. Land was typically purchased with
small down payments. Speculators bought and sold options to buy
land, and an option on a particular parcel often changed hands several
times before it expired. Prices skyrocketed. A thirty-​two-​acre parcel just
south of what is now downtown Los Angeles that sold in 1885 for $8,500
was resold in 1887 for $40,000, to be subdivided. In Hollywood, land
worth $100 per acre in 1886 was worth $600 per acre a year later. In
Boyle Heights, parcels that had been worth $150 only a few years earlier
sold in 1887 for $10,000. By the end of the year, the bubble had burst.
Many of the lots purchased at fabulous prices were abandoned by the
buyers as worthless.4
It happened again in the 1920s. As the stock market rose to new
heights in the middle of the decade, so did the real estate market in
much of the country. National real estate prices peaked in 1926 and
dropped sharply thereafter. In New  York, real estate prices closely
tracked stock prices, peaking in 1929 and then declining rapidly be-
tween 1929 and 1932. The boom of the 1920s gave rise to considerable
speculation in real estate and also in a new kind of financial asset, bonds
backed by real estate, the forerunners of the mortgage-​backed securi-
ties that would play a big part in the boom of the early 2000s. By the
mid-​twenties, issuances of real estate bonds constituted nearly a quarter
of all corporate debt issues, an even larger share than bonds issued by
railroads. Just as contemporaries worried that there was too much stock
speculation in the mid-​1920s, they worried that there was too much real

4 Joseph Netz, “The Great Los Angeles Real Estate Boom of 1887,” Annual
Publication of the Historical Society of Southern California 10 (1915–​16): 54–​68.
222

222 Speculation

estate speculation. In 1925, for example, the economist A. C. Miller, a


member of the Federal Reserve Board, urged the bankers of Boston to
be more conservative in making loans for the speculative purchase of
land. “There is evidence,” Miller argued, “that a section of the public
is losing its bearings and being drawn into the arena of thoughtless
speculation.”5
Nowhere was real estate speculation more spectacular than in
Florida. In Miami, subdivisions that existed only on paper sold out the
first day they were offered, most to investors who resold within days,
before they had to put any money down. It was hard to walk in down-
town Miami because the sidewalks were clogged with realtors. In Palm
Beach, reported Harper’s magazine, “conversation was full of stories of
houses built for $5,000 and sold for $18,000; of sea-​front lots bought at
$20 a foot, and now worth $700.” As one correspondent put it in late
1925, “the smell of money in Florida, which attracts men as the smell
of blood attracts a wild animal, became ripe and strong last spring.
The whole United States began to catch whiffs of it.” The result was a
“Florida frenzy.”6

5 Eugene N. White, “Lessons From the Great American Real Estate Boom and
Bust of the 1920s,” NBER Working Paper 15573 (2009); Alexander James Field,
“Uncontrolled Land Development and the Duration of the Depression in the
United States,” Journal of Economic History 52 (1992): 785–​805; Tom Nicholas
and Anna Scherbina, “Real Estate Prices During the Roaring Twenties and
the Great Depression,” Real Estate Economics 41 (2013): 278–​309; William N.
Goetzmann and Frank Newman, “Securitization in the 1920s,” NBER Working
Paper 15650 (2010); New York Times, 18 Nov. 1925, 36.
6 William Frazer and John J. Guthrie Jr., The Florida Land Boom: Speculation,
Money and the Banks (Westport, Conn.: Quorum Books, 1995); Mark S. Foster,
Castles in the Sand:  The Life and Times of Carl Graham Fisher (Gainesville:
University Press of Florida, 2000); Raymond B.  Vickers, Panic in Paradise:
Florida’s Banking Crash of 1926 (Tuscaloosa: University of Alabama Press, 1994);
Homer B. Vanderblue, “The Florida Land Boom,” Journal of Land & Public
Utility Economics 3 (1927):  118–​19; Paul S.  George, “Brokers, Binders, and
Builders: Greater Miami’s Boom of the Mid-​1920s,” Florida Historical Quarterly
65 (1986):  34; W.  L. George, “Humanity at Palm Beach,” Harper’s Monthly
Magazine 150 (1924): 221; Gertrude Mathews Shelby, “Florida Frenzy,” Harper’s
Monthly Magazine 152 (1925): 177.
  223

Land and Onions 223

While Florida land prices were on the rise, there was no shortage of
sober justifications for them. “The bubble will not burst for the very good
and sufficient reason that there is no bubble,” insisted Elmer Youngman
in early 1926, just before the bubble burst. Youngman, the editor of the
Bankers’ Magazine, declared that “some of the ablest men in America
are pouring millions into Florida.” He attributed the new values being
placed on Florida real estate to the increased prosperity of the country,
which gave the masses the means to enjoy leisure and travel, and to the
invention of the automobile, which gave them an easy way to reach a
warm climate. The journalist Frank Stockbridge emphasized the recent
drainage of Florida’s swamps, which eliminated the mosquito and the
attendant threats of malaria and yellow fever. Others simply liked the
weather. “One may swim in the surf, pick strawberries and wear roses in
December,” marveled one correspondent. “You can go fishing, boating
and surf bathing when the waters of the north are covered with ice.”7
Florida’s real estate prices came crashing down in 1926. The New York
Times summed up the situation at the year’s end: there was “a large class
of investors who, having made payments ranging from a few hundred
to several thousand dollars, now have nothing to show for their money
except parcels of waste land.” Banks all over the state failed within a few
months, after making real estate loans that now had no hope of being
repaid.8 The banks and the speculators alike were, as Groucho Marx
put it, stuck-​o.
By the early 1930s, after similar if less colorful episodes in many
cities, there was a general recognition that real estate speculation had
gotten out of hand in the previous decade. “Real estate values, we were
told, could never go down—​only up. You simply could not lose out
on any real estate investment,” the North American Review recalled.
“Therefore money poured into the real estate field.” Chicago was awash
in foreclosures because of overspeculation, lamented the vice president

7 Elmer H. Youngman, “Florida: The Last Pioneer State of the Union,” Bankers’


Magazine 112 (1926): 14, 7–​8; Frank Parker Stockbridge, “The Florida Rush of
1925,” Current History 23 (1925):  181–​82; J.  D. Burke, “Florida–​The Material
Progress From 1912 to 1926,” Lawyer & Banker and Southern Bench and Bar
Review 19 (1926): 129–​30.
8 New York Times, 3 Jan. 1927, 3; 16 Mar. 1927, 24.
224

224 Speculation

of the Chicago Title and Trust Company. Land speculation had left
“wreckage” all over the New  York area, a regional board concluded,
in the form of hundreds of thousands of undeveloped lots whose pur-
chasers had defaulted on their loans. The economist Herbert Simpson
observed that local governments were victims as well. Most of their
revenue came from property taxes, and many had embarked on ambi-
tious public works based on the assumption that property values would
continue to rise. As the Atlantic magazine despaired in 1931, the nation
was now reaping the “whirlwinds of speculation.”9
Analogous events in the securities markets gave rise to an extensive
debate in the late 1920s and early 1930s about how to regulate the mar-
kets to prevent another crash. There was no parallel debate concerning
real estate. Occasionally, to be sure, someone would call for greater reg-
ulation. During the midst of the Florida land boom, for example, at the
annual meeting of the Investment Bankers Association, the Chicago
banker Walter Greenbaum proposed a national blue sky law for real
estate. Greenbaum had become exasperated with the glowing public
relations material touting land in Florida. He wanted a federal agency
to screen out the bad investments before real estate could be offered
to the public, just like the first wave of state blue sky commissions
screened out unsound stock offerings.10 This idea went nowhere. And
such calls for regulation of the real estate market seem to have been
exceedingly rare. Certainly they were much rarer than calls to regulate
the stock market.

The Only Hope Here Is Increased Caution


One obvious form of regulation that would have deterred real estate
speculation, if such deterrence had been desired, was a tax on real estate

9 Gerhard Hirschfeld, “A Morgue of Mortgages,” North American Review 232


(1931): 255; Kenneth E. Rice, “A Study of Real Estate Cycles,” Lawyer & Banker
and Central Law Journal 24 (1931):  254; New  York Times, 11 Feb. 1929, 14;
Herbert D.  Simpson, “Speculation and the Depression,” American Economic
Review: Papers and Proceedings 23 (1933): 165–╉66; Samuel Spring, “Whirlwinds
of Speculation,” Atlantic, 1 Apr. 1931.
10 New York Times, 23 Oct. 1925, 31.
  225

Land and Onions 225

transfers. In the early 1930s, there were many calls to tax stock transfers
for the same purpose, to deter speculation in stock. In more recent
times, a few jurisdictions imposed taxes on real estate transfers in an
effort to dampen real estate speculation. Vermont established such a tax
in 1973, for example, and Ontario did the same in 1974. In earlier eras,
frontier residents had clamored for special property taxes on absentee
land speculators, and the available evidence suggests that at least some
jurisdictions satisfied that demand indirectly, by assessing absentee-​
owned property at higher values than comparable property owned
by residents. Taxing land speculation to reduce its quantity was thus
well within the realm of possibility in the late 1920s and early 1930s.
D. E. French, a West Virginia lawyer, suggested in 1931 that the federal
government could curtail real estate speculation by imposing a special
income tax on the profits of sales of real estate held for less than twelve
months. Booms were always followed by depressions, French pointed
out, so “if we can find a way to eliminate booms we will thereby elim-
inate the depression.”11 But no such tax was seriously considered by
Congress or any state legislature.
There was a debate about real estate taxation during the period, one
that bore tangentially upon speculation, but it had little to do with the
boom and bust of the 1920s. In the late nineteenth century the political
economist Henry George had proposed his famous “single tax,” a tax
on increases in the value of land. The primary reason George favored
this tax was to reduce inequality, but, he contended, the single tax had
the additional advantage of deterring real estate speculation. “That land
speculation is the true cause of industrial depression is, in the United
States, clearly evident,” George argued. He thought his tax would “tend
to increase production, by destroying speculative rent,” because “if land
were taxed to anything near its rental value, no one could afford to
hold land that he was not using, and, consequently, land not in use

11 R. Lisle Baker and Stephen O. Andersen, “Taxing Speculative Land Gains: The


Vermont Experience,” Urban Law Annual 22 (1981):  3–​ 69; Lawrence
B. Smith, “The Ontario Land Speculation Tax: An Analysis of an Unearned
Increment Land Tax,” Land Economics 52 (1976): 1–​12; Robert P. Swierenga,
“Land Speculation and Frontier Tax Assessments,” Agricultural History 44
(1970): 253–​66; D. E. French to Carter Glass, 28 Apr. 1931, CG, box 278.
226

226 Speculation

would be thrown open to those who would use it.” By the 1920s, this
claim does not appear to have been widely believed, at least among
economists and those with experience collecting taxes. But there was
always a small cohort of ardent Georgists who kept the faith, like the
economist Harry Gunnison Brown, who still argued that the single
tax really would discourage speculation.12 The debate over the single
tax had mostly petered out by the 1920s, however, and the effect of the
single tax on speculation was, in any event, a minor issue within that
debate.
Why were contemporaries so much more interested in limiting spec-
ulation in securities than in real estate? Any answer to this question re-
quires some guessing, because there does not appear to have been much
discussion of the question at the time, but it is not hard to identify five
related reasons for the disparity. None provides a full explanation, but
together they can explain why there was no new regulation for land
comparable to the New Deal limits on speculation in securities.
First, while securities were intangible, man-​made, and a bit mys-
terious to the average person, land was a physical, natural thing with
qualities that were easily understood. Many more people owned land
than securities, and while most people had never seen securities and
may have had only a vague sense of what they were, everyone had seen
land and knew exactly what it was. The buying and selling of land was a
well-​known part of everyday life, from the biggest cities to the smallest
towns. The same was not true of securities.
Second, speculators in securities engaged in all sorts of arcane and
unnatural-​seeming transactions. They made short sales, of items they
did not even own, to profit from price declines, which gave some
speculators the unneighborly incentive to make their fellows poorer.

12 Henry George, Progress and Poverty:  An Inquiry Into the Cause of Industrial
Depressions and of Increase of Want with Increase of Wealth (1879) (San
Francisco:  National Single Tax League, 1905), 266, 411; Adam Shortt,
“Municipal Taxation in Relation to Speculative Land Values,” Annals of the
American Academy of Political and Social Science 58 (1915): 214–​21; J. Hamilton
Ferns, “Single Tax in Theory and Practice,” Bulletin of the National Tax
Association 5 (1919): 75–​80; Harry Gunnison Brown, “Land Speculation and
Land-​Value Taxation,” Journal of Political Economy 35 (1927): 390–​402.
  227

Land and Onions 227

They entered into futures contracts, for things they might never even
see. They hedged against price movements by buying and selling at
the same time. The repertoire of transactions in real estate was much
smaller. There were no short sales, so there was no way to profit from
a decline in real estate prices, and there were no real estate speculators
hoping for a crash. There were no futures and there was no hedging.
Real estate speculators could buy options, and in the bubblier loca-
tions, like Florida, options were bought and sold in their own right,
but this was small stuff compared with the transactions on the stock
and commodity exchanges. The main way to speculate in real estate was
straightforward: buy land and hope it rose in price so it could be sold
to someone else. The transactions in securities that seemed artificial and
even pernicious had no parallels in real estate.
Third, real estate seemed like a more level playing field than securities
for the average investor. An ordinary person venturing into the stock
market was competing with corporate insiders, who knew much more
about their own firms than anyone on the outside possibly could. And
he was competing with professionals who were widely suspected to have
tricks up their sleeves that allowed them to manipulate prices, at least
in the short run, to fleece the amateurs. Not so with land. “Generally
speaking, the prudent purchase of land is a better investment for the or-
dinary man than stocks and bonds, because in the former case he does
not pit his judgment against the machinations of a board,” advised the
economist Richard Ely. “For example, you buy some stock. You know
some of the facts with regard to the stock, but you cannot know all the
facts which are known to the president and the board of directors, even
supposing they are honest and do not want to swindle anyone. But, if
they do not want to be honest, the buyer is playing with dice loaded
against him.”13 Real estate was more transparent. Its value might fluctu-
ate, but the average investor could see why.
Fourth, crashes in securities markets affected the country as a whole,
all at once, while crashes in real estate markets, although often corre-
lated, were local events. When the Florida land bubble burst in 1926,
it harmed only people who had bought land in Florida. When the

13 Richard T. Ely, “Land Speculation,” Journal of Farm Economics 2 (1920): 123.


228

228 Speculation

stock market crashed three years later, it harmed everyone. Real estate
downturns were less likely than stock market downturns to mobilize
political support for regulation.
Finally, with real estate, there was no interest group that was always
on one side of a transaction. Anyone who was a buyer of land one day
could be a seller of land another day; there was no group of people who
were always one or the other. That made land very different from com-
modities, where farmers were always sellers and their customers were
always buyers. Arguments that speculation pushed prices up or down
had a sharp impact for commodities, because if those arguments were
true, there would be clear winners and losers. Not so with real estate,
where if speculation pushed prices up, any given participant would gain
as a seller whatever he lost as a buyer, and vice versa if speculation
pushed prices down. Corporate stock was somewhere in the middle.
Anyone who bought stock one day might sell it another day, so in that
sense stock was like land. But when new stock was issued, corporate
insiders were always sellers, and members of the general public were
always buyers. In that sense, stock was like commodities. All else equal,
one would expect interest groups to form most readily to regulate
speculation in commodities, least readily to regulate speculation in real
estate, and somewhere in the middle to regulate speculation in stock.
Real estate speculation thus seemed more natural and less dangerous
than speculation in securities. “Real-​estate booms probably will con-
stantly recur,” the Atlantic observed in 1931, “because the real-​estate
men involved—​drifters from other industries—​naturally will cooperate
with local interests to sell through high-​pressure sales methods as many
lots as the public can possibly be induced to buy.” But the magazine did
not think any regulation was needed to prevent the next bubble from
inflating and bursting. “The only hope here is increased caution, born
of experience, on the part of the public,” the Atlantic concluded, “and
the influence of the established and more experienced local real-​estate
firms in checking boom tendencies.”14 With land, unlike with the stock
market, contemporaries perceived nothing for regulators to do.

14 Spring, “Whirlwinds.”
╇ 229

Land and Onions 229

An Onion€Ring
To gain some perspective on the lack of interest in limiting speculation
in real estate, it is useful to look at a very different market, one in which
Congress, in relatively short order, completely banned speculation. In
the Onion Futures Act of 1958, Congress made it a crime to enter into
a contract for the sale of onions for future delivery.15 To this day, onions
are the one agricultural commodity in which futures trading is illegal.
Onion futures were relatively new in 1958. The Chicago Mercantile
Exchange began trading onion futures in 1942, and the New  York
Mercantile Exchange followed suit in 1946. By the mid-╉1950s, the
Chicago exchange hosted nearly all of the transactions, in what was still
a tiny market compared with more established commodity futures like
wheat or corn.16
Then disaster struck—╉disaster from the perspective of an onion
farmer. In August 1955, a fifty-╉pound bag of onions was worth $2.75. By
March 1956, the same onions were worth 10 cents, less than the cost of
the bag. It turned out that two onion dealers, Sam Siegel of Chicago and
Vincent Kosuga of New York, had outmaneuvered the onion market.
In the fall of 1955, Siegel and Kosuga bought virtually the entire onion
crop available on the Chicago market. Once they had the onions, they
threatened the leading onion growers that they would flood the market,
and depress the price, unless the growers bought about a third of the
onions back from them at high prices. The growers agreed to buy a
third of the onions back. Siegel and Kosuga then double-╉crossed the
growers. They sold onions short on the Chicago Mercantile Exchange,
to the point where they acquired short positions that exceeded the
amount of onions they had originally purchased. They then flooded the
market with the remaining two-╉thirds of the onions. They even shipped
some of their onions out of town, repackaged them, and shipped them
back to Chicago to make it appear as if still more new onions were
coming on the market. Onion prices fell through the floor. The growers

1 5 72 Stat. 1013 (1958).


16 Emily Lambert, The Futures: The Rise of the Speculator and the Origins of the
World’s Biggest Markets (New  York:  Basic Books, 2011), 33–╉44; Prohibiting
Futures Trading in Onions, H.R. Rep. No. 1036, 85th Cong., 1st Sess. (1957).
230

230 Speculation

were left holding very expensive onions, while Siegel and Kosuga made
millions on their short positions.17
Onion farmers were devastated by the fall in prices. Within a few
months, the Commodity Exchange Authority (the agency within the
Department of Agriculture that administered the Commodity Exchange
Act of 1936) set sharp limits on trading in onion futures. Under the new
rules, no trader could hold net long or short positions of more than two
hundred car lots of onions, about a sixth of the short position accumu-
lated by Siegel and Kosuga.18 But this was not enough for the onion
farmers, who urged Congress to ban the futures trade completely. Their
testimony before Congress demonstrated that the market in onions was
very different from the market in real estate in each of the respects that
made limiting speculation in real estate so unlikely.
Onion farmers, unlike real estate investors, were a tightly focused
interest group who were always on the same side of the transaction.
They sold onions and never bought them. Their livelihood thus de-
pended on keeping onion prices up. Anything that was perceived to
depress onion prices was a menace. Farmers had long believed that
commodity speculation depressed prices, and the events of 1955–​56
had confirmed for the onion farmers that this was indeed the case.
There was no comparable interest group on the other side of the trans-
action. Lower onion prices were good for consumers, who were always
onion buyers and never onion sellers, but of course onions made up
a negligible portion of their food budget, so they had little reason to
lobby for lower prices.
Onions, unlike real estate, traded in a national market. A fall in prices
affected onion farmers everywhere. And onions were grown all over the
country. Letters poured into the Senate from onion growers in Oregon
and Wisconsin, Texas and Michigan, Idaho and New York. The growers
were represented by the National Onion Association, which gained the
support of larger agricultural organizations like the Vegetable Growers
of America and the National Farm Bureau, who had members in all

17 “Odorous Onions,” Time, 2 July 1956, 70; Chicago Daily Tribune, 22 June
1956, B7.
18 Wall Street Journal, 23 July 1956, 15.
  231

Land and Onions 231

but the most urban congressional districts.19 The Florida real estate bust
was well known throughout the country, but it personally affected only
those who had bought land in Florida. The great onion bust was prob-
ably less famous, outside the onion trade. It did not serve as the setting
for any Marx Brothers movies or Irving Berlin songs. But it affected
onion growers nationwide.
Onions, like land, were an easily understood physical commodity,
but onion futures were not. Most onion farmers did not hedge, testified
Veril Baldwin of Stockbridge, Michigan, the president of the National
Onion Association. “Onions are not processed or used in manufactur-
ing to any extent,” he explained, “so buying to hedge for future use is
almost nonexistent,” and “the average grower cannot successfully use
the market to hedge his sales.” The onion farmers had little or no expe-
rience with onion futures, which were primarily the province of specu-
lators. Onion futures, sneered the Texas grower Austin Anson, were
traded by “smart boys who had never owned an onion in their lives.”20
And those smart boys were engaging in unnatural-​seeming transac-
tions that seemed to flout the laws of supply and demand, by causing
the price of onions to fluctuate wildly when neither the supply of onions
nor the demand for them was changing. “It is only in a futures market,”
Baldwin complained, that a person “can make money by depressing
the market, and that is what is the matter with it.” John Hardwicke,
another Texas grower, emphasized how little the onion farmers were
asking for. They were “not asking for a handout or subsidy, not asking
for price supports,” he noted. The industry was “simply asking for its
God-​given right to conduct its own free business according to the laws
of supply and demand without some speculators being in a position to
often make a lot of money by scheming up to demoralize and ruin the
onion growers’ market.”21

19 Onion Futures Trading: Hearing Before a Subcommittee of the Committee on


Agriculture and Forestry, United States Senate (Washington, D.C.: Government
Printing Office, 1957) (“1957 Hearings”), 2–​6, 23–​24.
20 Onion Futures Trading: Hearings Before the Committee on Agriculture and
Forestry, United States Senate (Washington, D.C.: Government Printing
Office, 1958) (“1958 Hearings”), 141–​43; 1957 Hearings, 61.
21 1958 Hearings, 143, 168.
232

232 Speculation

Onion futures were thus a much more likely target of antispecula-


tive regulation than real estate. The farmers’ case was not very strong
from an economic perspective. Onion prices were actually more stable
during the period in which onion futures were traded on the Chicago
Mercantile Exchange than they had been before or would be after. As
the economist Roger Gray pointed out, wide swings in prices were in-
dicative of too little speculation in onions rather than too much, be-
cause a thicker market in onion futures would have dampened price
movements in both directions. A thicker market would also have been
harder for Siegel and Kosuga to manipulate. But this kind of argu-
ment had little effect on Congress, where the prevailing view was that
trading in onion futures was an injustice to the farmers. The farmer
did all the work, reasoned the Michigan congressman Clare Hoffman,
but “it’s those Chicago birds, those city slickers, who get the money.”22
The point of banning onion futures was to give the money back to the
farmers.

The Subprime€Crisis
Real estate speculation returned to the headlines in the first decade
of the twenty-╉first century, which saw the biggest rise and fall in real
estate since the 1920s. Nationally, home prices more than doubled
between 1995 and 2007. As in the 1920s, rising prices attracted pur-
chasers who hoped to own real estate only briefly before selling it on
to someone else. By 2005, according to the National Association of
Realtors, 28 percent of the people buying homes were seeking invest-
ments rather than places to live. Perhaps the best barometer of the
public interest in real estate speculation was the proliferation between
2005 and 2007 of reality television programs devoted to “flipping”
houses, in which purchasers with no real estate experience were de-
picted as making quick profits by speculating in single-╉family homes.

22 Roger W. Gray, “Onions Revisited,” Journal of Farm Economics 45 (1963): 273–╉


76; Roger W. Gray, “Some Current Developments in Futures Trading,” Journal
of Farm Economics 40 (1958):  344–╉51; Washington Post and Times Herald, 14
Mar. 1958, A19.
  233

Land and Onions 233

Lenders extended credit to homebuyers who had no prospect of ever


repaying their loans unless the value of their homes rose enough so
they could borrow even more a few years later. But the bubble burst in
2007. Within three years, national housing prices fell by a third. And
these were national averages; in some cities, prices rose and fell much
more sharply. By 2010, four million homes were in foreclosure, and
several million more foreclosures seemed to be on the way. Nearly a
quarter of the surviving borrowers owed more on their mortgages than
their homes were worth.23
Unlike previous real estate bubbles, this one nearly brought down
the entire financial system, because of innovations in the ways that
money flowed from lenders to borrowers. Mortgages existed at least
as early as the nineteenth century, and in the later part of the century
eastern financial institutions began investing in western farm mort-
gages. But the mortgage market remained very small by modern stan-
dards until the middle of the twentieth century. During the real estate
boom of the 1920s, the main mortgage lenders were banks, insurance
companies, and building and loan associations, cooperative organiza-
tions in which members would pool their funds to lend to members
buying homes. Mortgages were not the norm; only around 40 percent
of owner-​occupied housing units had a mortgage. And these mort-
gages were stingy by modern standards. They typically covered no more
than half of a home’s value, lasted only three to five years, and left the
homeowner owing a large lump sum at the loan’s expiration. In the
1920s, many of these loans were securitized—​broken into small pieces
and sold to investors—​but these were relatively simple transactions, in
which investors essentially bought shares of the mortgage for a single

23 National Association of Realtors survey quoted in Todd J. Zywicki and Joseph


D. Adamson, “The Law & Economics of Subprime Lending,” University of
Colorado Law Review 80 (2009): 33; New York Times, 18 Mar. 2007, F90; S&P/​
Case-​Shiller U.S. National Home Price Index, http://​us.spindices.com/​index-​
family/​real-​estate/​sp-​case-​shiller; The Financial Crisis Inquiry Report: Final
Report of the National Commission on the Causes of the Financial and Economic
Crisis in the United States (Washington, D.C.: Government Printing Office,
2011), 402, 23.
234

234 Speculation

building (usually a large apartment building or commercial building)


rather than for an assortment of properties.24
The collapse of the real estate market and the failure of many lenders
in the late 1920s and early 1930s gave rise to new government insti-
tutions intended to encourage mortgage lending. The Federal Home
Loan Bank Act of 1932 established regional home loan banks, which
could borrow from the government and issue tax-​free bonds to raise
money that could be lent to homeowners. The Home Owners’ Loan
Act of 1933, part of the flurry of New Deal legislation in the first hun-
dred days of the Roosevelt administration, created the Home Owners’
Loan Corporation, which used government funds to refinance exist-
ing mortgages that were near foreclosure. A  year later, the National
Housing Act of 1934 established the Federal Housing Administration
as an insurer of mortgages. In 1938, the Housing Act was amended
to establish another federal agency, the Federal National Mortgage
Association (“Fannie Mae”), for the purpose of buying mortgages from
private lenders. Federal government support transformed housing fi-
nance. By the middle of the century, the standard mortgage lasted up
to thirty years, covered up to 80 percent of the home’s value, and left
the homeowner owing nothing when it came to an end. With the avail-
ability of these mortgages, homeownership rates rose from 44 percent
in 1940 to 64 percent by 1980.25
The pervasive securitization of mortgages has its roots in the 1968
reorganization of government support of the mortgage market, when
Fannie Mae was converted into a formally private firm, and the
Government National Mortgage Association (“Ginnie Mae”) was cre-
ated as a federal agency to securitize the mortgages insured by the
Federal Housing Administration. The Emergency Home Finance Act

24 Jonathan Levy, Freaks of Fortune: The Emerging World of Capitalism and Risk in


America (Cambridge, Mass.: Harvard University Press, 2012), 150–​90; Kenneth
A. Snowden, “The Anatomy of a Residential Mortgage Crisis: A Look Back to
the 1930s,” NBER Working Paper 16244 (2010).
25 Richard K.  Green and Susan M.  Wachter, “The American Mortgage in
Historical and International Context,” Journal of Economic Perspectives 19
(2005): 93–​114; Kent W. Colton, “Housing Finance in the United States: The
Transformation of the U.S. Housing Finance System,” Harvard University
Joint Center for Housing Studies, W02-​5 (2002).
  235

Land and Onions 235

of 1970 established another nominally private firm, the Federal Home


Loan Mortgage Corporation (“Freddie Mac”), to purchase mortgages.
Before long, Freddie Mac and Fannie Mae began securitizing mort-
gages just like Ginnie Mae did. By the 1990s, both organizations were
securitizing most of the mortgages they purchased—​that is, they were
selling to investors bonds that were backed by the flow of payments
from homeowners. Private companies (typically investment banks)
unaffiliated with the government soon began securitizing mortgages
that were not guaranteed by Fannie Mae, Freddie Mac, or Ginnie
Mae, a development encouraged by the Secondary Mortgage Market
Enhancement Act of 1984, which removed regulatory obstacles to such
“private label” securitization. All this securitization allowed homebuy-
ers to tap into vast sums of money that had previously been unavail-
able for mortgage lending, which almost certainly caused more money
to be borrowed, at lower interest rates, than would otherwise have
been possible.26
Securitization also increased the number and size of the participants
in the financial system who depended on homeowners paying their
mortgages. Previous downturns in the housing market had bankrupted
mortgage lenders, but now there were, in effect, many more mortgage
lenders than ever before. The number of actors potentially exposed
to mortgage defaults multiplied in the first decade of the twenty-​first
century, when banks began assembling collateralized debt obligations
(CDOs)—​slices of pools of mortgage-​backed securities—​which were
then sold to other investors. That exposure multiplied again with the
development of credit default swaps (essentially insurance against de-
faults on CDOs), and then synthetic CDOs, which were slices of pools
of the credit default swaps. By the time the housing bubble burst, an
enormous financial superstructure had been built on a rickety base of
borrowers’ ability to make their mortgage payments.
This was why the downturn in real estate prices that began in
2007 had such a dramatic effect on the financial system. Major finan-
cial institutions were more tightly connected to the housing market,

26 Congressional Budget Office, Fannie Mae, Freddie Mac, and the Federal Role
in the Secondary Mortgage Market (2010), S1–​S4, 4–​7.
236

236 Speculation

through the securitization of mortgages and the reassembly of the re-


sulting bonds into various flavors of derivatives, than they had been in
earlier episodes of real estate speculation. The Florida land bust of 1926
wiped out some Florida banks, but the nationwide land bust of 2007
nearly wiped out the largest financial institutions in the country.
Before one could think intelligently about how to prevent such a
catastrophe from occurring again, one had to understand why real
estate prices had risen and fallen so sharply. On this question there
was, unsurprisingly, a wide range of opinion. Some blamed land use
regulation that limited the supply of buildable land, thus forcing up
its price. If such regulation was indeed the disease, the cure would be
to have less of it. Then again, while land use regulation was a plausible
explanation for why land was more expensive in some places than in
others, it was far less plausible as an explanation for the sharp rise in
real estate prices (because regulation had not become any more intense
while prices were on the way up), and it could scarcely explain why
real estate prices had plummeted (because regulation had not become
any less intense during the fall). Others, especially in the immedi-
ate aftermath of the financial crisis, blamed government policies that
encouraged homeownership in less affluent communities, which they
contended led to the deterioration of lending standards. But more
careful study afterward revealed that the homeowners in the com-
munities in which these policies were implemented were actually less
likely to default on their loans.27
A more persuasive set of explanations focused on the rapid expan-
sion of the practice of securitizing mortgages. The housing bubble co-
incided with the influx of capital from the “private label” securitization

27 Edward L. Glaeser et al., “Housing Supply and Housing Bubbles,” Journal of


Urban Economics 64 (2008): 198–​217; Randal O’Toole, “How Urban Planners
Caused the Housing Bubble,” Policy Analysis, no. 646 (2009); Thomas Sowell,
The Housing Boom and Bust (New York: Basic Books, 2009); R. Christopher
Whalen, “The Subprime Crisis—​Cause, Effect and Consequences,” Indiana
State University Networks Financial Institute Policy Brief 2008-​PB-​04 (2008);
Peter J.  Wallison and Arthur F.  Burns, “Dissenting Statement,” Financial
Crisis Inquiry Report, 443–​533; Robert B. Avery and Kenneth P. Brevoort, “The
Subprime Crisis:  Is Government Housing Policy to Blame?” (2011), http://​
ssrn.com/​abstract=1726192.
  237

Land and Onions 237

of mortgages that were too risky to be insured or securitized by the gov-


ernment or by the nominally private government-​sponsored entities.
Securitization reduced lending standards, as a greater supply of money
became available for lending. Borrowers took on more debt than they
had a rational hope of repaying, in part because many of them did not
fully understand the terms of their loans, and in part because unscru-
pulous lenders took advantage of their ignorance by steering them into
loans at higher interest rates than they could otherwise have obtained.
Many of the originators of these loans had little incentive to inquire
into the creditworthiness of the borrowers, either because they were
mortgage brokers who were compensated per loan regardless of repay-
ment or because the securitization process allowed them to pass the risk
of default on to investors.28
These institutional changes had been facilitated by legal changes. It
was lawful to make such subprime loans because of the liberalization
of federal banking laws in the early 1980s. The Depository Institutions
Deregulation and Monetary Control Act of 1980, the statute that re-
moved interest rate caps on bank accounts, also deprived the states of
their traditional power to limit the interest rates that mortgage lend-
ers could charge. The availability of higher rates made lenders more
willing to extend credit to riskier borrowers. The Garn–​St. Germain
Depository Institutions Act of 1982 permitted “alternative mortgage
transactions,” including loans with adjustable rates and balloon pay-
ments.29 These became common features of subprime loans.

28 Atif Mian and Amir Sufi, “The Consequences of Mortgage Credit Expansion:
Evidence From the U.S. Mortgage Default Crisis,” Quarterly Journal of
Economics 124 (2009): 1449–​96; Benjamin J. Keys et al., “Did Securitization
Lead to Lax Screening? Evidence From Subprime Loans,” Quarterly Journal
of Economics 125 (2010): 307–​62; Oren Bar-​Gill, “The Law, Economics and
Psychology of Subprime Mortgage Contracts,” Cornell Law Review 94 (2009):
1073–​151; Sumit Agarwal et al., “Predatory Lending and the Subprime Crisis,”
NBER Working Paper 19550 (2013); Alan M. White, “The Case for Banning
Subprime Mortgages,” University of Cincinnati Law Review 77 (2008): 617–​44;
Antje Berndt et al., “The Role of Mortgage Brokers in the Subprime Crisis,”
NBER Working Paper 16175 (2010).
29 Cathy Lesser Mansfield, “The Road to Subprime ‘HEL’ Was Paved With
Good Congressional Intentions: Usury Deregulation and the Subprime Home
Equity Market,” South Carolina Law Review 51 (2000):  473–​ 587; Patricia
238

238 Speculation

Of course, none of these factors could have brought about the


financial crisis without the investors who were willing to buy the
mortgage-​backed securities and CDOs that were made up of sub-
prime mortgages. These were sophisticated investors, including some
of the large banks that would be bailed out by the government when
these investments went sour. Why were they so eager to buy securities
that rested on such a flimsy base of mortgages? There are two likely
answers to this question.
First, while it might not have been difficult to estimate the value of
any single loan, the outputs of securitization were pools consisting of
thousands of heterogeneous loans. It was not a simple matter to figure
out what they were worth. Assessing their value required constructing
models of the probability that given numbers of individual borrowers
would default, but these securities were so new that there was not much
historical data against which to test the models, and the existing data
were mostly from the recent past, when real estate prices were consis-
tently rising. Mortgage-​backed securities were rated by the credit-​rating
agencies, but in retrospect it is clear that these agencies did a very poor
job, perhaps in part because they were compensated by the issuers and
so had an incentive to downplay the risk. As a result, even sophisticated
investors seem to have underestimated the risk of these securities and
thus paid too much for them.30
Second, the human psychology of this bubble was no different from
that of previous bubbles. Just as in Florida in 1925 or Los Angeles in
1887, participants genuinely believed that prices would continue to rise.
This belief seems to have been just as prevalent among the sophisticated
suppliers of finance as among the unsophisticated borrowers. While
inexperienced “flippers” were increasing their exposure to a real estate
market that was nearing unsustainable heights, so were many of the
managers of the Wall Street banks that were securitizing the flippers’

A. McCoy, Andrey D. Pavlov, and Susan M. Wachter, “Systemic Risk Through


Securitization: The Result of Deregulation and Regulatory Failure,” Connecticut
Law Review 41 (2009): 1327–​75; 94 Stat. 161 (1980); 96 Stat. 1545–​46 (1982).
30 Adam J. Levitin and Susan M. Wachter, “Explaining the Housing Bubble,”
Georgetown Law Journal 100 (2012): 1177–​258.
  239

Land and Onions 239

mortgages.31 As in the 1920s, the insiders were just as optimistic as the


outsiders.
Previous bursts of real estate speculation had not given rise to any
substantial proposals for regulating the market, but this episode was
simultaneously a real estate bubble and a much more dangerous finan-
cial bubble in securities backed by real estate. Congress was busy in the
aftermath, and its response to the twin bubbles followed the pattern of
previous crises. The flurry of ensuing legislation did little or nothing to
limit speculation in land, but it placed new limits on speculation in the
securities whose value derived from land.
The Housing and Economic Recovery Act of 2008 was the primary
legislation for the housing market. It was intended to ease the foreclo-
sure crisis rather than to restrain speculation in real estate, and to that
end it provided government insurance for new mortgages to replace
loans to homeowners at risk of foreclosure. Indeed, to the extent the
new law had any effect on speculation, it was to make purchasing real
estate even easier than it had been before, by raising the limits on the size
of loans that could be insured by the Federal Housing Administration,
and by providing a tax credit for first-​time homebuyers.32
By contrast, the Dodd-​Frank Act of 2010, the primary legislation
addressing securitization, was more restrictive. The act required secu-
ritizers to retain at least 5 percent of the risk of the mortgages being
securitized, in the hope that being forced to hold this risk would make
securitizers more careful about the soundness of the mortgages. The act
also required issuers of mortgage-​backed securities to disclose informa-
tion about the securities. The precise nature of that information was
left to the Securities and Exchange Commission, but the point was to
help investors assess the risk that the underlying mortgages would go

31 Robert J. Shiller, The Subprime Solution: How Today’s Global Financial Crisis
Happened, and What to Do about It (Princeton, N.J.: Princeton University
Press, 2008), 39–​68; Yuliya Demyanyk and Otto Van Hemert, “Understanding
the Subprime Mortgage Crisis,” Review of Financial Studies 24 (2011): 1848–​
80; William M. Goetzmann et al., “The Subprime Crisis and House Price
Appreciation,” Journal of Real Estate Finance and Economics 44 (2012): 36–​
66; Ing-​Haw Cheng et al., “Wall Street and the Housing Bubble,” NBER
Working Paper 18904 (2013).
32 122 Stat. 2264 (2008).
240

240 Speculation

into default. The Dodd-​Frank Act also included provisions intended to


abolish some of the darker arts of lending that took place during the
bubble. Mortgage originators were barred from receiving compensation
that varied based on the terms of the loan other than the amount, to
remove the incentive to steer borrowers into loans with higher inter-
est rates. Lenders were required to make a good-​faith effort to verify
the borrower’s ability to pay. Exceptionally large balloon payments and
fees were banned. These measures were intended in part to protect less
sophisticated borrowers against what had come to be called “predatory”
lending, but also in part to protect investors in mortgage-​backed securi-
ties, by weeding out the lowest-​quality loans.33
Had there been more interest in limiting speculation in real estate
itself, in addition to limiting speculation in securities backed by real
estate, a few measures would have been available for consideration. The
phenomenon of “flipping” might have been deterred by imposing a
special tax on gains from short-​term holdings of real estate. Congress
might have reduced the availability of subprime mortgages by real-
lowing states to set maximum interest rates for home loans, a power
Congress had taken away in 1980. The terms of mortgage loans that
seemed to cause the most trouble, like adjustable rates and balloon pay-
ments, might have been further regulated by revisiting the 1982 statute
that authorized such terms. None of this was done, because there was
no public outcry for restricting land speculation.
The experience of 2007–​10 thus replicated the experience of 1926–​33.
The real estate market and the securities market rose together and then
fell together. It was widely believed in retrospect that there had been
too much speculation in both. The law quickly responded with new
rules governing speculation in securities. But there were no new rules
governing speculation in land, which has always seemed more natural
and less dangerous than speculation in intangible assets.

33 124 Stat. 1376 (2010).


7
Q
Inside Information

Rodolphe Agassiz was the consummate insider. He was an elite


Bostonian, grandson of the distinguished Harvard naturalist Louis
Agassiz and son of Alexander Agassiz, the founder of the Calumet and
Hecla Mining Company, one of the leading mining firms of the early
twentieth century. As a young man, Rodolphe Agassiz was one of the
best polo players in the world. He became chairman of the board of the
family mining company after his father’s death. Meanwhile, he held
prominent positions in several Boston institutions: he was a director of
the State Street Trust Company, the Old Colony Trust Company, the
First National Bank, and the Edison Electric Illuminating Company.
Agassiz was also an active member of the Republican Party until the
mid-​1920s, when he broke with the Republicans because of their sup-
port for Prohibition. Few Americans were better placed to profit from
inside information than Rodolphe Agassiz.1
The Calumet and Hecla Mining Company owned a bit less than half
of the shares of the Cliff Mining Company, a firm with copper mines
in northern Michigan. The rest of the shares were dispersed among a
large number of stockholders, so Cliff Mining was in effect a subsidiary
of Calumet and Hecla. Agassiz was the president of Cliff Mining. In
1924, he hired the Harvard geologist Louis Graton to study the prop-
erty in Michigan and to report on the likelihood of finding copper ore.2

1 New York Times, 1 Aug. 1933, 17.


2 Wall St. Journal, 22 Sept. 1928, 7; 18 Oct. 1929, 13; C. S. Hurlbut Jr., “Memorial
of Louis Caryl Graton,” American Mineralogist 57 (1972): 638–​43.

241
242

242 Speculation

Graton came back with a highly positive confidential report in March


1926. The Michigan mines were far more valuable than anyone had
realized.
Agassiz put this information to immediate use. He and James
MacNaughton, the local manager of Cliff Mining, promptly purchased
on their own behalf nine thousand shares of the company, which was
more than a quarter of the outstanding shares that were not owned
by Calumet and Hecla. They were able to make these purchases at
approximately $2.50 per share. While Agassiz and MacNaughton were
buying up the stock of Cliff Mining, they shut down Cliff Mining’s
exploratory operations in Michigan and announced that the mine
was being abandoned as unprofitable, an announcement that pre-
vented Cliff Mining’s stock price from rising. All the while, Calumet
and Hecla was quietly obtaining options to buy the land adjacent to
the Cliff Mining properties. In October 1926, once they had finished
their stock purchases, Agassiz and MacNaughton reopened the mine
and made the geologist’s report public. Shares in Cliff Mining rose to
$12.50. Agassiz and MacNaughton sold their shares and made roughly
$90,000, or well over $1 million in today’s money. Agassiz most likely
did not think he had done anything wrong. He had not tried to con-
ceal his identity as the purchaser or seller of the stock in his company.
It was simply a matter of common sense to speculate where one had
the best information.
Homer Goodwin was furious. Goodwin was a local stockbroker, a
member of the Boston Stock Exchange. He had owned seven hundred
shares of the Cliff Mining Company, but he sold them in May, while
Agassiz and MacNaughton were snapping up the firm’s shares, and
while the share price was still $2.50. He sued Agassiz to get the shares
back. (Goodwin initially sued MacNaughton too, but he never followed
through, most likely because MacNaughton lived in Michigan while
Agassiz was close at hand in Boston.) Had he known what Agassiz had
known, Goodwin argued, he never would have sold his stock at $2.50
per share. He would have waited for the inevitable rise in price once the
mine was reopened and the geologist’s report was disclosed. Goodwin
contended that Agassiz had an obligation to the shareholders of the
Cliff Mining Company to refrain from using his superior information
  243

Inside Information 243

to take advantage of them by purchasing their stock for less than its
true value.
The Massachusetts Supreme Court held that Agassiz had not broken
the law. Corporate directors and officers owed a duty to the corpora-
tion, explained Chief Justice Arthur Prentice Rugg in his opinion for
the court, but not to each of the corporation’s shareholders as indi-
viduals. Besides, directors bought and sold their own company’s stock
all the time. “An honest director would be in a difficult situation,”
Rugg worried, “if he could neither buy nor sell on the stock exchange
shares of stock in his own corporation without first seeking out the
other actual ultimate party to the transaction and disclosing to him
everything which a court or jury might later find that he then knew.”
Disclosure might be morally right, but it was not a legal requirement.
“Business of that nature is a matter to be governed by practical rules,”
Rugg observed. “Law in its sanctions is not coextensive with morality.”
And who would serve as the director of a corporation if he had to pay
the heavy price of not dealing in the corporation’s stock? The court
concluded: “Fiduciary obligations of directors ought not to be made
so onerous that men of experience and ability will be deterred from
accepting such office.”3
Rugg pointed out that most transactions in the economy were just like
the one between Goodwin and Agassiz, in that they involved one party
who possessed more information than the other. Inequality was inherent in
contracting. The law “cannot undertake to put all parties to every contract
on an equality as to knowledge, experience, skill and shrewdness,” Rugg
insisted. Sometimes the party with superior information got the better of
the deal, but that was life. A court “cannot undertake to relieve against
hard bargains made between competent parties without fraud.” Agassiz
knew more about the financial condition of the Cliff Mining Company
than Goodwin did, but Goodwin’s remedy was not to force Agassiz to
share what he knew. Rather, it was to find out the information for him-
self. Goodwin “was no novice,” Rugg lectured. “He was a member of the
Boston stock exchange.” Before selling his stock, “he made no inquiries
of” Agassiz “or of other officers of the company.” If he had asked about the

3 Goodwin v. Agassiz, 186 N.E. 659, 661 (Mass. 1933).


244

244 Speculation

mines in Michigan, and if Agassiz had lied to him, that would have been
fraud, and Goodwin would have been entitled to rescind the transaction.
As Rugg put it, “directors cannot rightly be allowed to indulge with impu-
nity in practices which do violence to prevailing standards of upright busi-
ness men.” But Agassiz had not lied to anyone, much less to Goodwin. He
had merely stayed quiet about what he knew, and that did not disable him
from dealing in the stock of his own company.4
Seventy years later, Samuel Waksal found himself in the same posi-
tion as Rodolphe Agassiz. Waksal, a former immunologist, had started
a company called ImClone, whose primary product was a promising
cancer treatment called Erbitux. But Waksal learned, to his disap-
pointment, that the Food and Drug Administration would not permit
Erbitux to be sold. When the rest of the world found out, ImClone’s
share price would plummet. Just as Rodolphe Agassiz had taken ad-
vantage of his access to good news, Waksal sought to take advantage
of his own access to bad news. He immediately tried to sell $5 million
of ImClone stock before the FDA announced its decision. But times
had changed. Waksal’s brokers refused to execute his sell order, because
they knew he was not allowed to sell his ImClone stock until the news
reached the public. Waksal was arrested a few months later. He pled
guilty to securities fraud and several other crimes, and he was sentenced
to seven years in prison. While he was trying to sell his own shares,
Waksal had urged friends and family members to sell their shares too.
One of those friends was the celebrity homemaker Martha Stewart,
who ended up serving a five-​month prison sentence for making false
statements during the investigation of the incident.5
Over the course of the twentieth century, the law’s treatment of inside
information had changed dramatically. At the beginning of the century,
speculators’ ability to exploit their superior access to knowledge was
considered at worst inevitable, and at best a positive good. By the end of
the century, at least in certain circumstances, exploiting one’s knowledge
in this way was a serious crime. Rodolphe Agassiz had been esteemed
a shrewd businessman. Although he died a mere two months after the
Massachusetts Supreme Court decided his case, his lengthy obituary in

4 Id. at 661–​62.
5 Daniel Kadlec, “Sam’s Club,” Time, 17 June 2002.
╇ 245

Inside Information 245

the New York Times discussed his polo playing and his service on corpo-
rate boards but did not even mention Homer Goodwin’s lawsuit against
him. Samuel Waksal did the same thing Agassiz had done, but when his
obituary is written he will likely be remembered as a criminal. Between
Agassiz’s era and Waksal’s, speculating on inside information had been
transformed from a normal, sound business practice into a vice.

One-╉Sided Knowledge
Insider trading in the shares of a corporation is just one instance of a
much broader and much older question of commercial morality. Where
a buyer or a seller knows more than his counterpart, when should he
be able to use that knowledge to his advantage, and when should he
be required to disclose the information before entering into a transac-
tion? This was an important and unsettled question in the early United
States.
The issue was sharply presented to the US Supreme Court in 1817 in
Laidlaw v. Organ, which became one of the court’s most well-╉known com-
mercial cases in its early years. During the War of 1812, tobacco prices were
much lower than normal due to the British blockade of southern ports,
which deprived tobacco growers of foreign markets. The war formally
ended in December 1814 with the signing of the Treaty of Ghent, but it
took some time for news of the treaty to reach the United States. In the eve-
ning of February 18, 1815, Hector Organ, a New Orleans tobacco merchant,
learned about the treaty. He realized that the news would spread quickly,
and that tobacco prices would rise with the lifting of the blockade. Early
the next morning, just after sunrise, he called on another tobacco mer-
chant in the city, Peter Laidlaw & Company, and offered to buy approxi-
mately 120,000 pounds of tobacco. Francis Girault, a member of Laidlaw
& Company, asked Organ if there was any news that might influence the
price of tobacco. Organ did not respond. Girault nevertheless sold Organ
the tobacco. Within a day, as news of the treaty filtered through New
Orleans, the price of tobacco rose 30 to 50 percent. Laidlaw & Company
seized the tobacco it had just delivered to Organ. Organ sued.6

6 Laidlaw v.  Organ, 15 U.S. 178 (1817); M.  H. Hoeflich, “Laidlaw v.  Organ,
Gulian C. Verplanck, and the Shaping of Early Nineteenth Century Contract
246

246 Speculation

Was Laidlaw & Company bound by the contract? The company’s


lawyer insisted it was not. “The parties [were] treated on an unequal
footing,” he argued, “as the one party had received intelligence of the
peace of Ghent, at the time of the contract, and the other had not.”
Organ’s lawyer was Francis Scott Key, who had written the lyrics to
The Star-​Spangled Banner while watching the British bomb Baltimore
only a few months before his client had bought the tobacco. Organ’s
failure to disclose that the war was over might have been immoral,
Key acknowledged, but “human laws are imperfect in this respect,
and the sphere of morality is more extensive than the limits of civil ju-
risdiction.” Indeed, he continued, Organ deserved praise rather than
censure. He had done nothing wrong, “unless rising earlier in the
morning, and obtaining by superior diligence and alertness that intel-
ligence by which the price of commodities was regulated, be such.”
In any event, Key contended, it would be futile to require disclosure,
because virtually every contract involved one party who knew things
the other did not. “It is a romantic equality that is contended for on
the other side,” Key declared. “Parties never can be precisely equal in
their knowledge.”7
The court agreed with Key. Chief Justice John Marshall wrote a very
short opinion, only a few sentences long. Organ “was not bound to
communicate” his knowledge of the treaty, the court concluded. “It
would be difficult to circumscribe the contrary doctrine within proper
limits, where the means of intelligence are equally accessible to both
parties. But at the same time, each party must take care not to say or do
any thing tending to impose upon the other.”8 That is, a rule requiring
disclosure in this particular case would lead to chaos. It would open
up all contracts to challenge, because there would be no way to draw a
principled line distinguishing the situations in which information had
to be disclosed from those in which information could be kept secret.
Contracting parties could not lie to each other, but they were under no

Law: A Tale of a Case and a Commentary,” University of Illinois Law Review


1991 (1991): 55–​66.
7 Laidlaw, 15 U.S. at 185–​86, 193.
8 Id. at 194.
  247

Inside Information 247

obligation to reveal what they knew. As the old Latin expression put it,
caveat emptor—​let the buyer beware.
But this view was controversial. In a similar case a few years ear-
lier, for example, a Kentucky court had reached the opposite result.
A Virginian named Bowman owned a tract of land in Illinois that he
had never seen. (This was not unusual at the time. Bowman was appar-
ently a war veteran who had received the land as part of the compensa-
tion for his military service.) John Bates lived near the tract. He knew
that it included salt water, which made it more valuable, and he knew
that Bowman was unaware of the salt water. Bates communicated this
information to his brother James, who lived near Bowman. James Bates
purchased the land from Bowman without telling him about the salt
water. Bowman sued, and the court set aside the sale. To allow one
person to take advantage of another’s ignorance “might become perni-
cious in practice,” the court worried. It would allow “the dishonesty
and cunning of some to operate upon the honest credulity of others.”9
Indeed, a year after Laidlaw v.  Organ, the Mississippi Supreme
Court disagreed with the US Supreme Court even more directly. Frazer
v. Gervais was nearly a carbon copy of Laidlaw: a merchant who knew of
the Treaty of Ghent bought cotton from a merchant who did not, at a
price that seemed far too low once the news of the treaty had spread. The
Mississippi court refused to enforce the transaction. Once the purchaser
heard information that would affect the price of cotton, the court con-
cluded, he could not buy cotton without first divulging the information
to the seller.10
The lawyer and future congressman Gulian Verplanck was so out-
raged by Laidlaw v. Organ that he wrote an entire book criticizing it.
Verplanck’s Essay on the Doctrine of Contracts was devoted to answering
the question posed by Laidlaw and similar cases. “We shall find num-
berless cases of the most opposite moral character, from the arts of de-
liberate knavery to the fairest gains of industry and enterprise, all agree-
ing in this one prominent feature: that the profits of the transaction
arose from the superiority of one party to the other in some material

9 Bowman v. Bates, 5 Ky. 47, 53 (1810).


10 Frazer v. Gervais, 1 Miss. 72, 73 (1818).
248

248 Speculation

knowledge respecting the subject of the contract,” Verplanck noted.


“Where then shall we draw the line of fair and unfair, of equal and un-
equal contracts?” Verplanck thought the line should be drawn between
information about a person’s own desires and capacities, which would
not have to be disclosed, and information about the external world,
which would. “My knowledge of my own interests, and my personal
necessities, my sagacity, natural or acquired, in forming judgments of
the state of the market,” he reasoned, “can never be expected by the
other party to be communicated; and in most instances it would be
impracticable or silly to do so.” People differed widely in their moti-
vations and their skills, but it could not be unlawful to exploit such
advantages, or else commerce would cease entirely. “Not so with regard
to the common facts, which immediately and materially affect price,”
Verplanck continued. “It is true, that strict equality of knowledge, as to
these points, is just as difficult as with regard to other matters. But the
contract is entered into on the supposition, that whatever superiority
of knowledge one may have over the other, no advantage will be taken of
it.” Verplanck provided some examples. When he purchased a hat from
his hatter, the hatter knew much more about the value of hats than he
did. But if the hatter tried to exploit this informational advantage by
charging him $10, when better-​informed customers paid only $5, the
hat-​buying contract would be unenforceable, because hat purchasers
were entitled to presume that the hatter would treat them equally. And
of course, if a purchaser of tobacco or cotton knew that the War of
1812 had ended, he could not gain an edge on a seller who had not yet
heard the news. On the other hand, if a merchant discovered a new
use for a product, or if a scientist developed a new invention, there was
nothing wrong with exploiting these informational advantages, because
customers would “take it for granted, that such advantage, if possessed,
would be used.”11
Verplanck was hardly alone in believing that knowledge should not
be kept secret to the commercial detriment of others. Caveat emptor
was “a most demoralizing principle,” complained the lawyer-​scientist

11 Gulian C. Verplanck, An Essay on the Doctrine of Contracts (New York: G. &


C. Carvill, 1825), 10, 119–​21.
  249

Inside Information 249

Thomas Cooper. “It is a disgrace to the law that such a maxim should
be adopted.” Cooper thought the court opinions applying the doc-
trine “ought to be classed as cases in support of falsehood and fraud.”
An admiring reviewer of Verplanck’s book blamed the doctrine on
the American inheritance of English common law, which had been
formed long before under primitive circumstances. “A savage race, like
the Saxons, and a nation of military barbarians, like the Normans,” he
scoffed, “could hardly be expected to derive their law of contracts from
a very pure and enlightened source, or to found it upon the enlarged
principles of civilized morality.” A Kentucky court refused to enforce
a contract for the sale of a horse when the seller failed to tell the buyer
that the horse was blind. “It will better comport with morality and
sound policy,” the court insisted, “to subject the seller to an action at
law, and thereby impose upon him the legal, as well as moral duty of
telling the whole truth.” James Kent, whose four-​volume treatise on
American law was a standard reference work for decades after its publi-
cation in the 1820s, declared that “as a general rule, each party is bound
in every case to communicate to the other his knowledge of material
facts, provided he knows the other to be ignorant of them, and they be
not open and naked, or equally within the reach of his observation.”12
On this view, information was not a weapon to be wielded in the mar-
ketplace. The law, like common morality, should forbid the informed
from taking advantage of the ignorant.
But if there were many voices speaking against the caveat emptor
doctrine in the early nineteenth century, there were also many, prob-
ably more, speaking in its favor. “It would certainly be well, for the
best interests of society, if the pure principles of morality … could be
established as the law of the land, and practically enforced in the or-
dinary transactions of life,” one Mississippi lawyer acknowledged. But
requiring that contracting parties tell each other all they knew would be
“replete with difficulties.” Was a buyer bound to tell a seller “every fact
or rumor calculated to enhance the price of the article—​or if not every

12 Thomas Cooper, The Institutes of Justinian (Philadelphia:  P.  Byrne, 1812),


610–​11; Book Review, North American Review 22 (1826):  262; Hughes
v. Robertson, 17 Ky. 215, 217 (1824); James Kent, Commentaries on American
Law (New York: O. Halsted, 1826–​30), 2:377.
250

250 Speculation

fact, what facts, and who is to be the judge?” The law had to facilitate
commerce, not just promote morality, and the most moral rule was
not always the most practical. “In some special cases the law of the land
and morality are the same,” reasoned the Boston lawyer Nathan Dane,
but the two differed when “policy” was taken into account. Thus, every
“undue advantage in a bargain, to the hurt of another party, practised
by one, is an act of injustice in the eyes of morality; yet it is not the
means of restitution in the eyes of the law.” As another commentator
put it, caveat emptor “furnishes a plain, simple and practical rule for
the decision of all controversies of this nature,” even if “it is said by
some to be a hard, inequitable doctrine, repugnant to sound morality,
and productive of fraud, injustice and oppression.”13 Buyers or sellers
could not lie, but they had no duty to volunteer the truth.
Between the 1830s and 1850s, several courts accordingly declared that
American law embodied the principle of caveat emptor. A contract for
the sale of coffee beans was enforceable, held the US Supreme Court
Justice Henry Baldwin, even if one party failed to disclose his superior
knowledge of the state of the coffee market. “A purchaser may avail
himself of information which affects the price of the article, though it is
not known to the seller,” Baldwin explained. “Though the latter inquires
if there is any news which affects the price, the purchaser is not bound
to answer, and the contract is binding, though there was news then
in the place which raised the price.” As Baldwin saw it, acquiring and
exploiting information was a normal part of conducting business. “The
buying and selling [of ] merchandize being for mutual profit,” he rea-
soned, “one is not bound to impart to another his views of speculation,
his opinion of the effect of news or events, the bearing of the rise of one
article on another, or the results of his mercantile skill and knowledge
of the markets.” A few years later, the Indiana Supreme Court agreed,
in enforcing a contract in which a professional pork dealer bought ap-
proximately a thousand hogs, at a price well below market value, from
an elderly man unfamiliar with the value of pork. “However justly the

13 Note appended to Frazer v. Gervais, 1 Miss. 72, 73 (1818); Nathan Dane, A


General Abridgment and Digest of American Law (Boston: Cummings, Hilliard
& Co., 1823–​29), 1:100; “Caveat Emptor—​The Rule of the Common Law—​
Not of the Civil Law,” American Jurist 12 (1834): 95.
  251

Inside Information 251

moralist may censure the address sometimes resorted to by men of


keen business habits to effect advantageous contracts,” the court ob-
served, “misrepresentations as to the value or quantity of a commodity
in market, when correct information on those subjects is equally within
the power of both contracting parties with equal diligence, do not in
contemplation of law constitute fraud.” A contrary rule, the court wor-
ried, would lead to results much worse than the hard bargain it was
enforcing. “To deprive the better informed, more enterprising, and
more cautious party of the benefit of his contract, on account of repre-
sentations, of the correctness of which the other party ought to judge
for himself, would tend more to encourage ignorance, sloth, and reck-
lessness, than to repress dishonesty.” The Pennsylvania Supreme Court
took the same view while enforcing a land sale in which an experienced
miner, who knew that the land contained valuable chrome deposits,
purchased it from a family of poor and sickly farmers who were un-
aware of the chrome, at a price suitable for barren farmland rather than
for a valuable mining property. “A person who knows that there is a
mine on the land of another may nevertheless buy it,” the court held.
“The ignorance of the vendor is not of itself fraud on the part of the
purchaser. A purchaser is not bound by our laws to make the man he
buys from as wise as himself.”14
As caveat emptor became the general rule, however, many of the
judges and commentators who espoused it remained uncomfortable
with some of the more unsavory transactions it seemed to permit.
They increasingly began to suggest that the doctrine should not apply
in circumstances where the less knowledgeable person had a particu-
lar reason to place trust in the more knowledgeable. The US Supreme
Court Justice Joseph Story was one of the first to spell out these cir-
cumstances at length, in his widely read 1836 treatise on equity jurispru-
dence. “If a vendor should sell an estate, knowing that he had no title to
it,” Story argued, the purchaser should be able to void the transaction,
because “the very purchase implies a trust and confidence” on the part
of the purchaser that the seller actually owns what he purports to sell.

14 Blydenburg v. Welsh, 3 F. Cas. 771, 773 (C.C.D. Pa. 1831); Foley v. Cowgill, 5
Blackf. 18, 20 (Ind. 1838); Harris v. Tyson, 24 Pa. 347, 359–​60 (1855).
252

252 Speculation

“The like reason would apply,” Story added, “where the vendor should
sell a house, situate[d]‌in a distant town, which he knew at the time to
be burnt down.” Story maintained that cases like these were exceptions
to the general rule of caveat emptor, because “there are circumstances of
peculiar trust and confidence, or relation between the parties.”15
Story’s view soon became orthodoxy. It was a middle ground be-
tween a strict rule of caveat emptor and a rule requiring disclosure.
In the 1820s, James Kent had argued that disclosure should always be
required, but in later editions of his treatise, beginning in the 1840s, he
conceded that such a requirement, “though one undoubtedly of moral
obligation, is perhaps too broadly stated, to be sustained by the practi-
cal doctrine of the courts.” He agreed with Story that disclosure was
required only where the person knowing the facts was “under some spe-
cial obligation, by confidence reposed, or otherwise, to communicate
them truly and fairly.” The courts agreed as well. In Maine, for example,
the state supreme court considered the case of a man who had paid for
a carriage with a note (that is, a promise to pay) from a third party. He
failed to tell the carriage seller that it would be impossible to collect on
the note. The court required him to give the carriage back to the seller.
A person had to disclose such information “if the means of information
are not equally accessible to both” the buyer and the seller, the court
held, “and especially when one of the parties relies upon the other to
communicate to him the true state of facts to enable him to judge of the
expediency of the bargain.”16 By the middle of the nineteenth century,
it was fairly well settled that one party to a contract had no obligation
to disclose information to the other, even information that might have
a big effect on the price of whatever was being sold, unless there was
some special relationship between the parties that imposed such a duty.

15 Kim Lane Scheppele, Legal Secrets: Equality and Efficiency in the Common Law
(Chicago: University of Chicago Press, 1988), 269–​98; Paula J. Dalley, “The
Law of Deceit, 1790–​1860:  Continuity Amidst Change,” American Journal
of Legal History 39 (1995):  405–​42; Joseph Story, Commentaries on Equity
Jurisprudence (Boston: Hilliard, Gray & Co., 1836), 1:218–​19, 221.
16 James Kent, Commentaries on American Law, 5th ed. (New York: Printed for
the author, 1844), 2:482 note a; Prentiss v. Russ, 16 Me. 30, 32–​33 (1839); William
W. Story, A Treatise on the Law of Contracts Not Under Seal (Boston: Charles
C. Little and James Brown, 1844), 109.
  253

Inside Information 253

As shares in business corporations became more frequently bought


and sold, they were just one more asset to which general commercial
principles would apply. The question of how to deal with unequal in-
formation often arose in stock transactions. Everyone recognized the
importance of timely information to stock prices. A successful investor
“must watch the telegraph instruments which bring news from vari-
ous parts of the world, and be prepared to act promptly,” advised the
president of the New York Stock Exchange. “Opinions formed in the
morning must be changed at noon, and the evening’s news furnishes no
safe basis for the next day’s operations.”17 But some shareholders could
acquire information before others, because some were actively involved
in directing the corporation’s activities, while others were mere inves-
tors who knew scarcely anything about the day-​to-​day operations of the
company. When insiders bought stock from or sold stock to outsiders,
did they have an obligation to reveal their superior knowledge of the
company? If the general rule of caveat emptor applied to these transac-
tions, corporate directors could buy and sell without disclosing what
they knew. But if the directors of a corporation owed a duty to the
corporation’s shareholders not to take actions that would harm them,
the general rule of caveat emptor would not apply, and the directors
would have to disclose before transacting. The disclosure question thus
hinged on a question of corporate law. Did directors owe any duties to
shareholders?
Corporate law was itself in the process of forming, so this too was an
unsettled question. In 1836, for example, the Alabama Supreme Court
considered a lawsuit filed by the estate of Joseph Spence, who had
owned six shares of the Courtland Land Company, a corporation that
owned land it planned to subdivide into lots and sell. The company was
managed by William Whitaker, who although nominally the treasurer,
failed to keep any accounts. Shortly before Spence’s death, Whitaker
purchased Spence’s shares for $93.50 each, an amount far lower than
their actual value. The Alabama Supreme Court held that the transac-
tion was void because of Whitaker’s failure to disclose the true financial

17 Edwin T. Freedley, A Practical Treatise on Business (Philadelphia: Lippincott,


Grambo & Co., 1853), 168–​69; Brayton Ives, “Wall Street as an Economic
Factor,” North American Review 147 (1888): 569.
254

254 Speculation

condition of the company. The court determined that Whitaker was “a


trustee, in whom faith and confidence have been reposed, and whose
agency imparted to him a more intimate knowledge of the condition
and value” of the shares than Spence, an ordinary shareholder, could
possess. Whitaker had “taken advantage of his superior knowledge, to
force an unequal bargain.” The court accordingly rescinded the sale of
Spence’s shares.18
But the large majority of cases in the second half of the century held
precisely the opposite—​that a corporate director had no obligation to
disclose anything when transacting with shareholders, because directors
owed a duty to shareholders with respect only to the management of
the corporation, not with respect to property (including shares in the
corporation) owned by the shareholder or the director as individuals. In
1862, for instance, the administrator of Lloyd Glover’s estate was selling
off Glover’s assets, after Glover accidentally killed himself while hunting
when he tried to board a boat with a loaded gun. Glover had been an
employee of the National Bank Note Company, and among his assets
were 136 shares of the company’s stock. Glover’s administrator sold the
shares to George Danforth, one of the company’s directors. Not long
after, the company declared an extraordinary dividend, most likely be-
cause it had the lucrative contract to print the federal government’s new
paper currency, and the administrator realized he had sold the stock
for far less than its actual value. His suit against Danforth raised the
same question the Alabama Supreme Court had considered: was caveat
emptor the governing rule, or did Danforth, as a director, owe a duty
to shareholders not to exploit his superior information?19
A New York court decided that there was nothing unlawful about
the transaction. There is “a certain trust relation between the sharehold-
ers and the directors of a corporation,” the court reasoned, “but the
trust put in the directors usually extends … only to the management
of the general affairs of the corporation.” Danforth owed shareholders
a duty to manage the corporation properly, but not a duty to help the

18 Spence v. Whitaker, 3 Port. 297, 325 (Ala. 1836).


19 New  York Times, 7 Aug. 1862, 3; Carpenter v.  Danforth, 52 Barb. 581 (N.Y.
Sup. 1868).
  255

Inside Information 255

shareholders sell their shares at an appropriate price. That was their


business. Danforth knew much more about the company’s financial
condition than Glover’s administrator, but it was the administrator’s
“right to judge for himself as to the value of the stock, and to enable
him to do so, it was his right, and perhaps his duty, to inform himself
as to facts material on the question of value.” If a shareholder desired
information about the state of a corporation, he had to find it out for
himself before buying or selling. Transactions with directors were no
different from transactions with anyone else.20
This view soon became standard. Directors were trustees for share-
holders “in respect to the property of the corporation,” not the property
of the shareholders, declared the Indiana Supreme Court, in the course
of upholding a transaction in which the president of a railroad com-
pany bought up company stock while, unbeknownst to the sharehold-
ers, he was negotiating a sale of the corporation to another railroad at
a much higher price per share. The Tennessee Supreme Court agreed
in another case in which several directors of a railroad purchased stock
from shareholders because, due to their positions, they knew the stock
was worth more than the sale price. “A director or the treasurer of a
corporation is not, because of his office, in duty bound to disclose to an
individual stockholder, before purchasing his stock, that which he may
know as to the real condition of the value of that stock,” New Jersey’s
highest court concluded. It was thus not unlawful for a director of a
printing company to buy shares when he knew, but the shareholders
did not, that the company had just “made a favorable sale of property,
which enhanced the value of its stock.”21 When the New York lawyer
William Cook summed up the cases in 1887, in his Treatise on the Law
of Stock and Stockholders, it took him only a few sentences. “A director
of the corporation itself may buy and sell its stock like any other indi-
vidual,” Cook explained. “The information which he has of the affairs
of the corporation, whereby he is enabled to buy or sell at an advantage

2 0 Id. at 584, 589.
21 Board of Commissioners of Tippecanoe County v.  Reynolds, 44 Ind. 509, 513
(1873); Deaderick v. Wilson, 67 Tenn. 108 (1874); Crowell v. Jackson, 23 A. 426,
427 (N.J. 1891). See also Grant v. Attrill, 11 F. 469, 470 (C.C.S.D.N.Y. 1882);
Gillett v. Bowen, 23 F. 625, 626 (C.C.D. Colo. 1885).
256

256 Speculation

over the person with whom he deals, does not affect the validity of the
transaction.”22
While the judges tended to emphasize the doctrinal justification for
allowing corporate directors to trade on inside information—​the lack
of any duty they owed to shareholders with respect to their ownership
of shares—​the judges also sometimes provided practical reasons for
permitting insider trading. Stocks were constantly fluctuating in value,
the Utah Supreme Court reasoned, and one “who assumes to know the
most about them is frequently the first one to be deceived.” Information
was money in this uncertain world, so a director, like anyone, “is en-
titled to the benefit of his facilities for information.” The solution to the
problem of unequal information was not compelled disclosure, argued
the Michigan Supreme Court. It was for the less-​informed party to find
out what the insiders already knew. “The books of the corporation are
open to all stockholders alike,” the court noted, “and each may inform
himself of the condition of the company.” If a shareholder did not avail
himself of this opportunity, lectured Maryland’s highest court, he had
only “his own lack of business prudence” to blame.23
In reaching this conclusion, American courts were both ratifying a
pervasive practice and ensuring that it would continue. By all accounts,
insider trading was very common. Guides for investors routinely
warned that the market was an uneven playing field, where newcomers
would be up against better-​informed insiders. In eighteenth-​century
England, when government debt securities (then also called “stocks”)
were traded more widely than shares of businesses, Thomas Mortimer’s
oft-​reprinted Every Man His Own Broker cautioned that government of-
ficials “have the earliest intelligence of all events that can tend either to
raise or fall the Stocks,” and that an official, “if he should be a Jobber,”
might “settle a great account … to his advantage.” In the nineteenth
century, as corporate shares became an ever-​larger part of the market,
American authors repeated the same warning. “Sometimes the funds
of a Bank are employed in purchasing its own stock,” lamented the

22 William W. Cook, A Treatise on the Law of Stock and Stockholders (New York:


Baker, Voorhis & Co., 1887), 330.
23 Haarstick v. Fox, 33 P. 251, 153 (Utah 1893); Walsh v. Goulden, 90 N.W. 406, 410
(Mich. 1902); Boulden v. Stilwell, 60 A. 609, 612 (Md. 1905).
  257

Inside Information 257

economist and journalist William Gouge, “and then, if the price of-
fered be sufficiently high, those who have the management contrive
to sell their own shares.” The same was true of all corporations, which
were “liable to be abused,” cautioned the economist Daniel Raymond.
“This is mostly done by speculators and stock-​jobbers, getting control
and management of them, and then managing them with a view to
their own speculations,” a feat they could accomplish by keeping “the
stockholders and the public in the dark respecting the conditions of the
corporation, while they are themselves in the light.”24
The difficulty of matching wits with wily insiders was a staple of
the breathless exposés of Wall Street published at midcentury. “Certain
Stockholders are the losers,” one such publication reported, while “those
most active in getting up the concern are greatly improved in circum-
stances.” For this reason, “we advise those not acquainted with active
business, widows, orphans, &c. to leave all new undertakings to busi-
ness men.” Another cautioned that Wall Street was no place for “small
victims and outsiders.” “Don’t let speculators have anything to do with
the management,” insisted another. “Ask for the accounts; examine well
the details. … You have certainly a right to inquire about investing
your own money.”25 It was well understood, although often deplored,
that one of the customary perks of being the director of a corporation
was the opportunity to profit from trading in the corporation’s stock.
Two New  York stockbrokers published books in 1888 that were
part memoir and part investment guide, and both brokers discussed
the prevalence of insider trading. “Incomplete or insufficient informa-
tion is especially dangerous,” Henry Clews explained in an account of
his twenty-​eight years on the New York Stock Exchange. “One-​sided
knowledge is nowhere as deceiving as here.” Clews recalled examples

24 Thomas Mortimer, Every Man His Own Broker (London: S. Hooper, 1761), 44;
William M. Gouge, A Short History of Paper Money and Banking in the United
States (Philadelphia: T. W. Ustick, 1833), 76; Daniel Raymond, The Elements
of Constitutional Law and of Political Economy, 4th ed. (Baltimore: Cushing &
Brother, 1840), 276.
25 William Armstrong, Stocks and Stock-​ Jobbing in Wall-​ Street
(New York: New York Publishing Co., 1848), 28; George G. Foster, New York
in Slices (New  York:  W.  F. Burgess, 1849), 16; George Francis Train, Young
America in Wall-​Street (New York: Derby & Jackson, 1857), 212–​13.
258

258 Speculation

in which “a stock is bought up freely at New  York because London


is taking large amounts of it; a day or two later, the deliveries show
that large holders connected with the management are unloading on
the foreign market upon knowledge of facts damaging to the prospects
of the property.” J. F. Hume likewise revealed “how certain great rail-
road magnates, who are leading operators in Wall street, have amassed
such colossal fortunes.” It was by speculating in the stock of their own
companies. “Had they, like ordinary speculators, confined themselves
to other people’s goods,” Hume remarked, “it is questionable whether
they would have grown exceptionably rich.”26
In 1915, a magazine called The Annalist interviewed several directors
of large corporations to learn their opinions of insider trading. Nearly
all of them believed that there was nothing wrong with the practice.
They offered a variety of defenses. One bank director reasoned that it
was only fair. If a director “knows of a profitable contract about to be
closed,” he argued, “he has more right than any one else to benefit by a
rise in the stock on the announcement, because he has helped to get the
contract.” Another who was a director of many large corporations char-
acterized the profit from trading as a form of compensation to which
directors were due. “A Director gives his time and ability to the man-
agement of a corporation and gets $20 a week if weekly meetings are
held,” he explained. “That is surely not compensation enough. What
the Director can make in the market on the basis of what he learns as a
Director is part of the pay which he gets for his work.” Another director
argued that shareholders were not harmed when directors speculated in
their companies’ stock. A director should “buy stocks on the knowledge
which he has of good things to come,” he suggested. “The more stock
he holds the better qualified he is to serve on the board.” Finally, the
directors feared that a ban on buying and selling shares would make it
difficult for firms to find directors. “If it were held that a man’s holdings
of a stock should be frozen up the moment he becomes a Director,” one
worried, “I fancy very few could be induced to become Directors.” But

26 Henry Clews, Twenty-​Eight Years in Wall Street (New York: Irving Publishing


Co., 1888), 202–​3; J. F. Hume, The Art of Investing (New York: D. Appleton &
Co., 1888), 128–​29.
╇ 259

Inside Information 259

whether one welcomed or deplored insider trading, everyone recog-


nized it as, in the words of one director, something that “as a practical
matter nearly all Directors do.”27

A Contrariety of€Judicial Opinion


Insider trading was legal and commonplace, but it had its critics, who
thought it gave an unfair advantage to a privileged few. “Directors and
officers of our large corporations have peculiar means for obtaining
information with regard to the company with which they are con-
nected,” the Commercial & Financial Chronicle editorialized in 1872.
“This speculating upon information which all the stockholders are
equally entitled to is a very great evil.”28 On this view, insider trading
was common in the same way that rudeness or lying was common: it
was an ethically dubious but seemingly inescapable part of life. In a
legal environment in which caveat emptor was the governing rule for
most commercial transactions, and in which corporate directors owed
no duties to shareholders that might render caveat emptor inapplica-
ble, there was no ground for requiring directors to disclose information
to shareholders before they bought and sold the stock of their own
companies. A prudent shareholder had to assume the worst when deal-
ing with an insider. Yet this conclusion was intuitively unappealing to
many, including many judges.
The judges accordingly pressed back against the rule. In case after
case in the late nineteenth and early twentieth centuries, judges identi-
fied special circumstances that took the case outside the doctrine of
caveat emptor and imposed on insiders a duty of disclosure. For ex-
ample, when the president of the Commercial Insurance Company
told shareholders that he had found a purchaser for their stock, but the
ostensible purchaser turned out to be an agent for the president, who
was really trying to buy up the shares for less than their true worth, the
Rhode Island Supreme Court refused to allow the transactions to stand.
If the president had merely offered to buy the stock himself, the court

2 7 “Should Directors Speculate?,” The Annalist, 19 July 1915, 65.


28 “Railroad Officers as Stockholders and Speculators,” Commercial & Financial
Chronicle, 8 Feb. 1872, 145.
260

260 Speculation

reasoned, a shareholder “could have no right to complain. As a matter


of common sense, the fact that the president of the company, presumed
to be fully acquainted with its condition, offered a certain sum, would
put the seller on his guard, and he might in most cases reasonably con-
sider that he might get more.” But this case was different, in the court’s
view, because the president had concealed his identity as the purchaser.
He had claimed to be helping the shareholders, and by doing so he
had caused each shareholder “to repose a trust and confidence” in him.
That was enough, the court held, to impose on the president a duty to
act on the shareholders’ behalf. Caveat emptor was the ordinary rule,
“yet if there be any peculiar relation implying confidence or leading to
confidence, it would take the case out of the ordinary rule,” the court
concluded. “In the present case there are peculiar circumstances.”29
In another case, the directors of a mining company had intentionally
driven down the value of the stock by refusing to develop the mines,
so that they could buy up stock from shareholders at an artificially de-
pressed price. This would not have been illegal, explained the judge, a
young William Howard Taft, if there had been no relationship between
the buyers and sellers other than director and shareholder. A director
owed a duty to the corporation, not to the corporation’s shareholders.
But in this case the shareholders had pledged the stock to the directors
as collateral for a loan, so there was another relationship. The directors
owed a duty to the shareholders—​not as directors but as custodians of
the shares—​to refrain from taking actions that would reduce the value
of the stock.30
Judges were quick to find special circumstances that would prohibit
insider trading in a given case, without casting doubt on the general
principle permitting insiders to buy and sell without disclosing infor-
mation. In Kansas, the president of the Topeka Water Supply Company
bought stock from shareholders without telling them that he had
received an offer to buy the entire company at a much higher price
per share. But the president was not just a corporate officer, the state

29 Paula J.  Dalley, “From Horse Trading to Insider Trading:  The Historical
Antecedents of the Insider Trading Debate,” William & Mary Law Review 39
(1998): 1289–​1353; Fisher v. Budlong, 10 R.I. 525, 527–​29 (1873).
30 Ritchie v. McMullen, 79 F. 522, 532–​34 (6th Cir. 1897).
  261

Inside Information 261

supreme court determined. Once he had received the offer to buy the
company, he also had “a special agency for the sale,” under which “he
was specially authorized to find a purchaser,” and that special agency
imposed an obligation of candor toward the shareholders greater than
that of an ordinary insider. In Utah, when Alviras Snow, the treasurer
of a mining company, bought shares from C. D. Morrison for much
less than their true value, the state supreme court rescinded the transac-
tion. Snow was “a dealer in mines and mining stock” who had “almost
absolute control of the business affairs.” Morrison, by contrast, “was
a poor man, with a family, and had to keep constantly at work. He
resided at a wayside station on the railroad 500 miles from Salt Lake
City, where the books of the company were kept.” Morrison had never
even been to Salt Lake City. He knew little or nothing about the busi-
ness. He was entirely dependent on Snow for information. The stock
sale was “the culmination or final act of a scheme conceived by Snow,
weeks before, to defraud Morrison out of his stock.” These special facts
sufficed to render the normal caveat emptor rule inapplicable. “Under
these circumstances,” the court held, “Snow was both morally and le-
gally bound to refrain from doing anything … to mislead or deceive
Morrison.”31 Where an insider’s conduct seemed especially reprehen-
sible, or where a shareholder was particularly sympathetic, it did not
take much for courts to find special circumstances that allowed them to
disregard the usual rule.
The most famous of these “special facts” cases, Strong v. Repide, was
decided by the US Supreme Court in 1909. It came from the Philippines,
which had become an American colony shortly before. When the
United States assumed control of the Philippines, one of its first acts
was to purchase much of the colony’s land. One of the big landowners
was the Philippine Sugar Estates Development Company, which, after
protracted negotiations, reached an agreement in 1903 to sell a large
parcel to the government at a price favorable to the company, a deal
that would cause the company’s stock price to rise. Francisco Repide,
a director of the company and the owner of nearly three-​quarters of

31 Mulvane v. O’Brien, 49 P. 607, 613 (Kan. 1897); Morrison v. Snow, 72 P. 924,


927–​28 (Utah 1903).
262

262 Speculation

the company’s stock, conducted the negotiations. While doing so, he


purchased shares in the company from Eleanor and Richard Strong.
Repide bought the stock in a sneaky way, so that the Strongs would
not realize he was the purchaser and raise their asking price. He hired
a relative by marriage named Kauffman, who hired a broker, who told
the Strongs that the stock was for a member of Kauffman’s wife’s family.
The Supreme Court concluded that while, in the ordinary case, a di-
rector need not disclose inside information when buying shares, this
was not an ordinary case. “Concealing his identity when procuring the
purchase of the stock,” the court held, “was strong evidence of fraud.”
Repide may not have committed fraud in the strict sense—​he really
was a member of Kauffman’s wife’s family, so the broker’s statement was
technically true—​but his actions were devious, and that was enough.
A director normally owed no duty of disclosure to shareholders, “yet
there are cases where, by reason of special facts, such duty exists.” This
was one of them.32
At the turn of the century, the general rule was that insiders need not
disclose information when buying and selling stock, but that rule was
heavily qualified by judges’ willingness to find special circumstances
that required disclosure. Beginning in the first decade of the century,
courts in several states went one step further. They began to say that
corporate insiders did have a duty of disclosure, in all circumstances.
In the view of these courts, the directors of corporations were trustees
not just for the corporation, but for the shareholders as well. The courts
that took this step were in the South and the rural Midwest, far from
the nation’s financial centers.
The first court to prohibit insider trading was the Georgia Supreme
Court, which in 1903 considered a case involving the Gate City Oil
Company. The company’s president, William Oliver, had acquired op-
tions to purchase stock in the company from several shareholders for
$110 per share, while he was negotiating the sale of the company’s plant
for a sum that made the stock worth $185 per share. Oliver had not dis-
closed the impending transaction to the shareholders. Once he signed

32 Strong v. Repide, 213 U.S. 419, 431–​33 (1909).


  263

Inside Information 263

the contract to sell the plant, he exercised the options and immediately
resold the shares, yielding him an instant profit of nearly $50,000.
This scheme did not sit right with Joseph Lamar of the Georgia
Supreme Court. Lamar had just been appointed to the court. He was a
corporate lawyer from Augusta, who would stay on the court for only
two years before returning to his law practice. (In 1911, President Taft
would appoint him to the US Supreme Court, where he would serve
until his death five years later.) Lamar knew full well that under existing
law there was nothing illegal about Oliver’s transactions. But he did not
like it. “It is a matter of common knowledge,” Lamar noted, “that the
market value of shares rises and falls, not only because of an increase
or decrease in tangible property, but by reason of real or contemplated
action on the part of managing officers.” To allow an insider to trade on
his superior knowledge, “without making a full disclosure, and putting
the stockholder on an equality of knowledge as to those facts, would
offer a premium for faithless silence, and give a reward for the suppres-
sion of truth.” Lamar was dismayed that the law would allow a director
“to take advantage of his own wrong—​a thing abhorrent to a court.”
So he and his colleagues established a new rule. In Georgia, Lamar ex-
plained, the inside information obtained by a corporate director “is a
quasi asset of the company, and the shareholder is as much entitled to
the advantage of that sort of asset as to any other regularly entered on
the list of the company’s holdings.” Information about corporate deal-
ings belonged to the corporation, not to the director. “Where the direc-
tor obtains the information giving added value to the stock by virtue
of his official position,” Lamar declared, “he holds this information in
trust for the benefit of those who placed him where this knowledge was
obtained.” The director was thus “a quasi trustee as to the shareholder’s
interest in the shares.” He could not use inside information to take ad-
vantage of the shareholders.33
Kansas followed suit the next year. “The managing officers of a cor-
poration are not only trustees in relation to the corporate entity,” the
state’s supreme court held, “but they are also to some extent and in
many respects trustees of the corporate shareholders.” John Stewart,

33 Oliver v. Oliver, 45 S.E. 232, 233–​35 (Ga. 1903).


264

264 Speculation

the president of the Wellington National Bank, had purchased shares in


the bank for much less than their actual value from an eighty-​year-​old
stockholder who knew nothing about banking. The Kansas Supreme
Court rejected existing law, which, in the court’s view, left stockhold-
ers “the legitimate prey” of insiders. In a later case, the court required
insiders not merely to disclose information before they traded, but also
to explain it for the benefit of shareholders who would not otherwise
understand it.34
By the 1910s, insider trading was illegal in Nebraska and Iowa as
well. The law had once been clear, but now lawyers faced “a contrariety
of judicial opinion,” one law professor lamented. Some thought the
trend a bad one. “If a seller understands that the buyer is in a position
to know more about the subject of sale than he does,” insisted Clarence
Laylin of Ohio State University, “there is nothing morally or legally
wrong, according to present-​day standards, in laying upon him the
burden of making inquiry of the buyer.” Roberts Walker, a New York
corporate lawyer and a director in several large corporations, sniffed
that the newer view, “particularly at a distance from salt water, seems
to have beclouded rather than clarified the state of the law.” Walker
complained that “courts, in the Mississippi basin especially, seem to
cherish a mental picture of shrewd, sharp, scheming directors craft-
ily trading with inexperienced, female, infant, defective stockholders.”
Walker worried that a ban on insider trading “would be most unsettling
if enforced in financial centers, or respecting great corporations with
hosts of stockholders.”35
Others thought the trend was a positive development. Insider trad-
ing “offends the moral sense,” declared H. L. Wilgus of the University
of Michigan. “No shareholder expects to be so treated by the director he
selects; no director would urge his friends to select him for that reason;

34 Stewart v. Harris, 77 P. 277, 279–​81 (Kan. 1904); Hotchkiss v. Fischer, 16 P.2d


531, 534 (Kan. 1932).
35 Jacquith v.  Mason, 156 N.W. 1041 (Neb. 1916); Dawson v.  National Life
Ins. Co., 157 N.W. 929 (Iowa 1916); Clarence D.  Laylin, “The Duty of a
Director Purchasing Stock,” Yale Law Journal 27 (1918): 732, 739–​40; Roberts
Walker, “The Duty of Disclosure by a Director Purchasing Stock from His
Stockholders,” Yale Law Journal 32 (1923): 639–​40.
  265

Inside Information 265

that the law yet allows him to do this, does more to discourage legiti-
mate investment in corporate shares than almost anything else.” Adolf
Berle approvingly predicted that the recent cases classifying directors
as trustees for the shareholders “are merely specific applications of a
much wider fiduciary relationship which will ultimately transform the
publicly financed corporations of today into organisms having many of
the characteristics of investment trusts.”36
The Securities Exchange Act of 1934 addressed the propriety of in-
sider trading only indirectly. The act did not prohibit insiders from
buying or selling stock or require them to disclose their inside informa-
tion before trading. Instead, section 16(a) of the act required directors,
officers, and large shareholders (those who own more than 10 percent
of the stock) to file monthly statements revealing any purchases or sales.
This requirement was “based upon the theory that if such transactions
were publicized they would be discontinued,” explained Securities and
Exchange Commission Chairman Donald Cook. In addition, “for the
purpose of preventing the unfair use of information,” section 16(b) of
the act barred directors, officers, and large shareholders from profiting
from purchases and sales (or sales and purchases) within a six-​month
span. The corporation itself could sue to recover any such short-​swing
profits. Finally, section 16(c) of the act prohibited directors, officers,
and large shareholders from making short sales of shares they did not
yet own.37
This was an extremely oblique way of preventing insider trading.
The Securities Exchange Act applied only to the largest corporations,
those traded on a stock exchange, so it did not limit insider trading in
the shares of smaller companies. As for the largest corporations, the act
made insider trading more difficult. Insiders could not profit secretly or

36 H.  L. Wilgus, “Purchase of Shares of a Corporation by a Director from a


Shareholder,” Michigan Law Review 8 (1910): 297; A. A. Berle Jr., “Publicity
of Accounts and Directors’ Purchases of Stock,” Michigan Law Review 25
(1927):  831. Berle would later make a similar argument in Adolf A.  Berle
Jr. and Gardiner C.  Means, The Modern Corporation and Private Property
(New York: Macmillan, 1932), 327–​30.
37 Donald C. Cook and Myer Feldman, “Insider Trading Under the Securities
Exchange Act,” Harvard Law Review 66 (1953): 386; 48 Stat. 881, 896–​97 (1934).
266

266 Speculation

quickly. But they could still profit, by buying shares before publicizing
good news or selling before publicizing bad, so long as they were willing
to hold the shares (or refrain from repurchasing them) for more than six
months. And as a committee of the American Bar Association pointed
out a few years after the law’s enactment, the act “does not meet the
problem squarely, for it neither compensates the losing party to a trade
nor confines its punitive effect to actual instances where information
confidentially obtained is used for private profit.”38 Section 16 failed to
reach many instances of insider trading, while it forced insiders to dis-
gorge profits even if they were not acquired by virtue of inside informa-
tion. In retrospect, the strategy pursued in the Securities Exchange Act
can be understood as a snapshot of informed opinion during an era of
transition. Just as the states were dividing over the propriety of insider
trading, the federal government occupied an intermediate position as
well, neither fully approving nor fully forbidding it.

We Are Against Fraud, Aren’t€We?


The legality of insider trading before the mid-╉twentieth century was
a special case of a more general rule of caveat emptor governing com-
mercial transactions. The general rule would change over the course of
the century, however, and as it did, so did the rules governing insider
trading.
One of the overarching legal developments of the twentieth century
was the gradual loss, in a wide variety of contexts, of the ability to
exploit an informational advantage to the detriment of one’s contrac-
tual counterparts. Merchants were no longer allowed to sell defective
products: they were held to an implied warranty of the quality of their
wares, and they were made strictly liable for the harms their products
caused to consumers. House builders and house sellers were required to
reveal unfavorable information to prospective purchasers. Mandatory
disclosures became familiar parts of commercial life—╉disclosures of the
ingredients in food, the attributes of automobiles, the terms of credit

38 “Report of the Special Committee on Securities Laws and Regulations,”


Annual Report of the American Bar Association 66 (1941): 356.
  267

Inside Information 267

cards, and much more. If caveat emptor was the background rule of
commercial law in the late nineteenth century, disclosure—​the oppo-
site of caveat emptor—​had become the background rule by the late
twentieth.39
These legal changes reflected a growing ethical interest in protect-
ing consumers against sharp practices by knowledgeable insiders in the
marketplace. In the stock market, ordinary shareholders were consum-
ers of a sort; they were buying (and selling) a product they often knew
much less about than the person on the other side of the trade. The
law’s treatment of insider trading began to change accordingly.
Courts in rural states had already condemned insider trading, but in
the late 1940s, the practice came in for condemnation by judges in urban
states and eastern financial centers as well. The California Supreme
Court held that “an officer, in buying or selling to a shareholder, must
inform him of those matters relating to the corporate business of which
the officer has knowledge.” A court in Chicago required a director to
reveal his advance knowledge of the sale of the corporation’s assets before
buying stock. The Delaware Court of Chancery, an important decision
maker because of the many corporations chartered in Delaware, held
that a corporate employee could not purchase the corporation’s stock
when he knew that the price would soon rise because the corporation
itself planned to purchase its own stock. Louis Loss, the associate gen-
eral counsel for the Securities and Exchange Commission, concluded
in 1951 that “the so-​called ‘majority’ view,” which permitted insiders to
buy and sell freely without disclosure, “is gradually giving way to the
generally growing feeling of responsibility of corporate insiders—​the
development of a status of ‘trusteeship’ in a non-​technical sense.”40
The SEC helped this development along. In 1942, the SEC pro-
mulgated Rule 10b-​5, a catch-​all provision that prohibited “any act,

39 William L.  Prosser, “The Implied Warranty of Merchantable Quality,”


Minnesota Law Review 27 (1943): 117–​68; Alan M. Weinberger, “Let the Buyer
Be Well Informed? Doubting the Demise of Caveat Emptor,” Maryland Law
Review 55 (1996): 387–​424.
40 Hobart v. Hobart Estate Co., 159 P.2d 958, 970 (Cal. 1945); Agatucci v. Corradi,
63 N.E.2d 630, 632 (Ill. App. 1945); Brophy v. Cities Service Co., 70 A.2d 5, 8
(Del. Ch. 1949); Louis Loss, Securities Regulation (Boston: Little, Brown and
Co., 1951), 826.
268

268 Speculation

practice, or course of business which operates or would operate as a


fraud or deceit upon any person.” The rule was not written with insider
trading in mind. “I was sitting in my office,” recalled the SEC lawyer
Milton Freeman,

and I received a call from Jim Treanor who was then the Director of
the Trading and Exchange Division. He said, “I have just been on
the telephone with Paul Rowen,” who was then the S.E.C. Regional
Administrator in Boston, “and he has told me about the president of
some company in Boston who is going around buying up the stock
of his company from his own shareholders at $4.00 a share, and he
has been telling them that the company is doing very badly, whereas,
in fact, the earnings are going to be quadrupled and will be $2.00 a
share for the coming year. Is there anything we can do about it?”

The Boston company president was committing a garden-​variety fraud


by lying to the shareholders about the company’s prospects, but noth-
ing in the SEC’s rules covered the situation. The SEC lawyers hast-
ily put together a new rule prohibiting all frauds and presented it to
the agency’s five commissioners. “All the commissioners read the rule,”
Freeman remembered. “Nobody said anything except Sumner Pike
who said, ‘Well,’ he said, ‘we are against fraud aren’t we?’ That is how
it happened.”41
Although the new rule did not mention insider trading, within a year
the SEC had already concluded that insider trading violated the rule.
In 1943, the SEC investigated the Ward La France Truck Corporation,
which was making unprecedented profits from truck sales to the military
during the war. The company planned to sell its entire business to the
Salta Corporation, a transaction that would greatly increase the value
of the shares in the Ward La France Corporation. Without disclosing
this plan to shareholders, both companies were busy buying up Ward
La France shares at prices much lower than the shareholders would have
received if they had known to wait for the merger. The SEC determined
that these share purchases “unaccompanied by appropriate disclosure of

41 Milton Freeman, “Administrative Procedures,” Business Lawyer 22 (1967): 922.


  269

Inside Information 269

material facts” constituted a violation of the new Rule 10b-​5.42 This was
just an administrative decision, not the judgment of a court, but it was
important nonetheless, because as a practical matter many of the cases
before the SEC never reached a court. In the view of the SEC, at least,
insider trading was now illegal.
Investors soon began using the rule to sue insiders to recover the
insiders’ profits from such trades. In 1946, the managers of the Western
Board and Paper Company bought company stock from fellow share-
holders without telling the shareholders that they had reached an agree-
ment to sell the company’s plant and equipment to another firm. The
shareholders sued the managers under Rule 10b-​5 and won. “Under
any reasonably liberal construction” of the rule, the judge explained,
“these provisions apply to directors and officers who, in purchasing the
stock of the corporation from others, fail to disclose a fact coming to
their knowledge by reason of their position, which would materially
affect the judgment of the other party to the transaction.” In a similar
case a few years later, another judge declared that “the rule is clear. It is
unlawful for an insider, such as a majority stockholder, to purchase the
stock of minority stockholders without disclosing material facts affect-
ing the value of the stock, known to the majority stockholder by virtue
of his inside position.” The judge reasoned that “the duty of disclosure
stems from the necessity of preventing a corporate insider from utiliz-
ing his position to take unfair advantage of the uninformed minority
stockholders. It is an attempt to provide some degree of equalization of
bargaining position.” By the late 1950s, the leading treatise on closely
held corporations advised the victims of insider trading to sue under
Rule 10b-​5 rather than state common law, because the rule imposed
stricter duties of disclosure on insiders.43
The norms of corporate insiders changed accordingly. In the early
twentieth century, Donald Cook recalled, profits from insider trading

42 In the Matter of Ward La France Truck Corp., 13 S.E.C. 373 (1943).


43 “The Prospects for Rule X-​ 10B-​5:  An Emerging Remedy for Defrauded
Investors,” Yale Law Journal 59 (1950): 1140–​49; Kardon v. National Gypsum
Co., 73 F.  Supp.  798, 800 (E.D. Pa. 1947); Speed v.  Transamerica Corp., 99
F. Supp. 808, 828–​29 (D. Del. 1951); F. Hodge O’Neal, Close Corporations: Law
and Practice (Chicago: Callaghan & Co., 1958), 2:156; see also Elvin R. Latty,
“The Aggrieved Buyer or Seller or Holder of Shares in a Close Corporation
270

270 Speculation

had been “regarded by members of the financial community as one of


the usual emoluments of office.” But a 1961 survey conducted by the
Harvard Business Review revealed a very different picture. The Review
posed this scenario to seventeen hundred executives: “Imagine that you
are a member of the board of directors of a large corporation. At a
board meeting you learn of an impending merger with a smaller com-
pany which has had an unprofitable year, and whose stock is presently
selling at a price so low that you are certain it will rise when news of the
merger becomes public knowledge.” A few decades earlier, this question
would probably not even have been asked in a survey of ethics, because
it would hardly have been thought to raise an ethical issue. But in 1961,
56 percent of respondents declared that they would “do nothing” in this
situation. Only 42 percent said they would buy shares in the smaller
company for themselves. Only 61  percent thought that the “average
executive” would buy shares for himself. A  few decades earlier, both
figures would have been close to 100 percent.44
While the Harvard Business Review was conducting its survey, the
SEC was extending the reach of its prohibition on insider trading. At a
1959 board meeting of the Curtiss-​Wright Corporation, a manufacturer
of airplane parts, the directors voted to cut the company’s quarterly
dividend by nearly half. Curtiss-​Wright’s normal policy, in keeping
with the rules of the New York Stock Exchange, was to send telegrams
immediately to the stock exchange and to the Dow Jones News Ticker
Service, so that everyone would know about the dividend. (The stock
exchange was marketing itself to small investors as a safe place for their
money.45 The notification rule was part of that effort.) In this instance,
however, there was an inadvertent delay in sending the telegrams. The
board vote took place at 11:00 a.m., but the telegram did not reach Dow

Under the S.E.C. Statutes,” Law & Contemporary Problems 18 (1953): 527–​32;


Robert M.  Desky, “Corporations:  Fiduciary Duties:  Application of SEC
Rule X-​10B-​5 to Prevent Nondisclosure in the Sale of Corporate Securities,”
California Law Review 39 (1951): 429–​39.
44 Cook and Feldman, “Insider Trading,” 386; “How Ethical Are Businessmen?,”
Harvard Business Review, July–​August 1961, 16.
45 Janice M. Traflet, A Nation of Small Shareholders: Marketing Wall Street after
World War II (Baltimore: Johns Hopkins University Press, 2013).
  271

Inside Information 271

Jones until 11:45, and it did not reach the stock exchange until 12:29.
In the interim, a Curtiss-​Wright director named J.  Cheever Cowdin
left the meeting. Cowdin was also a representative of a brokerage firm
called Cady, Roberts and Company. He phoned Robert Gintel, a col-
league at Cady Roberts, and told Gintel about the reduction of the div-
idend. Gintel promptly sold Curtiss-​Wright stock on behalf of several
of the brokerage firm’s customers, and sold the stock short on behalf of
several others. One of the customers was Gintel’s wife. When the news
of the dividend was finally made public, Curtiss-​Wright’s stock price
plummeted. By selling as soon as Cowdin told him about the dividend,
Gintel had saved his clients a bundle.
William Cary had just been appointed by President Kennedy as
chairman of the SEC. Cary had spent most of his career as a law profes-
sor at Columbia. He had a particular interest in stamping out insider
trading. The SEC’s earlier cases had involved insiders who traded with
shareholders in face-​to-​face transactions, and Cary recognized that the
Cady Roberts case presented a chance to extend the ban on insider trad-
ing to trades on a stock exchange, where the person on the other side
of the trade was unknown. He took the opportunity to write a lengthy
opinion spelling out the SEC’s view of the law. “The securities acts
may be said to have generated a wholly new and far-​reaching body of
Federal corporation law,” Cary declared. Part of that law, he explained,
was a ban on insider trading. “We and the courts have consistently
held that insiders must disclose material facts which are known to them
by virtue of their position but which are not known to persons with
whom they deal,” he explained. This was just as true for transactions on
a stock exchange as for face-​to-​face transactions. “It would be anoma-
lous indeed,” Cary insisted, “if the protection afforded by the antifraud
provisions were withdrawn from transactions effected on exchanges,”
which were the “primary markets for securities transactions.” Because
neither Cowdin nor Gintel had disclosed the dividend reduction before
Gintel sold Curtiss-​Wright stock, they had broken the law. Cowdin had
died shortly after the incident, so the SEC could not punish him, but
the SEC suspended Gintel from the stock exchange for twenty days.46

46 Joel Seligman, The Transformation of Wall Street (Boston: Houghton Mifflin,


1982), 344–​45; In the Matter of Cady, Roberts & Co., 40 S.E.C. 907 (1961).
272

272 Speculation

The case caused considerable alarm in the New York financial com-


munity. It was the first in which an upstanding member of that com-
munity was punished for insider trading. Robert Gintel was a recent
graduate of Harvard Business School who had served in the air force.
He would go on to run a small mutual fund named after himself,
a fund small enough that he could invite all the shareholders to an
annual barbecue at his estate in Greenwich, Connecticut.47 J. Cheever
Cowdin had been the president of Universal Pictures and the Aqueduct
Racetrack, and he had been a Curtiss-​Wright director for more than
thirty years. Unlike some of the insiders in the earlier cases, these were
prominent people.
The case was also the first in which the person punished was not
himself a director or officer of the corporation, but rather someone who
had received a tip from a director. Robert Gintel was a stockbroker who
happened to have a partner who sat on the board of a major corpora-
tion. Wall Street was full of brokers and bankers and lawyers in the same
position, people who knew directors but were not directors themselves.
They talked to each other all the time, at work and at play, and some of
their conversation involved information that could affect stock prices.
Indeed, in an era when the commissions stockbrokers charged their cus-
tomers were fixed by the exchange, the brokers could not compete on
price, so one common way of competing was by offering tips to their
customers.48 If Gintel had broken the law, who else might have done
the same?
“The decision raises many troublesome questions and encompasses
various shadowy areas,” two New York lawyers worried. Investment
advisors were supposed to gather information about companies, so
they could recommend which stocks to buy. But now the SEC was
punishing a broker for doing just that. If giving good advice would
only “expose the well advised to potential penalties,” they feared,
“the precept of ‘investigate and invest’ may be thwarted.” The lawyer
Carlos Israels wondered whether ordinary small talk had become

4 7 New York Times, 19 Mar. 1990, D6; New York Times, 13 Aug. 1982, A14.


48 Stanislav Dolgopolov, “Insider Trading, Chinese Walls, and Brokerage
Commissions:  The Origins of Modern Regulation of Information Flows in
Securities Markets,” Journal of Law, Economics & Policy 4 (2008): 311–​67.
  273

Inside Information 273

illegal. “How about the case in which the partner of this brokerage
house has lunch with his friend, the vice-​president, and is told, ‘You
know, Bill, that new combustion engine that we wrote about in our
last report to stockholders, we spent fifteen million dollars on it,
but it turned out to be a complete dud.’ ” Had the broker become
an “insider” who owed a duty to the corporation not to trade in
its stock? And what about his duty to the clients whose money he
was investing—​which took priority? The Chicago lawyer W. McNeil
Kennedy was a former regional administrator of the SEC, but even
he was positive that the decision “foreshadows grave problems for the
securities industry.”49
The view was quite different outside Wall Street. Louis Loss had left
the SEC for Harvard Law School, where he would teach for decades. “I
think if Professor Cary does nothing else at the SEC he has earned his
pay in Cady,” Loss exclaimed. “I view that as a landmark in the law.” As
Loss saw it, Cary had said “officially what needed to be said a long time
ago. Of course it opens problems—​what does not?”50
Another set of problems was opened a few years later, when the US
Court of Appeals for the Second Circuit, the federal appellate court
with jurisdiction over New York and thus the most important to secu-
rities law, decided the Texas Gulf Sulphur case. Directors and officers
of a mining company bought company stock when they, but not the
public, knew that the company had found unusually rich ores and thus
that the stock price was certain to rise. These directors and officers were
paradigmatic insiders, and their knowledge of the ores was informa-
tion as important to the stock price as any information could be. The
difficult thing about the case was not the result, although the case did
mark the first occasion in which a court of appeals blessed the SEC’s
now two-​decades-​old view that Rule 10b-​5 barred insider trading. The
problems were caused instead by the court’s broad language, which
suggested a prohibition on insider trading even wider than the SEC

49 F. Arnold Daum and Howard W. Phillips, “The Implications of Cady, Roberts,”


Business Lawyer 17 (1962): 940, 959; William L. Cary, “Recent Developments in
Securities Regulation,” Columbia Law Review 63 (1963): 866; William L. Cary,
“The Direction of Management Responsibility,” Business Lawyer 18 (1962): 79.
50 Cary, “Recent Developments,” 861.
274

274 Speculation

had urged. “The Rule is based in policy on the justifiable expectation of


the securities marketplace that all investors trading on impersonal ex-
changes have relatively equal access to material information,” the court
declared. “Insiders, as directors or management officers are, of course,
by this Rule, precluded from so unfairly dealing, but the Rule is also
applicable to one possessing the information who may not be strictly
termed an ‘insider.’ ” It was not one’s status as an insider that triggered
a ban on trading; it was rather the possession of information that the
public did not yet have. “Thus, anyone in possession of material inside
information must either disclose it to the investing public, or, if he is
disabled from disclosing it in order to protect a corporate confidence,
or he chooses not to do so, must abstain from trading.” The purpose of
the rule was that “all investors should have equal access to the rewards
of the participation in securities transactions,” the court concluded. “It
was the intent of Congress that all members of the investing public
should be subject to identical market risks.”51
This was as strong an assertion of commercial morality as the
American legal system had seen since the cases rejecting the caveat
emptor doctrine in the early nineteenth century. Read literally, it would
mean that no securities could be bought or sold unless both parties had
equal knowledge of all information affecting the price. The Texas Gulf
Sulphur rule extended far beyond the devious director who conned his
shareholders by buying up their stock on the cheap. It covered everyone
who knew more than the person on the other side of the transaction.
Once the law had allowed anyone to speculate with superior informa-
tion, but now the pendulum had swung completely in the opposite
direction.
“It’s going to have a hell of an impact on the financial world,” pre-
dicted Sterry Waterman, the judge who wrote the opinion. He was
right. The printed text of the opinion broke all the sales records for
the company with the Second Circuit printing contract. “This thing
is hot as hell,” exclaimed the printing company’s manager. “Almost ev-
erybody and his uncle who knows anything about law or the stock

51 Securities and Exchange Commission v. Texas Gulf Sulphur Co., 401 F.2d 833,
848, 851–​52 (2d Cir. 1968) (en banc).
  275

Inside Information 275

market wants it.” Lawyers, brokers, investors—​everyone had to read


Texas Gulf Sulphur. The brokers did not like it one bit. “This makes for
utter confusion,” one complained. Another reported that “it’s got the
whole street disturbed.” Brokers had long been accustomed to ferreting
out news and passing tips along to their clients, because the brokers
had not considered themselves insiders with respect to the corporations
whose stock they bought and sold. But if the brokers could not use
the information they discovered, how would they make a living? One
broker lamented that he and his colleagues would be nothing but “ticket
sellers,” simply buying whatever stocks their clients requested. Their
alarm grew two weeks later, when the SEC commenced administrative
proceedings against fourteen employees of the Merrill Lynch broker-
age firm for telling clients about an earnings decline at the Douglas
Aircraft Company before the decline was reported to the public. The
SEC had “put Wall Street on notice,” reported the New  York Times,
“that a relatively widespread practice no longer will be tolerated.” But
a famine for brokers could be a feast for members of other professions.
Public relations firms foresaw an increased need for their services, be-
cause of the specter of liability from the mistimed release of corporate
information. “Every sensible company will have to consider its finan-
cial public relations in a new light,” chortled the chairman of Hill &
Knowlton. Lawyers flocked to conventions to discuss how to advise
their clients. The Practising Law Institute held a special session on Texas
Gulf Sulphur and got the biggest turnout the organization had ever
seen. “Nearly every company is re-​examining its policy in this area,”
one satisfied lawyer explained.52
Eventually, however, the Supreme Court narrowed the class of
people barred from insider trading. In two opinions in the early 1980s,
the court rejected the view that the law required both parties to a
transaction to have equal information. Both opinions were written by
Justice Lewis Powell, who had been a corporate lawyer before joining
the court and who had formed a strong belief that the law imposed
too many restrictions on businesses. In Chiarella v.  United States,

52 Wall Street Journal, 14 Aug. 1968, 3; 19 Aug. 1968, 5; New York Times, 15 Aug.
1968, 54; 18 Aug. 1968, F1; Wall Street Journal, 3 Sept. 1968, 30; New  York
Times, 1 Sept. 1968, E5; 25 Aug. 1968, F14; 13 Oct. 1968, F1; 11 Oct. 1968, 67.
276

276 Speculation

the court reversed the conviction of a printer who had learned of an


impending takeover while printing the necessary documents. There
was no “general duty between all participants in market transactions
to forgo actions based on material, nonpublic information,” Powell
declared. Rather, insider trading was unlawful only where the trader
owed a duty to disclose the information to another. Three years later,
in Dirks v. S.E.C., the court ruled in favor of a broker who had re-
ceived tips about a corporate fraud from corporate employees, and
who then spread news about the fraud to others, who sold the com-
pany’s stock before the news became public. Such “tippees” broke the
law, the court explained, only where the insider provided the tip to
profit from it. Powell again took the opportunity to reject the conten-
tion that the law requires equal information among all traders.53
Many more difficult questions remained.54 Who owed duties to
whom not to exploit inside information? What was the source of these
duties, and what was their content? Were there also non-​duty-​based
grounds for imposing liability for insider trading? These questions were
rendered even more difficult by Congress’s failure to answer them in
legislation. By the early twenty-​first century there was a large and com-
plex body of insider trading law, all of which had been created by judges
who had little to guide them apart from the 1942 SEC rule banning
“fraud or deceit.” The judges had rejected both extremes. The law did
not permit insider trading across the board, as it had for much of the
nineteenth century, nor did it completely forbid traders from exploit-
ing their superior knowledge, as it briefly had in the mid-​twentieth.
The law of insider trading was somewhere in the middle, but it was no
easy matter to say exactly where. “The closer one looks at insider trad-
ing law,” one expert despaired, “the messier it appears.”55

53 Adam C. Pritchard, “Justice Lewis F. Powell, Jr., and the Counterrevolution in


the Federal Securities Laws,” Duke Law Journal 52 (2003): 841–​949; Chiarella
v. United States, 445 U.S. 222, 233, 230 (1980); Dirks v. S.E.C., 463 U.S. 646,
655–​62 (1983).
54 For an engaging discussion of these questions, see Stephen M.  Bainbridge,
Securities Law: Insider Trading (New York: Foundation Press, 1999).
55 Sung Hui Kim, “Insider Trading as Private Corruption,” UCLA Law Review
61 (2014): 945.
  277

Inside Information 277

A handful of high-​profile prosecutions in the 1980s made insider


trading the white-​collar crime perhaps most familiar to the general
public. Prominent people charged with insider trading during the era
included the entrepreneur Ivan Boesky and the investment bankers
Michael Milken and Dennis Levine. Insider trading was a major theme
of the popular 1987 film Wall Street, which earned Michael Douglas an
Academy Award for best actor. The public seemed to take a greater in-
terest in insider trading than in more humdrum offenses like embezzle-
ment or tax fraud, most likely because the existence of insider trading
confirmed the widespread sense—​one that has always been deeply in-
grained in American culture—​that the financial markets are rigged in
favor of crafty insiders. Despite the attention these cases received, the
empirical evidence suggests that they did little to reduce the frequency
or the profitability of insider trading.56
Meanwhile, there was a raging debate among economists and law
professors over whether insider trading should be illegal at all. The
debate was sparked in 1966 by the law professor Henry Manne, who
argued that insider trading made the stock market more efficient by
aligning the price of stocks with their true value, and that allowing
entrepreneurs to profit from trading was the most effective way of com-
pensating them for their contributions. Other market participants were
not harmed by insider trading, Manne contended. Long-​term inves-
tors would not care whether stock prices moved before or after the
announcement of news, because they were unlikely to buy or sell in the
interim. And short-​term investors who transacted with an insider were
not hurt, because they would have bought or sold anyway.57 In the old
days, when an insider dealt directly with a handful of shareholders, any
gain to the insider was a loss to the shareholder, who would have held
on to his shares had he not sold them to the insider. But in the modern
impersonal market, an insider could buy shares only from a shareholder
who had already decided to put them up for sale.

56 H.  Nejat Seyhun, “The Effectiveness of the Insider-​Trading Sanctions,”


Journal of Law and Economics 35 (1992): 149–​82.
57 Henry G.  Manne, Insider Trading and the Stock Market (New  York:  Free
Press, 1966).
278

278 Speculation

Manne’s argument was attacked immediately, by critics who charged


that investors were harmed by insider trading when they sold at prices
less advantageous than they would have otherwise received. The critics
worried about the loss of public confidence and participation in the
market that might result from a perception that insiders had an advan-
tage.58 The debate has continued in the pages of economics journals
and law reviews for half a century.
But this academic debate about the economic effects of insider trad-
ing has made scarcely any dent in the law. Insider trading became il-
legal because it was widely perceived as unfair, not because of its effects
on the efficiency of markets. Whether it is right to use one’s superior
knowledge to gain an edge in the marketplace is one of our oldest ques-
tions of commercial morality. As the law of insider trading has devel-
oped over the past century, speculators have been held to the same
standards as other merchants. They have gradually lost much, but not
all, of their former ability to exploit inside information.

58 A good example of the early criticism of Manne’s thesis is Roy A. Schotland,


“Unsafe at Any Price: A Reply to Manne, Insider Trading and the Stock Market,”
Virginia Law Review 53 (1967): 1425–​78.
8
Q
Speculation or Investment?

What is the difference between speculation and investment? As we


have seen, this question lurked behind the commercial regulation of
the nineteenth and twentieth centuries, much of which was intended
to restrain the former but encourage the latter. But there were also
some contexts in which the question was in the foreground. In these
situations, lawyers, judges, and legislators routinely had to classify par-
ticular transactions as instances of speculation or investment, because
important legal consequences flowed from the decision. Over the nine-
teenth and twentieth centuries, a great deal of thought was devoted
to an immediately practical question: where does investment stop and
speculation begin?
One of these contexts was the “prudent man” rule that governed
the conduct of trustees. A trustee is a person who manages assets for
the benefit of someone else. Courts declared over and over again that
a trustee was required to act as a prudent man would. In particular,
a trustee had to invest the trust’s assets, but under no circumstances
could he speculate with them. When a trustee used the trust’s funds
to buy an asset, like stock in a corporation, and the value of that asset
declined, the trustee was not personally responsible for the loss if his
purchase of the asset was an investment, but if it was a speculation, the
trustee had to make up the loss out of his own pocket. A lot turned on
precisely how to draw the line between speculating and investing. But
the prudent man rule was much easier to state in the abstract than to

279
280

280 Speculation

apply in concrete cases. How was a judge supposed to tell whether this
or that transaction was speculation or investment?
A second context was the capital gains tax. A capital gain is income
from the sale of an asset at a price higher than was paid for it. Capital
gains have been taxed as income since the birth of the modern federal
income tax in the early twentieth century. For nearly as long, Congress
has sought to impose higher tax rates on speculators than on investors,
so as not to deter investment. Of course, to do so requires some method
of telling the two groups apart. The drafters of the tax code have thus
faced the same question as judges evaluating the transactions of trust-
ees. How could they distinguish speculation from investment?
A third context in which the question arose was in the design of a
securities transaction tax. Since the late nineteenth century, the federal
government and the state of New York have, off and on, imposed small
taxes on the sale of stock. The primary goal of these taxes has been
to raise revenue, not to deter speculation. There have been recurring
suggestions, however, that the tax should be increased to the point
where it would make speculation uncomfortably expensive, as a way
of reducing the amount of it. And these suggestions have inevitably
raised the same old nagging questions:  Is it possible to deter specu-
lation without also deterring investment? Can one be distinguished
from the other?

The Prudent€Man
When the wealthy Boston merchant John McLean died in 1823, he left
the then-╉enormous sum of $50,000 to a trust, the income from which
was to support his widow, Ann, for the rest of her life. Upon Ann’s
death, the fund would be divided evenly between Harvard College and
the Massachusetts General Hospital. McLean placed the money in the
care of two other Boston merchants, Jonathan and Francis Amory, who
used it to purchase the stock of a bank, an insurance company, and
two textile firms. A  few years later, the value of the insurance stock
had fallen from $16,000 to $12,000, and the shares in the two textile
companies, which the Amorys had bought for $25,000, were worth
only $17,000. Harvard and the Massachusetts General Hospital were
not pleased. They sued Francis Amory (Jonathan had died), and they
  281

Speculation or Investment? 281

argued that he was responsible for making up the difference. Rather


than making “safe and productive” investments in government bonds
or bank shares, they contended, Amory had speculated “in trading
companies, whereby the principal sum was exposed and still continues
to be exposed to great loss.”
But the Massachusetts Supreme Court drew a different line between
investing and speculating. Government bonds could be risky too, Justice
Samuel Putnam pointed out. In times of war, for example, when the gov-
ernment borrowed so much that bondholders feared a default, the value
of outstanding bonds could plummet. And “bank shares may be subject
to losses which sweep away their whole value.” There was simply no place
to put money with absolute safety. “It will not do to reject those stocks
as unsafe, which are in the management of directors whose well or ill
directed measures may involve a total loss,” Putnam reasoned. “Do what
you will, the capital is at hazard.” John McLean himself had owned man-
ufacturing stock—​so much that it amounted to nearly half his fortune—​
and McLean was well known as “a man of extraordinary forecast and
discretion.” The court concluded that Amory had done nothing wrong.
He had been investing, not speculating.1
By the time Ann McLean died in 1834, the trust had not recovered.
Harvard and the Massachusetts General Hospital ended up with less
than $20,000 each. For many years after, they would resent what Amory
had done with their money. The real estate lawyer and Harvard gradu-
ate Nathaniel Ingersoll Bowditch reported decades later: “It is believed
that every Trustee of the Hospital and every Corporator of the College
coincided in opinion, that this investment of the trust-​funds, though
adjudged to be legal, was not made in the exercise of a sound discretion,
and with a due regard to the rights of all parties.”2
But if Harvard and the hospital lost their case, their argument
would eventually win general acceptance. Only a handful of American
courts in the nineteenth century would agree with Judge Putnam that
shares of corporations were investments suitable for trustees. One
was the highest court in Mississippi, which recalled that stock in the

1 Harvard College v. Amory, 26 Mass. 446, 460–​63 (1830).


2 Nathaniel I. Bowditch, Extracts from a History of the Massachusetts General
Hospital 1810–1851 (s.l.: privately printed, 1899), 18.
282

282 Speculation

Commercial Bank of Natchez “was considered safe and profitable …


by the most prudent men in the country” when a trustee bought
$5,000 of it, even after the stock became worthless a few years later.
Most American judges, however, would agree with Harvard and the
Massachusetts General Hospital that corporate stock was too specula-
tive for trustees to touch. “Stocks are liable to great depression,” cau-
tioned New Jersey chancellor Peter Vroom. “The prospects of peace or
war, to say nothing of the agitations caused by the spirit of restless and
unprincipled speculation, are constantly causing a fluctuation in the
stocks.” Vroom warned the state’s trustees that “the stock of private
companies is not considered safe.” A New York court ordered a trustee
for four children to compensate the children for the trust’s losses on
shares in the Dutchess County Bank. The property of a trust “cannot
be jeopardized and wasted by hazardous speculations,” the court lec-
tured. “In a great majority of cases speculations in this country have
been productive of disastrous consequences.” Such transactions might
sometimes yield high incomes for the trust’s beneficiaries, but “the law
regards the certainty of an income for persons thus situated, more than
its magnitude.”3 Real estate and government bonds were investments,
but corporate stock was a speculation.
This view was orthodoxy by the middle of the nineteenth century, as
was made evident during the protracted litigation over the will of the
Philadelphia banker Stephen Girard, who was probably the wealthiest
person in the United States when he died childless in 1831. Girard’s will
is most famous for its bequest of $2 million to the city of Philadelphia
for the establishment of a school for orphans, at which Girard specified
that no minister of any sect would be allowed to teach. This aspect of
the will was challenged by some of Girard’s relatives, who hoped to get
the $2 million for themselves, but it was upheld by the US Supreme
Court in 1844.4 Litigation over the will did not end there, however.
Another clause of the will left $50,000 to a trust for the benefit of
Girard’s niece, Marie Antoinette Hemphill. Hemphill would receive

3 Smyth v. Burns, 25 Miss. 422, 427 (1853); Gray v. Fox, 1 N.J. Eq. 259, 266–​67
(N.J. Ct. Ch. 1831); Ackerman v. Emott, 4 Barb. 626, 646–​47 (N.Y. Super. 1848).
4 Vidal v. Philadelphia, 43 U.S. 127 (1844).
  283

Speculation or Investment? 283

the interest on this sum for her life, and then on her death the money
would be divided among her children. The trustees used part of the
money to buy shares of the Bank of the United States. Girard himself
had been intimately involved in creating the bank, had been one of
the bank’s directors, and had owned a significant fraction of the bank’s
stock. No financial institution, with the exception of Girard’s own pri-
vate bank, was as closely associated with Stephen Girard as the Bank of
the United States. But the trustees had the misfortune to buy the shares
just before the panic of 1837 shut the bank’s doors for good.
“We have not a doubt, that the investment was made in perfect good
faith,” explained the Pennsylvania Supreme Court. But the court re-
quired the trustees to compensate the Hemphills for the loss, because
they had speculated with the Hemphills’ money. “If a trustee may
throw off his responsibility by investing the trust fund in the stock
of a corporation, what security is left?” the court asked. “There is no
reason why the trustee should not make the investment in some secu-
rity which cannot fail. It is just as convenient.” The court held that the
trustees should have invested in land or in government bonds, which
were less likely to lose their value. “It is better for trustees that the rule
of their conduct should be clearly defined and well understood,” the
court concluded. “A plain path, though it may be a narrow one, is safer
to walk in than a trackless waste, where no man can be sure that he is
on the right course.”5
A few years later the New Hampshire Supreme Court was just as
dubious about a trustee’s purchase of stock in the Boston, Concord,
and Montreal Railroad, stock that soon became worthless. “Safety is
the primary object to be secured in an investment of this kind,” the
court noted, “and the trustee is not chargeable with an income that
cannot be realized without hazard to the fund.” The construction of
a new railway line was the very opposite of safe. “Now it seems to us
that the building of a railroad must, from the very inherent nature of
the enterprise, have at all times been regarded by prudent men as a
doubtful adventure,” clucked the court. Even if the purchase of the
stock “was advised and approved by … judicious and prudent men

5 Hemphill’s Appeal, 18 Pa. 303, 306 (1852).


284

284 Speculation

whom the trustee consulted,” it was simply not the role of a trustee to
put funds into any business corporation.6
By 1872, when the Salem, Massachusetts, lawyer Jairus Perry pub-
lished his Treatise on the Law of Trusts and Trustees, he could report
that nearly all courts drew a sharp line between business corporations,
which were too speculative for trustees, and safe investments like land
and government bonds. Money put into business corporations “runs
the risks and chances of trade, business, and speculation,” Perry ex-
plained. “Calamities that depress public credit seldom occur, while the
risks of trade are constant. It would seem to be the wiser course to
withdraw the funds, settled for the support of women, children, and
other parties who cannot exercise an active discretion in the protection
of their interests, as much as possible from the chances of business.”7
The prudent man would stay away from such risks.
What made corporate stock too speculative? The problem with cor-
porations, a New York court explained, was that a stockholder had to
put his faith in the wisdom of other people, the directors and officers of
the firm, who were beyond the effective supervision of the trustee. “The
moment the fund is invested in bank, or insurance, or railroad stock, it
has left the control of the trustees,” the court reasoned. “Its safety and
the hazard, or risk of loss, is no longer dependent upon their skill, care,
or discretion, in its custody or management.” A trustee would be abdi-
cating his solemn responsibility if he delegated his duty of care to the
managers of such an enterprise. Of course, prudent men bought shares
of corporations all the time, with their own money, but in the court’s
view that was no reason for trustees to speculate with other people’s
money. “If it be said, that men of the highest prudence do, in fact, invest
their funds in such stocks,” the court concluded, “the answer is” that “in
their private affairs, they do, and they lawfully may, put their principal
funds at hazard; in the affairs of a trust they may not.”8
In many cases, the problem courts perceived was that the particular
corporation involved was so new that there was not yet any way to

6 Kimball v. Redding, 31 N.H. 352, 374–​76 (1855).


7 Jairus Ware Perry, A Treatise on the Law of Trusts and Trustees (Boston: Little,
Brown and Co., 1872), 412.
8 King v. Talbot, 40 N.Y. 76, 88–​89 (1869).
  285

Speculation or Investment? 285

ascertain whether it would be successful. In Maine, the trustee for a


child bought shares of the Union Packing Company, a Portland firm
that was only a few years old. The company had scarcely any capital on
hand because it had spent nearly all of its money to lease factories. Its
business “was sensitive to changes in the markets, the crops, the migra-
tions of fish, the weather, etc., which are always variable and uncertain,”
lectured the Maine Supreme Court. The Union Packing Company
“was liable to be overwhelmed at the first unfavorable turn in affairs,”
which indeed happened, leaving its stock worthless. The court required
the trustee to cover the loss himself, because “a mere business chance or
prospect, however promising, is not a proper place for trust funds.” The
Alabama Supreme Court reached the same conclusion about a trustee’s
purchase of “stock of a recently-​formed land company, organized, as
is alleged, as a purely speculative venture.” When a gullible dentist in
Rochester, New York, acting as a trustee for his own son, bought stock
in the Tex-​Lahoma Oil Company and in Herschell-​Spillman Motors,
a judge mocked the dentist as a man “who believes that an investment
is something that pays abnormally high rates of interest and promises
excessive profits.” By the time of litigation, the two companies’ short
lives had long since ended. “Under no pretext whatever may trust funds
be used for engaging in wild speculation,” the judge intoned. “To so
use them is illegal.” Even the Union Pacific Railroad was too unproven
an enterprise for the Massachusetts Supreme Court to allow a trustee
to buy its shares in the late nineteenth century. “It must have been
manifest to any well-​informed person,” the court declared, “that the
Union Pacific Railroad ran through a new and comparatively unsettled
country; that it had been constructed at great expense, as represented
by its stock and bonds, and was heavily indebted; that its continued
prosperity depended upon many circumstances, which could not be
predicted, and that it would be taking a considerable risk to invest any
part of a trust-​fund in the stock of such a road.”9 Novelty meant risk,
and risk meant speculation rather than investment.

9 Mattocks v. Moulton, 24 A. 1004, 1006–​7 (Me. 1892); Randolph v. East Birmingham


Land Co., 16 So. 126, 129 (Ala. 1894); In re Cady’s Estate, 207 N.Y.S. 385, 390 (N.Y.
App. Div. 1925); Appeal of Dickinson, 25 N.E. 99, 100 (Mass. 1890).
286

286 Speculation

But there were other sources of risk besides novelty. One was dis-
tance. When a trustee in New York purchased a mortgage on land in
Toledo, Ohio, New  York’s highest court took the opportunity to es-
tablish a general rule forbidding trustees from making investments in
other states. When money was sent so far away, the court fretted, “the
proper and prudent knowledge of values would become more difficult
and uncertain; watchfulness and personal care would in the main be
replaced by confidence in distant agents, and legal remedies would have
to be sought under the disadvantages of distance, and before different
and unfamiliar tribunals.” The Missouri Supreme Court was positively
scornful, for the same reason, about a trustee who purchased bonds
issued by the Mexican state of Jalisco. The court was shocked that a
trustee in St. Louis would go “thousands of miles into a foreign country
and among a strange people to obtain for his brother an investment in
6 per cent bonds.”10 A prudent man would keep his money closer to
home, where he could keep an eye on it himself, and where the rules of
the game were familiar and enforced by his own neighbors.
Any number of circumstances could push an investment into the
forbidden territory of speculation. Making an investment with bor-
rowed money, for example, was “purely a speculation,” one New York
judge declared, even where the investment would have been permis-
sible without the borrowing. The judge explained that an investment
is “for the purpose of securing an income for the beneficiaries, while
a speculation is for the purpose of realizing a profit from the trans-
action.” Making an investment in a business the trustee knew little
or nothing about was another circumstance that suggested specula-
tion. When Ezra Curtis died in 1882, he left a friend named Potter
in charge of managing the funds of his widow, Mary, who “had no
business experience whatever” and thus left all the decisions to Potter.
Potter put most of the money into shares of the Housatonic Rolling-​
Stock Company, a manufacturer of railroad cars. “The stock of the
Housatonic Rolling-​Stock Company was a speculative stock, of the
character or value of which it is found that Mr. Potter knew nothing,”

10 Ormiston v. Olcott, 84 N.Y. 339, 344 (1881); Cornet v. Cornet, 190 S.W. 333, 340
(Mo. 1916).
  287

Speculation or Investment? 287

a Connecticut judge sneered, after the share price had plummeted


from $50 to $5. The firm had never even been properly incorporated, “a
fact that Mr. Potter … could easily have ascertained, and was bound
to have ascertained before assuming to advise an investment in it by
Mrs. Curtis.” The court concluded that this was “conduct on the part
of Mr. Potter that is really inexcusable.”11
But if the general rule was that the purchase of corporate stock was
speculation rather than investment, there were always cases in which
courts held the opposite. When Joseph Glover died in Knoxville, he
left a trust for his three little children, in the charge of the Fidelity
Trust & Safety Vault Company. At his death Glover had owned shares
of the East Tennessee National Bank that paid annual dividends of
14 percent, shares that would become part of the trust unless the trust
company had to sell them. The trust company accordingly went to
court to request permission to keep the stock. The court recognized
that the bank “is in a prosperous condition, and as the proof shows
is well managed.” The stock was “producing a considerable income
for the children.” The court gave its blessing. “An investment for pur-
poses of speculation, or in such stocks as pay a dividend one month
and none the next, and that are thrown day by day on the market for
purposes of speculation,” would be forbidden to trustees, the court
reasoned. But when a trustee invests in “a safe banking institution,”
where the trust funds are “secure and the dividends regularly paid,” it
made no sense to require the trustee to sell the stock and reinvest the
proceeds, which “might result in an investment not nearly as profit-
able or secure.” The purchase of stock, at least in this case, was an
investment suitable for trustees, not a speculation, because the trustee
was expecting a steady stream of dividends rather than a rise in the
share price.12
In other cases, courts likewise distinguished investment from specu-
lation, not by condemning all business corporations as speculative, but
by discerning the intentions of the trustee. Where a trustee bought
shares for their dividend income, that was investment, but where he

11 In re Hirsch’s Estate, 101 N.Y.S. 893, 899 (N.Y. App. Div. 1906); Appeal of Potter,
12 A. 513, 514 (Conn. 1888).
12 Fidelity T. & S.V. Co. v. Glover, 14 S.W. 343, 344 (Ky. Ct. App. 1890).
288

288 Speculation

bought shares in the hope that they would appreciate, that was specula-
tion. “The word ‘speculate’ means to purchase with the expectation of
an advance in price,” declared one Ohio judge. “An investment is the
laying out of money with the view of obtaining an income or profit
from the thing bought whether it be an interest in a business, a farm,
stocks or bonds; to place money so that it will be safe and yield a profit.”
On this definition, “investments can be made in stocks and stocks can
be taken as security for loans without the transactions becoming specu-
lations.” A judge in New Jersey similarly reasoned that “a speculative
investment is one in which there is a substantial danger of loss of prin-
cipal balanced by a prospect of appreciation of principal.” Many stock
purchases were speculative, but not all. “Some few common stocks,
because of the ample assets which they represent, the steady earnings
of the company, and the conservatism of its management, fall into the
class of safe investments, and such the trustee may properly retain.”13
There was more than one line that could be drawn between investment
and speculation.
Case by case, trustee by trustee, transaction by transaction, courts
were policing the boundary between investment and speculation, by
applying the vague standard of the prudent man to countless unique
factual situations. By the late nineteenth and early twentieth centuries
it was commonplace to observe that the lines the courts drew were not
all consistent. The prudent man standard “is wholly valueless, except as
an indication of the animus of the court,” complained the Philadelphia
lawyer Jonathan Merrill. “What court could or would name the securi-
ties in which, and in which alone, diligent, careful and prudent men of
discretion and intelligence might place their funds?” “The word ‘specu-
lation’ is not capable of exact legal definition,” admitted Mayo Adams
Shattuck, a Boston trust lawyer. “It is a very difficult task to state what
it is that provides the distinction between the fiduciary attitude and
that of the business man or speculator.” When the Harvard law profes-
sor Austin Wakeman Scott surveyed the field in his magisterial treatise
on the law of trusts, he had to concede that “the line between what

13 State v. Gibbs, 18 Ohio Dec. 694, 695–​96 (Ohio. Ct. Comm. Pleas 1908); Wild
v. Brown, 183 A. 899, 900 (N.J. Ct. Ch. 1936).
  289

Speculation or Investment? 289

constitutes speculation and what constitutes a business man’s risk and


what constitutes a prudent investment is drawn in different places by
different courts.”14 The judges who evaluated the decisions of trustees
were no more able to draw a definitive line between investment and
speculation than anyone else had been.
The line-​drawing problem was complicated by two circumstances that
were present in virtually all of these cases. On one hand, it was widely
recognized that trustees, unless closely watched, often had a greater in-
centive to take risks than did the beneficiaries whose money they were
managing. This was in part a matter of age, gender, and disposition.
Trustees were often men of working age accustomed to taking business
risks, while beneficiaries were often elderly widows or children with no
experience in the business world. One newspaper complained of “these
smart young men, who become the life of trade and speculation—​into
whose hands come the … sacred trusts of widows and orphans.” But it
was also a matter of how trustees were compensated. Many, particularly
the banks that took over much of the trust business, were paid according
to how much income the trusts in their care earned each year. “A banker
is constantly beset by temptation to yield a little of a customer’s safety
in favor of more profits for the bank,” contended the journalist Fred
C. Kelly. “In the trust department, a bank’s charges are based on income.
If an individual investor has only $50,000, his mode of life will be much
the same whether his income is 5 per cent or only 4 per cent. But to a
banker, handling many millions, his share of that extra one per cent of
other people’s incomes is worth thinking about.” As a result, Kelly al-
leged, “bankers became inclined more than ever to take chances” with
trust funds.15 These considerations counseled in favor of strict supervi-
sion of trustees by the courts, to rein them in when they tried to cross
the line from investment into speculation.

14 Jonathan Houston Merrill, “Investment of Trust Funds,” American Law


Register 34 (1886): 232–​33; Mayo Adams Shattuck, “The Massachusetts Prudent
Man in Trust Investments,” Boston University Law Review 25 (1945): 334, 337;
Austin Wakeman Scott, The Law of Trusts (Boston:  Little, Brown and Co.,
1939), 2:1206.
15 Zion’s Herald, 21 May 1884, 164; Fred C. Kelly, “Can Trust Companies Be
Trusted?,” Forum and Century 88 (1932): 241–​42.
290

290 Speculation

On the other hand, cases only reached the courts when investments
went sour. Beneficiaries never complained when trustees took risks and
made money. With hindsight it was easy to condemn a transaction as
unduly speculative, after it had already proven disastrous. Judges’ aware-
ness that they encountered a biased sample of investments should have
counseled in favor of deference to the decisions of trustees, for fear of
classifying as speculation a transaction that would have seemed a prudent
investment if it had been taken to court when it was made. If many judges
were aware of this problem, however, it is not evident in their opinions,
which contain few traces of humility. As one judge wrote in 1931, speak-
ing of the state of the stock market two years earlier, just before the crash,
“it was common knowledge, not only amongst bankers and trust com-
panies, but the general public as well, that the stock market condition at
the time of the testator’s death was an unhealthy one, that values were
very much inflated, and that a crash was almost sure to occur.” The judge
accordingly blamed the trustees for lacking the foresight to anticipate the
crash, a prescience that had been possessed by very few in 1929.16
The line between speculation and investment, always unstable,
moved conspicuously in the twentieth century. Virtually all courts
began to permit trustees to buy shares of business corporations, so long
as the trustees intended to own the shares for a long time. “The dif-
ference between speculation and investment,” declared an Alabama
court, was that “one who buys common stocks with the idea of sell-
ing them on the market for higher prices is speculating. One who is
making a prudent investment examines the stocks’ intrinsic values and
purchases them for a long-​term investment.” As Mayo Shattuck sum-
marized the cases at midcentury, “definitions of ‘speculative’ participa-
tions must vary with time and place and circumstances. The concept of
permanence of investment, as contrasted with transactions entered into
for a quick turnover or profit, is, however, basic.”17 In the nineteenth

16 Jeffrey J. Rachlinski, “Heuristics and Biases in the Courts: Ignorance or


Adaptation?,” Oregon Law Review 61 (2000): 79–​81; In re Chamberlain’s Estate,
156 A. 42, 43 (N.J. Prerogative Ct. 1931).
17 Harold B. Elsom, “The Law of Trust Investment,” Financial Analysts Journal 16,
no. 4 (1960): 32; First Alabama Bank of Montgomery, N.A. v. Martin, 425 So. 2d
415, 427 (Ala. 1983); Mayo Adams Shattuck, “The Development of the Prudent
  291

Speculation or Investment? 291

century, judges had tended to draw the line between corporate stock
and government bonds, but in the twentieth they tended to draw it
between short-​term and long-​term purchases of corporate stock.
One reason for the line’s migration was simply that by the middle
of the twentieth century there were some corporations that had proven
stable and profitable for a long period of time. The St. Louis–​based
International Shoe Company, for example, had been formed in 1911
by the merger of firms dating back to the 1890s. By the mid-​twentieth
century it was the largest shoe manufacturer in the nation. Its sales
and its revenue had grown continuously for decades, except for a brief
downturn at the beginning of the Depression. It had operated at a
profit every year since its inception, and its stock had paid dividends
every year. It had no debt. In 1946, when a St. Louis court had to
decide whether stock in the International Shoe Company was an ap-
propriate investment for a trustee, the court had no doubt that it was.
“Generally speaking,” explained the Philadelphia lawyer James Moore,
shares “in a ‘well-​seasoned’ corporation are acceptable for investment
and retention. By ‘well-​seasoned’ (or similar phrases) the courts seem
to mean corporations which have been in existence for a substantial
period of time, which are leaders in their field, which have a back-
ground of steady earnings and growth, and whose securities, listed on a
recognized exchange, are not particularly subject to rapid fluctuations
in market price.” In short, “the background of the corporation will go
far toward determining whether the investment is speculative.”18
Another reason for the courts’ growing acceptance of stock purchases
as investments rather than speculations was the pervasive inflation of
the twentieth century. There had been inflationary periods in the nine-
teenth century, especially during the Civil War, but these had been
more than balanced out by long periods of deflation. Price levels in
1900 were only around two-​thirds of what they had been in 1800. The

Man Rule for Fiduciary Investment in the United States in the Twentieth
Century,” Ohio State Law Journal 12 (1951): 517.
1 8 Louis S. Headley, “Trust Investments: Fundamental Principles Lawyers Should
Know,” Trusts and Estates 71 (1952): 740–​41; Warmack v. Crawford, 195 S.W.2d
919, 920–​21 (Mo. Ct. App. 1946); James A. Moore, “A Rationalization of the
Trust Surcharge Cases,” University of Pennsylvania Law Review 96 (1948): 668.
292

292 Speculation

twentieth century was very different. Prices nearly tripled between 1900
and 1950, even with the sharp deflation of the early 1930s, and in the
second half of the twentieth century they would rise even faster. This
new economic climate required a rethinking of trustees’ investment
strategies. In the nineteenth century, when even an investment that
paid no interest at all would leave the investor ahead after accounting
for deflation, the prudent man had no reason to take risks to seek a
higher return. But in the twentieth century, when a long-​term invest-
ment at low interest rates would lose money in light of inflation, the
avoidance of all risk was no longer prudent. “Hardship has overtaken
the income beneficiaries of many trusts which, when created, were
thought to provide adequate support,” one commentator observed after
the inflation of the late 1940s. “The trouble with trust funds,” agreed
the financial journalist H. J. Maidenberg, was that the old standards of
prudence had become obsolete. Because of inflation, “in recent years,
trust funds have generally performed far worse than the portfolios of
most investors who have handled their own trading and investments.”19
The prudent man had to invest in stocks to keep up with inflation.
Finally, in the second half of the twentieth century, advances in
economics made the old prudent man rule seem even more outdated.
Judges had been in the habit of evaluating investments one by one,
when by modern lights the relevant question was not the wisdom of
any single investment but rather the wisdom of the entire portfolio
of investments chosen by the trustee. A purchase of stock that might
have been too risky on its own could be deemed prudent as part of a
larger diversified package. Along with the new economic theory came
new investment vehicles like index funds, which made it much easier
for trustees to diversify their investments. These developments put ad-
ditional pressure on the always-​fragile line between speculation and
investment.20

19 Susan B. Carter et al., eds., Historical Statistics of the United States: Millennial
Edition On Line (Cambridge University Press, 2006), table Cc1–​2; Harry L.
Fledderman, “Prudent Man Investment of Trust Funds During Inflation,”
California Law Review 39 (1951): 380; New York Times, 18 Jan. 1981, F13.
20 John H.  Langbein and Richard A.  Posner, “Market Funds and Trust-​
Investment Law,” American Bar Foundation Research Journal 1 (1976):  1–​34;
Jeffrey N. Gordon, “The Puzzling Persistence of the Constrained Prudent Man
╇ 293

Speculation or Investment? 293

As a result, every state repealed the prudent man rule by the end
of the twentieth century. After nearly two centuries of trying to draw
a line between investment and speculation, the task was abandoned
as fruitless. “We no longer use speculation as a tool for dealing with
trustee investment misadventures because it is considered sloppy, sen-
timental, and inefficient,” one law professor explained. “The reforms
imported modern investing theory and practice into trust law.” Rather
than asking whether the trustee was investing or speculating, the new
laws tended instead to ask whether the trustee was advancing the pur-
pose of the trust with skill and diligence. Under this new rule, trusts
included more stock than they had before.21 It was simply too hard to
tell where investment stopped and speculation began.

Taxing Speculators
When the federal government began taxing income in 1913, capital
gains were not given any special treatment. They were taxed at the
same rates as other income. This was a recurring source of complaint
from people who owned land and other property that had appreciated
for many years and who thus faced very large tax bills when they sold
the property. “Under the present law,” explained the Senate Finance
Committee in 1921, “many sales of farms, mineral properties, and other
capital assets have been prevented by the fact that gains and profits
earned over a series of years are under the present law taxed as a lump
sum.” To provide relief to such taxpayers, Congress decided to count as

Rule,” New York University Law Review 62 (1987): 52–╉114; Edward C. Halbach


Jr., “Trust Investment Law in the Third Restatement,” Real Property, Probate
and Trust Journal 27 (1992):  407–╉65; Michael T.  Johnson, “Speculating on
the Efficacy of ‘Speculation’: An Analysis of the Prudent Person’s Slipperiest
Term of Art in Light of Modern Portfolio Theory,” Stanford Law Review 48
(1996): 419–╉47.
2 1 Joel C. Dobris, “Speculations on the Idea of ‘Speculation’ in Trust Investing: An
Essay,” Real Property, Probate and Trust Journal 39 (2004): 453; William P. Wade,
“The New California Prudent Investor Rule: A Statutory Interpretive Analysis,”
Real Property, Probate and Trust Journal 20 (1985): 1023; Max M. Schanzenbach
and Robert H. Sitkoff, “Did Reform of Prudent Trust Investment Laws Change
Trust Portfolio Allocation?,” Journal of Law and Economics 50 (2007): 681–╉711.
294

294 Speculation

income only 40 percent of the capital gains realized upon the sale of an
asset. The result was to make the tax on capital gains only 40 percent of
the tax on ordinary income.22
But this change raised a new problem. Many of the taxpayers who
would benefit from the reduced rate were not selling the family farm;
they were selling shares of corporations, sometimes shares they had
purchased but a short time before. Taxing capital gains at a lower rate
would be a windfall to speculators. “Is this desirable?” wondered the
economist Willford King. “Speculation tends to degenerate into gam-
bling and gambling is generally recognized as an anti-​social form of
activity.” Senator David Walsh of Massachusetts raised the same con-
cern on the floor of the Senate. “Under the proposed amendment and
bill a lawyer or any other professional man who derived as a fee from
a large case or a merchant who through a substantial increase in sales
derived an income of, say, $100,000 per year is taxable upon the full
amount of income,” Walsh complained. But “the speculator who de-
rives an income of $100,000 a year upon the New York Stock Exchange
or in any other manner would be taxable only on 40 per cent of his net
income, or $40,000.”23 Why should speculators pay lower taxes than
everyone else?
Walsh proposed a solution. “If the benefits of this amendment are
limited to cases of the sale of capital assets which have been owned by
the taxpayer for a period of three years or more,” he suggested, relief
would be afforded to those who deserved it, “but the speculator will not
be included and will be forced to pay a tax upon his entire net income,
the same as every other taxpayer.” The idea of making speculators pay
the full tax received instant approval from Walsh’s colleagues, who quib-
bled only about how best to achieve that goal. Porter McCumber of
North Dakota proposed that the law should require full taxation only
of “speculative stocks and securities” and give preferential treatment to
all other capital gains. But this idea was shot down by Irvine Lenroot
of Wisconsin, who pointed out that it would be very difficult for the

22 S. Rep. No. 275, 67th Cong., 1st Sess. (1921), 12.


23 Willford I.  King, “Earned and Unearned Income,” Annals of the American
Academy of Political and Social Science 95 (1921): 258; Congressional Record 61
(1921): 6575.
  295

Speculation or Investment? 295

Treasury Department, in collecting the tax, to determine which of a


taxpayer’s transactions were “speculative” and which were not. The only
way to distinguish speculation from investment, Lenroot counseled, was
to look to the length of time between purchase and sale, as Walsh had
suggested. In the end, Congress agreed on requiring a two-​year holding
period before a taxpayer could get the advantage of the lower rate.24
For the rest of the twentieth century and into the twenty-​first, the
principle that speculators should be taxed at a higher rate than inves-
tors remained a central part of the tax law. Congress tinkered with the
details every so often. The two-​year holding period was replaced in
1934 with a sliding scale of holding periods, ranging from one year to
ten years, with a lower tax rate the longer the time between purchase
and sale. In 1938, the sliding scale was replaced by an eighteen-​month
holding period, which was shortened to six months in 1942 and then
lengthened to twelve months in 1976. Tax rates changed even more fre-
quently, and as they did, so did the size of the preference given to inves-
tors over speculators. There were recurring changes in the tax treatment
of capital losses, in the provisions relating to the sale of homes, and in
many technical details that were attended to primarily by tax lawyers
and accountants. But the one consistent goal of the capital gains tax,
throughout all this change, was that the lower tax rate for capital gains
should be available to investors, not to speculators. “Capital gains are
as desirable a basis for taxation as any other kind of income,” explained
the economist R. S. Tucker, “and to the extent that they represent spec-
ulative or purely chance income, even more desirable a basis.”25

24 Congressional Record 61 (1921): 6575.


25 James R. Repetti, “The Use of Tax Law to Stabilize the Stock Market: The
Efficacy of Holding Period Requirements,” Virginia Tax Review 8 (1989): 591–​
637; C. Thomas Paschall, “U.S. Capital Gains Taxes: Arbitrary Holding Periods,
Debatable Tax Rates,” Southern California Law Review 73 (2000): 843–​78;
Congressional Research Service, Individual Capital Gains Income: Legislative
History (2006); Van Mayhall, “Capital Gains Taxation—​The First One Hundred
Years,” Louisiana Law Review 41 (1980): 81–​99; Anita Wells, “Legislative History
of the Treatment of Capital Gains Under the Federal Income Tax, 1913–​1948,”
National Tax Journal 2 (1949): 12–​32; R. S. Tucker, “Government Control of
Investments and Speculation,” American Economic Review 25 (1935): 148.
296

296 Speculation

But taxing speculators more than investors was easier said than done,
because of the difficulty of drawing a clear line between speculation
and investment. On one side, there were repeated calls to lengthen the
required holding period on the ground that the existing period was so
short that it permitted speculators to get the advantage of the lower
rates, and there were frequent arguments for raising the rates on short-​
term capital gains on the ground that higher rates were needed to deter
speculation. On the other side, there were repeated calls to shorten
the period and lower the rates, or even to exempt capital gains from
taxation completely, so as not to discourage investment.26 The age-​old
debate over the distinction between investment and speculation had
become a debate over the capital gains tax.
In early 1945, for example, Federal Reserve Chair Marriner Eccles
proposed a sharp increase in the short-​ term capital gains rate, to
90 percent, and a lengthening of the holding period from six months
to two years, to tame wartime inflation. His proposal was aimed at
“speculation—​not investment,” Eccles explained. It would “discour-
age all such speculative transactions, whether in homes, farms, stocks,
or commodities, and whether based upon credit or cash—​and would
do so without interference with normal, non-​speculative transactions.”
Eccles’s proposal aroused a storm of criticism. Taxing speculation would
not just be an ineffective way of reducing inflation, charged the econo-
mist Aaron Sakolski. It would also “stop progress,” because “without
speculation, would there be progress of any kind?” Barron’s, a business
newspaper, compared Eccles’s plan to a tax that had been imposed in
Nazi Germany.27 Eccles’s proposal was shelved.
Alongside these recurring political battles ran a long-​standing debate
over a more fundamental question: was the distinction between short-​
term and long-​term holdings a sensible way to divide speculators from
investors? “There is nothing in accounting, in economics, or in fiscal
policy which justifies the distinction of speculation and investment on

26 Atlanta Constitution, 1 May 1929, 8; New York Times, 20 Jan. 1938, 1; 4 Jan. 1931,
N20; Wall Street Journal, 24 June 1931, 8; Chicago Daily Tribune, 18 Oct. 1935,
39; Daily Boston Globe, 27 Nov. 1936, 14; Wall Street Journal, 21 Mar. 1942, 3.
27 Commercial and Financial Chronicle, 1 Mar. 1945, 963; 8 Mar. 1945, 1061; 1
Mar. 1945, 929, 943; Barron’s, 26 Mar. 1945, 5.
  297

Speculation or Investment? 297

the basis of time,” testified Elisha Friedman, an economist at New York


University, before the House Ways and Means Committee. “There is
no justification for segregating short-​term from long-​term gains.” On
this view, the tax law had adopted an utterly arbitrary distinction, be-
cause the line between short-​term and long-​term holdings had no cor-
respondence with the line between speculators and investors. Stanley
Surrey, who shuttled between Harvard Law School and the Treasury
Department throughout the mid-​twentieth century, agreed that the
distinction between short-​term and long-​term holdings “leaves some
strange results.” In the long-​term category were a heterogeneous collec-
tion of market participants—​“the professional speculator whose pur-
chases and sales are substantial and frequent but involve more than a six
months’ holding period, the large investor who is constantly perfecting
his portfolio through changes in its composition, the modest inves-
tor who occasionally changes his portfolio, and the amateur speculator
who takes a chance now and then.” As Surrey saw it, “we are thus left
with a congressional feeling that speculation and investment are dif-
ferent matters, but with no statutory differentiation between the two
except as respects the in-​and-​out, daily traders.”28
Others argued that even if speculators intended to hold assets for
shorter periods than investors did, the unpredictability of market con-
ditions made the actual holding period a poor way of distinguishing
between the two groups. When the market was down, as in the early
1930s, stock that had originally been purchased for speculative purposes
might be held for years, in the hope that its price might climb back
upward. The Chicago tax lawyer Herman Reiling thus complained in
1932 that confining “speculation” to short-​term holdings “appears to
ignore the facts.” And when the market was up, even a person who had
purchased with the intent to hold for a long time might be tempted
to sell. The Investment Bankers Association accordingly proposed de-
fining as a speculator only “the day-​to-​day trader,” who “makes his

28 Revenue Revision of 1942: Hearings Before the Committee on Ways and Means,
House of Representatives, 75th Cong., 2nd Sess. (Washington, D.C.: Government
Printing Office, 1942), 1:944; Stanley S. Surrey, “Definitional Problems in
Capital Gains Taxation,” Harvard Law Review 69 (1956): 999–​1000.
298

298 Speculation

living by stock transactions.” All others, the IBA argued, were investors
who deserved the lower rate.29
The standard response to such arguments was that the holding period,
while no doubt an imperfect way of distinguishing speculation from
investment, was the only feasible way. The economist Robert Murray
Haig noted that Weimar Germany had tried to distinguish specula-
tive from investment income on the basis of the taxpayer’s intent; if
he intended to speculate, he was a speculator. But this effort proved
impractical. “It led to vagueness, to debates between administrators and
taxpayers, and to litigation,” Haig recalled. After a few years, Germany
gave up. “In place of the vague test of ‘intent’ in the mind of the tax-
payer, appeared the definite test of the length of the time-​period cov-
ered by the transaction,” Haig explained. In Germany, as in the United
States, “a speculation is a short-​term transaction and an investment a
long-​term venture.” The short-​term/​long-​term distinction sometimes
produced anomalous results, acknowledged the New York tax lawyer
Peter Miller, but “the necessities of administration explain this anom-
aly.” Speculators differ from investors “only in state of mind and not in
easily recognizable overt behavior. If ‘the devil himself knoweth not the
thought of man,’ the Commissioner of Internal Revenue could hardly
be expected to.”30
There was little doubt, in any event, that the distinction between
short-​term and long-​term holdings mapped onto a widely held, if not
fully theorized, distinction in the public mind between speculators and
investors. When congressional committees tinkered with the length
of the holding period, they repeatedly explained that the holding pe-
riod’s purpose was “separating, in a practical way, speculative transac-
tions from investment transactions” (as a 1938 House report put it),
or to serve as “a sufficient deterrent to the speculator as contrasted
with the legitimate investor” (in the words of a 1942 Senate report).
“The underlying concept,” declared a 1976 report of Congress’s Joint

29 Herman T. Reiling, “Stock Transactions and the Income Tax,” Tax Magazine
10 (1932): 207; Los Angeles Times, 28 Nov. 1941, 29.
30 Wall Street Journal, 8 Apr. 1937, 4; Peter Miller, “The ‘Capital Asset’
Concept:  A  Critique of Capital Gains Taxation:  I,” Yale Law Journal 59
(1950): 843.
  299

Speculation or Investment? 299

Committee on Taxation, “is that a person who holds an asset for only a
short time is primarily interested in obtaining quick gains from short-​
term market fluctuations, which is a distinctive speculative activity. In
contrast, the person who holds an asset for a long time probably is
interested fundamentally in the income from his investment and in the
long-​term appreciation value.” All transactions involved an element of
investment and an element of speculation, conceded William Healy, a
federal judge in Idaho. But Healy, like many others, nevertheless per-
ceived a difference between the two. “The ‘in-​and-out’ market hanger-​
on who buys and sells through brokers on margin is a typical example
of the pure speculator,” he reasoned. “On the other hand, an investor is
ordinarily thought to be a person who acquires property for the income
it will yield rather than for the profit he hopes to obtain on a resale.”31
Of course, using the tax code to distinguish between speculation
and investment made sense only if there was a distinction between
speculation and investment, and opinion on that question was no
more settled than it had ever been. Beardsley Ruml, chairman of the
Macy’s department store company, thought “the discussion is some-
what confused by the use of the term speculation. It seems to me, as
an example, that at Macy’s when we buy some women’s dresses to sell
in six months or six weeks, that is speculation, if we want to use that
horrid term. What you have is buying and selling of things,” which was
the same no matter what name was attached to it. The Chicago Tribune
agreed that “the dividing line between investment and speculation is
often indistinct,” because even an ostensible investor hoped one day
to sell his investment for more than he had paid. “All investors are in-
terested in the probable price movements of their assets,” declared the
economist Lawrence Seltzer. “Like professional speculators, they try to
take future price movements into account in the prices at which they
buy and sell.”32 If investment and speculation were identical, there was

31 H.R. Rep. No. 1860, 75th Cong., 3rd Sess. (1938), 7; S. Rep. No. 1631, 77th
Cong., 2nd Sess. (1942), 50; Joint Committee on Taxation, General Explanation
of the Tax Reform Act of 1976 (Washington, D.C.: Government Printing Office,
1976), 426; United States v. Chinook Inv. Co., 136 F.2d 984, 985 (9th Cir. 1943).
32 Capital Gains Taxation (New York: Tax Institute, 1946), 87–​88; Chicago Daily
Tribune, 17 Nov. 1946, A11; Lawrence H. Seltzer, The Nature and Tax Treatment
300

300 Speculation

no reason to use the capital gains tax to encourage one and discourage
the other.
And even if speculation could be distinguished from investment,
there was always a considerable body of opinion that the two should be
treated identically, because speculation was just as good. In the 1950s,
Life magazine was willing to concede that speculation differed from in-
vestment, because “speculation is putting the desire for enhancement
ahead of the desire for generally predictable returns—​for growth, rather
than dividends. It is the difference between buying Ford stock in 1903
and buying Standard Oil, whose success was already established; or be-
tween buying a uranium stock today and buying G.M.” But Life insisted
that speculation and investment should both be encouraged. “People
who are willing and able to take such risks enable new industries to get
born, and in that sense speculation is a social good,” the magazine rea-
soned. “In a very real sense, America grew great on speculation.”33
The use of the capital gains tax to encourage investment but dis-
courage speculation was also attacked on technical grounds, from both
directions. In the early years of the capital gains tax, critics charged
that differential tax rates had little influence on the conduct of either
investors or speculators, who had more important things to worry
about. “The operations of such individuals are … relatively insensi-
tive to capital gains tax provisions,” Treasury staff concluded in 1937.
“It must be emphasized that tax factors operate in practice among a
welter of other considerations, and that, by reason of this fact, they are
commonly robbed of a great deal of their force.” When the Twentieth
Century Fund studied the tax system at approximately the same
time, it likewise concluded that tax rates were unlikely to affect de-
cisions to buy or sell securities, decisions taxpayers would base solely
on their expectations as to future prices. If capital gains taxes did not
affect the conduct of speculators or investors, it would be pointless to
fiddle with rates as a means of deterring the former or encouraging
the latter. The Treasury could focus instead on raising revenue, a goal

of Capital Gains and Losses (New York: National Bureau of Economic Research,


1951), 67.
3 3 “The Market and Its ‘Friends,’ ” Life, 21 Mar. 1955, 46.
  301

Speculation or Investment? 301

that stood in tension with that of discouraging speculation, because


the more that speculation was discouraged, the less revenue a tax on
speculation would raise.34
In later years, however, critics of the distinction between long-​term
and short-​term profits charged precisely the opposite—​that the distinc-
tion had too great an effect on the conduct of speculators and investors.
In 1948, a security analyst in New York named Goldie Stone interviewed
a small sample of brokers, accountants, and bankers to find out how
the capital gains tax affected the buying and selling of securities. One
obvious strategy, she noted, was not to sell securities on which there
were gains until the expiration of the six-​month holding period, but to
sell all securities on which there were losses before the period expired,
so that the short-​term losses could offset any short-​term gains. Aside
from that simple practice, Stone found that sophisticated taxpayers en-
gaged in several other “highly complicated and devious methods” of
manipulating the short-​term/​long-​term distinction to their advantage.
One technique was to use short sales to convert a short-​term gain into
a long-​term gain without fear of loss. The owner of the stock would sell
the stock short, wait for the six-​month period to pass, and then cover
the short sale with his own stock. He would get the same gain as if he
had sold the stock earlier, but now it was long-​term gain rather than
short-​term gain. The same goal could also be accomplished by buying
put options, which would allow the owner of the stock to lock in a gain
but defer it until after six months had passed. A second technique, the
mirror image of the first, was to take advantage of a short-​term loss
without having to lose one’s position in a stock. The owner of the stock
would sell it short, purchase additional stock, and then a month later
cover the short sale with his original stock. He would have a short-​term
loss, but he would still own as much of the stock as before. The same
goal could also be accomplished with call options, which would allow

34 “Tax Treatment of Capital Gains and Losses” (staff memo, Division of Tax
Research, Treasury Department, 1937), section V.A.2-​3, http://​www.taxhis-
tory.org/​Civilization/​Documents/​Surveys/​hst23731/​23731-​1.htm; Twentieth
Century Fund, Facing the Tax Problem:  A  Survey of Taxation in the United
States and a Program for the Future (New  York:  Twentieth Century Fund,
1937), 190; New York Times, 16 Feb. 1938, 20.
302

302 Speculation

him to sell the stock and buy it back at the same price thirty days later.
And this was just the beginning of a long list of clever transactions
savvy taxpayers engaged in, transactions with no purpose other than
taking advantage of the distinction between short-​term and long-​term
capital gains. “The dominant result of these interviews,” Stone con-
cluded, “appears to be that tax-​saving manipulations are restricted to
a fairly limited group of large stockholders with ample cash reserves
who trade in substantial blocks of securities. This type of individual
has, in addition, sufficient time to discover the many available devices
and sufficient money to employ expert advice to inform him of these
possibilities.”35
One of those expert advisors was M. Francis Bravman, a tax lawyer
in New York and the chair of the American Bar Association’s committee
on income taxation. Bravman admitted that the structure of the capital
gains tax “places a very high premium on the use of devices to trans-
mute ordinary income into capital gains, or capital losses into ordinary
deductions, or short-​term gains into long-​term gains.” A great deal of
time, money, and intelligence was devoted to gaming the system, be-
cause “this approach to the taxation of capital gains has resulted in
a complicated federal income tax law, containing very tempting in-
centives for taxpayers to reduce their taxes.”36 If speculators could use
tricks to make themselves look like investors at tax time, there was no
point in trying to treat the two groups differently.
But all this criticism, from all these directions, had no discernible
effect on the apparently widespread belief that speculators should be
taxed more heavily than investors. Unlike the law of trusts, which aban-
doned as futile the effort to distinguish investment from speculation,
the tax law continues to draw a line between the capital gains of inves-
tors and those of speculators.

35 Goldie Stone, “How the Capital Gains Tax Affects Buying and Selling
Securities,” Taxes 26 (1948): 1041–​48.
36 M. Francis Bravman, “Integration of Taxes on Capital Gains and Income,”
Virginia Law Review 37 (1951): 527–​28.
╇ 303

Speculation or Investment? 303

The Activities of€a€Casino


Wartime brings new taxes, as the government needs to finance new
expenditures. The Spanish-╉American War was no exception. In 1898,
Congress imposed a wide range of new taxes, including an estate tax
and taxes on tobacco, flour, cosmetics, and many other items. One of
the new taxes was a tax on stock transactions. It was very small, only
1 cent per $100 of stock. The purpose of all these taxes was “to raise
moneys for the support of the government,” as the Supreme Court
explained.37 The stock transfer tax was no more intended to deter stock
trading than the cosmetics tax was intended to deter the use of cosmet-
ics or the estate tax was intended to deter dying. There was a war on,
and the government needed the money.
The federal stock transfer tax was repealed when the war was over.
But the idea lingered on. In 1905, when the state of New York faced a
looming budget deficit, New York imposed a tax of 2 cents on each $100
of stock sold. The purpose was again to raise money, but as one eco-
nomics journal observed, “the tax probably falls in the main upon the
purchasers [of stock], but this means very often the speculative buyers,”
who engaged in the most frequent transactions. “The burden upon the
investing public is slight,” the journal concluded, drawing a distinction
between speculators who traded frequently and investors who traded
only rarely. The New York Times agreed that the tax would be felt pri-
marily by “large speculators whose daily purchases much exceed 1,000
shares.” The federal government resumed taxing stock transfers in 1914,
with the start of World War I. Both governments would continue to
impose small taxes on stock transfers for a long time. The federal tax
remained in effect until 1966, and New York’s until 1981.38
The stock market crash of 1929 prompted several calls for a much
stiffer transaction tax, intended not to raise revenue but to suppress

37 Steven A. Bank, Kirk J. Stark, and Joseph J. Thorndike, War and Taxes
(Washington, D.C.: Urban Institute Press, 2008); 30 Stat. 448–╉70 (1898);
Nicol v. Ames, 173 U.S. 509, 516 (1899).
38 “Changes in the Tax Laws of New York State in 1905,” Quarterly Journal of
Economics 20 (1905): 153; New York Times, 22 Apr. 1905, 10; Joseph J. Thorndike,
“Speculation and Taxation: Time for a Transaction Tax?” (2008), Tax History
Project Article Archive, http://╉www.taxhistory.org.
304

304 Speculation

speculation. “This terrible calamity occurred because we had no legis-


lation to prevent speculation,” averred Representative Jed Johnson of
Oklahoma, who proposed taxing stock and commodity futures trades
at rates high enough that “the exchanges would be virtually put out of
business.” The idea received considerable circulation among economists
when John Maynard Keynes suggested a milder version of it in his
widely read General Theory of Employment, Interest, and Money. Keynes
drew a distinction between speculation, which he defined as “forecast-
ing the psychology of the market,” and enterprise, defined as “forecast-
ing the prospective yield of assets over their whole life.” This distinction
was very much like the short-​term/​long-​term line drawn at the time by
American lawyers for purposes of trust law and the capital gains tax. As
Keynes saw it, the predominance of speculation over enterprise in the
1920s had been one of the causes of the rise and then fall of the market.
“Speculators may do no harm as bubbles on a steady stream of enter-
prise,” Keynes explained. “But the position is serious when enterprise
becomes the bubble on a whirlpool of speculation. When the capital
development of a country becomes a by-​product of the activities of a
casino, the job is likely to be ill-​done.” One possible solution, Keynes
suggested, was to raise the tax rate on stock transactions to reduce their
frequency. “The introduction of a substantial Government transfer tax
on all transactions might prove the most serviceable reform available,”
he concluded, “with a view to mitigating the predominance of specula-
tion over enterprise in the United States.”39
The idea of a transfer tax to dampen speculation re-​emerged among
economists in the 1970s, when the liberalization of foreign exchange
markets allowed traders to move quickly from one currency to another,
to a degree that seemed to threaten the ability of national governments
to set their own monetary policy. “How can some national monetary
autonomy be preserved?” worried the economist James Tobin. “Some
sand has to be thrown into the well-​greased channels of the Eurodollar
market.” Tobin’s solution was “an internationally agreed uniform tax,

39 Congressional Record 78 (1934): 8097–​98; John Maynard Keynes, The General


Theory of Employment, Interest, and Money (New York: Harcourt, Brace and
Co., 1936), 158–​60.
  305

Speculation or Investment? 305

say 1%, on all spot conversions of one currency into another.”40 The
idea was soon widely known as a “Tobin tax”—​a tax intended to reduce
the number of transactions in a market thought to be too active, by
making the transactions substantially more expensive to carry out.
In the late 1980s, after a brief period of extraordinary volatility in
stock markets all over the world, including the largest one-​day decline
in the history of the US market, several leading economists proposed a
similar tax on stock speculation. Joseph Stiglitz urged a tax “to discour-
age short-​term speculative trading,” which he contended was the cause
of a wasteful arms race in which competing speculators made “excessive
expenditures on gathering information and on financial innovation.”
In Stiglitz’s view, “resources devoted to gambling—​and to short-​term
speculation in the stock market—​could be devoted to more produc-
tive uses.” Lawrence Summers agreed that “the efficiency benefits of
curbing speculation are likely to exceed any costs of reduced liquidity
or increased costs of capital that come from taxing transactions more
heavily.” The Treasury Department contemplated proposing such a tax.
Members of Congress introduced narrower versions of a transaction
tax. A bill called the Excessive Churning and Speculation Act of 1989,
introduced by the two senators from Kansas, would have imposed a
10  percent tax on the gains of pension funds from the sale of assets
held for thirty days or less. Another bill introduced a few months later
would have barred pension funds from selling assets held for less than
three months. None of these proposed taxes was ever imposed, but the
idea of a tax to discourage stock speculation lingered on, to be debated
every few years by lawyers and economists.41

40 James Tobin, The New Economics One Decade Older (Princeton, N.J.: Princeton
University Press, 1974), 88–​ 89; see also James Tobin, “A Proposal for
International Monetary Reform,” Eastern Economic Journal 4 (1978): 153–​59.
41 Joseph E. Stiglitz, “Using Tax Policy to Curb Speculative Short-​Term Trading,”
Journal of Financial Services Research 3 (1989): 102–​ 3, 109; Lawrence H.
Summers and Victoria P. Summers, “When Financial Markets Work Too Well:
A Cautious Case for a Securities Transaction Tax,” Journal of Financial Services
Research 3 (1989): 263; Donald W. Kiefer, “The Security Transactions Tax: An
Overview of the Issues,” Tax Notes 48 (1990): 885; Joint Committee on Taxation,
Tax Treatment of Short-​Term Trading (Washington, D.C.: Government Printing
Office, 1990), 8; Scott W. MacCormack, “A Critique of the Reemerging Securities
Transfer Excise Tax,” Tax Lawyer 44 (1991): 927–​41; G. William Schwert and
306

306 Speculation

In the early 2010s, amid growing concern about the detrimental


effects of high-​frequency trading, attention turned once again to the
possibility of a transfer tax. High-​frequency traders pushed short-​term
speculation to its physical limits, buying and selling the same asset
within microseconds, to earn minuscule profits per trade on a very high
volume of trades. Critics suggested that a tiny transfer tax per trade, as
small as three hundredths of 1 percent, would be enough to deter high-​
frequency trading without causing much reduction in ordinary trading
at conventional time scales.42
The argument against a transfer tax has always been that a tax is a
blunt instrument that will discourage the good transactions along with
the bad. Opponents of a tax on high-​frequency trading argued that by
making trading more expensive, the tax would reduce the earnings of
ordinary investors in mutual funds, because even if the investors them-
selves do not trade regularly, the funds do. Opponents also contended
that the tax would fall heavily on market makers, who would com-
pensate by widening their bid-​ask spreads, a result that would likewise
harm ordinary investors.43
High-​frequency trading was a new phenomenon, but it raised a
perennial dilemma:  how to deter speculation without also deterring
investment. And that question rested in turn on a more fundamental
question: was it even possible to tell the two apart?

Paul J. Seguin, “Securities Transaction Taxes: An Overview of Costs, Benefits


and Unresolved Questions,” Financial Analysts Journal 49 (1993): 27–​35; Lynn
A. Stout, “Are Stock Markets Costly Casinos? Disagreement, Market Failure,
and Securities Regulation,” Virginia Law Review 81 (1995): 699–​702; Paul G.
Mahoney, “Is There a Cure for ‘Excessive’ Trading?,” Virginia Law Review 81
(1995): 727–​36; Anna Pomeranets and Daniel Weaver, “Security Transaction
Taxes and Market Quality” (2013), http://​ssrn.com/​abstract=1980185.
42 Lee Sheppard, “A Tax to Kill High Frequency Trading,” Forbes, 16 Oct. 2012;
New York Times, 8 Apr. 2014, B1.
43 Charles M.  Jones, “What Do We Know About High-​Frequency Trading?”
(2013), http://​ssrn.com/​abstract=2236201.
9
Q
Deregulation and Crisis?

“Time to man up and regulate these guys correctly,” complained John


Scollin Jr., of Winter Springs, Florida. “Somalian pirates have a better
sense of ethics and meth-​addled winos are more trustworthy than
these guys.” Scollin was referring to the speculators he held respon-
sible for the financial crisis of 2007–​08. He was hardly alone in his
certainty that the crisis had been caused by deregulation, which had
allowed banks and other financial intermediaries to engage in dan-
gerously speculative transactions. “I couldn’t be more angry over how
my government allowed regulations to become so lax,” fumed Charles
Yeargen of Beckley, West Virginia. Nathaniel Powell of Los Angeles
agreed. “It’s unfortunate that our congress, especially [House Speaker
Nancy] Pelosi, is spineless. Stop the gambling!” Powell urged. Barbara
Coulson of Marshall, North Carolina, observed that the government
had lost sight of an old truth—​“the awful consequences of risky trad-
ing.” The only conceivable explanation for why the government had
allowed so much speculation to take place, suggested Louis Spain Jr., of
Lakewood, Washington, was that the federal employees nominally re-
sponsible for overseeing the markets were actually “domestic terrorists
working secretly for the Taliban.” The speculators themselves were far
worse, worried Jerry Lujan of Albuquerque: “The Wall Street Leaches
and Banking Parasites are today’s equivalent of the money changers
Jesus was so upset about,” Lujan warned. “They are the anti-​christ!!”
Thomas Tague of Burien, Washington, had a simple solution to all
this speculation. “Stop it now,” he lectured federal officials. “Do your

307
308

308 Speculation

regulatory jobs. Pass and enforce laws that put the American people
first, not banks and corporations.”1
These critics of the government were not experts. They were ordinary
citizens with strong opinions on financial regulation. As their com-
ments suggest, the financial crisis focused more public attention on
speculation than at any time since the Great Depression. Critiques of
speculation carried an even sharper edge this time, however, because
of the widely shared view that overspeculation was a direct result of
the deregulation of the financial markets. When Time magazine listed
the “25 people to blame for the financial crisis,” the list included Bill
Clinton, whom Time held responsible for “financial deregulation, which
in many ways set the stage for the excesses of recent years”; George
W. Bush, who “embraced a governing philosophy of deregulation”; and
Senator Phil Gramm, “Washington’s most prominent and outspoken
champion of financial deregulation.” It was a charge leveled over and
over again: overspeculation had once been contained by laws prohibit-
ing the most risky transactions, but those safeguards had been rashly re-
moved, and now we were all suffering the consequences. “Welcome to
the Third World!” smirked the law professor Rosa Brooks. “Policies of
irresponsible government deregulation” had produced “an energy crisis,
a housing crisis, a credit crisis and a financial market crisis, all at once,
and accompanied (and partly caused) by impressive levels of corruption
and speculation.”2
Critics saw deregulation as the triumph of ideology over common
sense. “The meltdown of the financial industry across the spectrum is
due to the failed economics of Milton Friedman–style, unstructured
capitalism,” insisted one newspaper columnist who had seen his re-
tirement account lose a third of its value. “Deregulation gives those

1 Comments 1373, 1338, 1379, 1358, 1378, 1380, and 1372 on Financial Stability
Oversight Council Notice 2010-​0002, Public Input for the Study Regarding the
Implementation of the Prohibitions on Proprietary Trading and Certain Relationships
with Hedge Funds and Private Equity Funds (Nov. 2010), available at http://​www
.regulations.gov.
2 “25 People to Blame for the Financial Crisis,” Time, 11 Feb. 2009; Rosa Brooks,
“An Intervention Strategy for the U.S. Economic Mess,” L.A. Times, 18
Sept. 2008.
  309

Deregulation and Crisis? 309

opportunists a fast track to quick riches,” he reasoned, and “the result


is the worst financial crisis in history.” A member of the Missouri state
legislature declared that because the New Deal regulatory structure had
been dismantled “at the behest of Wall Street and some academics …,
the door was open for commercial banks, securities houses and insurers
to engage in risky speculative investments.” Even the United Nations
opined that “blind faith in the efficiency of deregulated financial mar-
kets” had “licensed profligacy through speculative finance.” The gov-
ernment-​appointed Financial Crisis Inquiry Commission, tasked to
explain what had happened, concluded that “the sentries were not at
their posts, in no small part due to the widely accepted faith in the
self-​correcting nature of the markets.” The commission lamented that
“more than 30 years of deregulation … had stripped away key safe-
guards, which could have helped avoid catastrophe.”3
Some critics focused their venom on particular measures. The law
professor Lynn Stout blamed the Commodity Futures Modernization
Act of 2000, which in her view caused a “sudden and wholesale removal
of centuries-​old legal constraints on speculative trading.” The financial
columnist James Rickards thought “the financial crisis might not have
happened at all but for the 1999 repeal of the Glass-​Steagall law that
separated commercial and investment banking.” The economist Joseph
Stiglitz also pointed the finger at the repeal of the Glass-​Steagall Act,
which in his view gave rise to “a demand for the kind of high returns
that could be obtained only through high leverage and big risk-​taking.”
As the subtitle of one popular account of the crisis put it, “Washington’s
wise men” had “turned America’s future over to Wall Street.”4

3 James Murr, “Financial Markets Must Be Regulated,” Santa Maria Times, 9 Jan.
2009; Jeanne Kirkton, “Financial Security Is National Security,” St. Louis Post-​
Dispatch, 8 July 2011; United Nations Conference on Trade and Development,
The Global Economic Crisis: Systemic Failures and Multilateral Remedies (New
York and Geneva: United Nations, 2009), iii; The Financial Crisis Inquiry
Report (Washington, D.C.: Government Printing Office, 2011), xviii.
4 Lynn A. Stout, “Derivatives and the Legal Origin of the 2008 Credit Crisis,”
Harvard Business Law Review 1 (2011): 4; James Rickards, “Repeal of Glass-​
Steagall Caused the Financial Crisis,” U.S. News & World Report, 27 Aug.
2012; Joseph E. Stiglitz, Freefall: America, Free Markets, and the Sinking of the
World Economy (New York: W. W. Norton, 2010), 163; Michael Hirsh, Capital
310

310 Speculation

Was it all true? What exactly had been deregulated, and why? And
what were the links between deregulation, speculation, and the finan-
cial crisis? Did deregulation give rise to unsafe levels of speculation? If
so, was speculation a cause of the crisis? Had we really forgotten the
lessons painfully learned in earlier depressions?

Badly€Burned
The idea of deregulating financial markets did not become a serious
possibility until the 1990s, by which time several other markets had
already undergone substantial and high-╉profile programs of deregula-
tion. Most of these programs yielded a similar pattern—╉greater com-
petition in the relevant industry, yielding enormous savings for con-
sumers and a correspondingly precarious position for producers and
their employees.
Perhaps the most conspicuous case was the airline industry. Before
1978, airfares and route changes had to be approved by a federal agency
called the Civil Aeronautics Board. The airlines were largely insulated
from price competition, so airfares were high and the airlines competed
instead to offer in-╉flight perks like food and drinks. Stephen Breyer,
who was counsel to the Senate Judiciary Committee in the 1970s, re-
called that when the committee held hearings about airline deregula-
tion, one of Senator Ted Kennedy’s constituents asked, “â•›‘Senator, why
are you holding hearings about airlines? I’ve never been able to fly.’
Kennedy replied:  ‘That’s why I’m holding the hearings.’â•›” The move
toward deregulation began during the Ford administration and gained
momentum in 1977, when President Carter appointed the economist
Alfred Kahn as chair of the Civil Aeronautics Board. In the Airline
Deregulation Act of 1978, Congress ended the regulation of fares and
route changes. The airline industry was opened to competition.5

Offense: How Washington’s Wise Men Turned America’s Future Over to Wall Street
(Hoboken, N.J.: John Wiley & Sons, 2010).
5 Stephen G. Breyer, “Airline Deregulation, Revisited,” Bloomberg Businessweek,
20 Jan. 2011; Anthony E.  Brown, The Politics of Airline Deregulation
(Knoxville: University of Tennessee Press, 1987).
  311

Deregulation and Crisis? 311

The results were not surprising. Consumers were the net beneficia-
ries from competition, producers the net losers. Airfares declined dra-
matically, adjusted for inflation, which brought air travel within the
reach of millions of moderate-​income passengers. This influx of new
price-​conscious customers would itself cause some new problems, in-
cluding crowded airports and crowded planes, but there is little doubt
that on balance, deregulation increased consumer welfare. Not so for
the welfare of producers. Some airlines, unable to compete in the new
marketplace, went out of business. The surviving airlines had to cut
costs, which led to a decline in the wages of airline employees. But the
point of deregulation was to benefit consumers by forcing the airlines
to compete. From consumers’ point of view, the deregulation of the
airlines was a tremendous success.6
Similar events took place, with similar results, in several other markets
in ensuing years. Until 1980, trucking rates were set by “rate bureaus”
composed of trucking firms and overseen by the Interstate Commerce
Commission. The Motor Carrier Act of 1980 brought price competition
to trucking. “The purpose of the legislation is clear,” declared President
Jimmy Carter in his signing statement. “Protective and wasteful regula-
tions are to be replaced wherever possible by competition and the dis-
cipline of the free market. These changes will work to the benefit of all
consumers.” They did just that. Trucking prices declined steadily over
the ensuing decade. Gains to consumers were losses to trucking firms
and to truck drivers, whose wages underwent a corresponding decline.7
But that was precisely the purpose of deregulation—​to subject produc-
ers to price competition, for the benefit of consumers.

6 Steven Morrison and Clifford Winston, The Economic Effects of Airline


Deregulation (Washington, D.C.: Brookings Institution, 1986); Elizabeth
E. Bailey, David R. Graham, and Daniel P. Kaplan, Deregulating the Airlines
(Cambridge, Mass.: MIT Press, 1985); Thomas Gale More, “U.S. Airline
Deregulation: Its Effects on Passengers, Capital, and Labor,” Journal of Law &
Economics 29 (1986): 1–​28.
7 John Richard Felton and Dale G. Anderson, eds., Regulation and Deregulation
of the Motor Carrier Industry (Ames: Iowa State University Press, 1989); Dorothy
Robyn, Braking the Special Interests: Trucking Deregulation and the Politics of Policy
Reform (Chicago: University of Chicago Press, 1987); Carter’s signing statement
at http://​www.presidency.ucsb.edu/​ws/​?pid=44689; Veiko Parming, Competition
and Productivity in the U.S. Trucking Industry Since Deregulation (Cambridge,
312

312 Speculation

The railroad industry was also deregulated, in the Staggers Rail Act
of 1980. Price competition among railroads benefited consumers by re-
ducing the cost of shipping goods by rail. Some railroad lines went out
of business because they were unable to compete, and the surviving
railroads paid lower wages than they had before. Markets in telephone
service and natural gas were also opened to price competition during
this period, with broadly similar results.8
The lesson seemed to be that elementary economics was exactly
right. Regulatory schemes dating back to the 1930s insulated produc-
ers from competition. Deregulation forced these firms to compete on
price, which made life more difficult for the firms and their employ-
ees, but which yielded enormous benefits for consumers. Finance was
one more heavily regulated industry, also dating back to the 1930s. The
consumers of finance were investors on one side and enterprises seek-
ing investment on the other; the producers were the banks, the stock
exchanges, and the various other intermediaries between the two. The
experience of deregulation in the last quarter of the twentieth century
suggested that here, too, deregulation could facilitate competition
among producers and thus gains for consumers.
The period was one in which economic conditions were unusually
conducive to this view. Between the early 1980s and the late 2000s,
the volatility of economic activity declined quite sharply. Virtually
every aggregate measure—​ output, employment, inflation—​ became
much more stable than it had ever been. The business cycle seemed
to have nearly disappeared. By 2002, economists were speaking of a
“Great Moderation” that was taking place, not just in the United States
but throughout the industrialized world. The “problem of depression
prevention has been solved,” declared the Nobel Prize winner Robert

Mass.: MIT Press, 2013); Michael H. Belzer, Sweatshops on Wheels: Winners and


Losers in Trucking Deregulation (New York: Oxford University Press, 2000).
8 Clifford Winston, The Success of the Staggers Rail Act of 1980, AEI-​Brookings Joint
Center for Regulatory Studies Publication 05-​24 (2005); Ernst R. Berndt et al.,
“Cost Effects of Mergers and Deregulation in the U.S. Rail Industry,” Journal
of Productivity Analysis 4 (1993):  127–​44; Clifford Winston, “U.S. Industry
Adjustment to Economic Deregulation,” Journal of Economic Perspectives 12
(1998): 89–​110; James Peoples, “Deregulation and the Labor Market,” Journal of
Economic Perspectives 12 (1998): 111–​30.
  313

Deregulation and Crisis? 313

Lucas. Ben Bernanke, then one of the governors of the Federal Reserve
System, was only slightly less triumphant. He cited “the increased depth
and sophistication of financial markets, deregulation in many indus-
tries, the shift away from manufacturing toward services, and increased
openness to trade and international capital flows,” along with “better
monetary policy,” as causes of what appeared to be a permanently stable
economy.9
One persistent (although persistently disputed) rationale for limit-
ing speculation had always been that speculators amplified the busi-
ness cycle—​that they made the highs even higher and the lows even
lower. But if the business cycle was no longer such a serious problem,
this rationale was no longer as important. The Great Moderation thus
contributed to an intellectual climate receptive to deregulating finan-
cial markets, by weakening one of the traditional arguments for regula-
tion. As the federal judge Richard Posner observed in his autopsy of the
crisis, deregulation “was a government failure abetted by the political
and ideological commitments of mainstream economists, who over-
looked the possibility that the financial markets seemed robust because
regulation had prevented previous financial crises.”10
The intellectual climate of the late twentieth century was also in-
fluenced by a view of financial markets called the efficient market
hypothesis. Here we need to be careful, because there is a mis-
understanding of the efficient market hypothesis that pervades
much of the post–​financial crisis critique of deregulation and that

9 Steven J. Davis, “Interpreting the Great Moderation: Changes in the Volatility


of Economic Activity at the Macro and Micro Levels,” Journal of Economic
Perspectives 22 (2008): 155–​80; James H. Stock and Mark W. Watson, “Has the
Business Cycle Changed and Why?,” in NBER Macroeconomics Annual 2002, ed.
Mark Gertler and Kenneth Rogoff (Cambridge, Mass.: MIT Press, 2003), 162;
Peter M. Summers, “What Caused the Great Moderation? Some Cross-​Country
Evidence,” Federal Reserve Bank of Kansas City Economic Review: Third Quarter
2005, 5–​32; Robert E. Lucas Jr., “Macroeconomic Priorities,” American Economic
Review 93 (2003): 1; Ben S. Bernanke, “The Great Moderation” (remarks at the
meeting of the Eastern Economic Association, 2004), http://​www.federalreserve
.gov/​BOARDDOCS/​SPEECHES/​ 2004/​20040220/​default.htm.
10 Richard A. Posner, A Failure of Capitalism: The Crisis of ’08 and the Descent Into
Depression (Cambridge, Mass.: Harvard University Press, 2009), 260.
314

314 Speculation

ascribes too much direct causal force to the EMH.11 In correcting this
mistake, however, it would be equally wrong to deny any connection
between the EMH and one’s view as to the propriety of regulation.
The point that deserves emphasis is that this connection was more dif-
fuse and perhaps even more emotional than is usually acknowledged.
The efficient market hypothesis simply states that the prices in a
capital market (such as a stock market) fully reflect all available rel-
evant information. The implication is that investing in stocks chosen
by so-​called experts will be no more profitable than investing in stocks
chosen at random. The EMH was a staple of the academic literature by
the early 1970s. Since then it has been criticized, defended, and restated
in various forms, but in all versions it is an empirical claim about how
quickly new information affects the prices of securities.12 The EMH is
not a normative claim that markets are good or that regulation is bad.
This point was sometimes lost in the aftermath of the financial
crisis, when critics blamed the efficient market hypothesis for caus-
ing regulators to fall asleep at the wheel. “The upside of the current
Great Recession is that it could drive a stake through the heart of the
academic nostrum known as the efficient-​market hypothesis,” thun-
dered the journalist Roger Lowenstein. “The mistaken faith in mar-
kets turned regulators into fawning groupies.” The investor Jeremy
Grantham insisted that the efficient market hypothesis “left our eco-
nomic and government establishment sitting by confidently, even as
a lethally dangerous combination of asset bubbles, lax controls, per-
nicious incentives and wickedly complicated instruments led to our
current plight.”13 This sort of criticism is misdirected, if taken liter-
ally. The efficient market hypothesis has no necessary relationship to

11 For a forceful expression of this view, see Ray Ball, “The Global Financial
Crisis and the Efficient Market Hypothesis: What Have We Learned?,” Journal
of Applied Corporate Finance 21, no. 4 (2009): 8–​16.
12 Eugene F. Fama, “Efficient Capital Markets: A Review of Theory and Empirical
Work,” Journal of Finance 25 (1970):  383–​ 417; Burton G.  Malkiel, “The
Efficient Market Hypothesis and Its Critics,” Journal of Economic Perspectives
17 (2003): 59–​82.
13 Roger Lowenstein, “Book Review:  ‘The Myth of the Rational Market,’ ”
Washington Post, 7 June 2009; Joe Nocera, “Poking Holes in a Theory on
Markets,” New York Times, 6 June 2009.
  315

Deregulation and Crisis? 315

the appropriate level of regulation. One can believe that stock prices
incorporate all available information and simultaneously believe that
certain transactions should not be allowed, for any of the reasons
conventionally offered for prohibiting transactions. For example, an
adherent of the EMH can still favor limits on speculating with bor-
rowed money, on the ground that highly leveraged transactions are too
risky, whether for individual speculators or for the financial system as
a whole. To call the market “efficient” is merely to say that stock prices
cannot be predicted; the term does not imply that whatever we observe
in the market must be good.
On the other hand, there is a less direct relationship between the effi-
cient market hypothesis and one’s attitude toward financial regulation,
one that may have facilitated deregulation. The EMH implies that if in-
vestors have no way of knowing whether prices are too high or too low,
neither do regulators. There is no way to identify a bubble while it is
occurring, or even afterward—​the best one can say is that prices at the
peak reflected the information available at the peak, and prices at the
trough reflected the information available at the trough. When it comes
to prices, regulators are no wiser than investors. A person accustomed
to thinking of markets in this way can easily grow skeptical that regula-
tors will be any wiser than investors when it comes to other matters,
like whether investors have taken on too much risk, or whether inves-
tors fully understand the securities they are buying. Attitudes toward
regulation are based on moods or general feelings as much as on careful
analyses of the pros and cons of particular regulatory programs. This
point obviously should not be taken too far. There were booms and
crashes, and arguments for and against regulation, long before anyone
thought of the efficient market hypothesis. The EMH did not cause the
financial crisis or even any of the legal developments that preceded it.
But the EMH may have helped to propagate a mood that was condu-
cive to deregulation.
There was considerable weakening, at the same time, in Americans’
traditional moral disapproval of gambling. In the 1970s, casino gambling
was lawful only in Nevada, but by the end of the century there were
casinos in nearly half the states and on many Indian reservations as well.
At midcentury there were no state-​run lotteries, but by the century’s end
almost every state ran a lottery. Denunciations of speculative financial
316

316 Speculation

transactions as gambling carried significant weight when gambling was


generally considered a social evil, but they gradually lost their force as
gambling was reconceived as a respectable activity.
Toward the end of the twentieth century, then, the intellectual cli-
mate was just right for financial deregulation. Meanwhile, the financial
markets were rapidly expanding, particularly markets for derivatives,
which increased the political support for deregulation.
The idea of a derivative was already centuries old. A  derivative is
simply an asset whose value is derived from the value of some other
asset. An option to buy a share of stock is a derivative, for example; its
value is determined in part by the value of the stock. The commodity
futures bought and sold on boards of trade beginning in the middle of
the nineteenth century were derivatives whose value depended in part
on the value of the underlying bushels of wheat and corn. But if the
concept of derivatives was not new, the particular kinds of derivatives
were. In the late twentieth century, banks invented all sorts of deriva-
tive contracts that would have been impractical to trade before comput-
ers became standard equipment. The notional value of the derivatives
market more than doubled between 1995 and 2001, more than doubled
again between 2001 and 2004, and then more than doubled again be-
tween 2004 and 2007. More than $500 trillion of derivative contracts
were outstanding by June 2007. Notional values are misleading, be-
cause they bear no necessary relationship to the values of any real assets
or potential losses, but there is little doubt that the derivatives market
was growing much faster than the real economy. By the end of 2007,
when the notional value of the derivatives market was nearing $600
trillion, the combined worldwide value of all publicly traded stocks was
less than $70 trillion.14
The new derivatives flourished in part because they facilitated the
hedging of risks that had previously been difficult or impossible to
hedge. Older derivatives like commodity futures had allowed farmers
and other participants in the grain trade to protect themselves against

14 These figures come from the triennial reports published by the Bank for
International Settlements. They are available at the bank’s website, http://​www
.bis.org.
  317

Deregulation and Crisis? 317

the risk that prices would change between one time period and another.
The new derivatives allowed companies to hedge all sorts of other risks,
like changes in currency exchange rates, changes in interest rates, or
even bad weather. For example, a firm that had to make investments
in the present in one currency but expected future revenue in a differ-
ent currency could enter into a derivative contract that would allow
it to remove the risk of loss from an adverse change in the exchange
rate between the two currencies, by paying a counterparty to absorb
that risk. A firm with obligations to pay debt at a variable interest rate
but with expected income at a fixed rate could enter into a derivative
contract that would allow it, in effect, to convert its variable interest to
fixed interest, by swapping debt obligations with a counterparty will-
ing to absorb the risk of adverse changes in the variable interest rate.
Unlike the older, simpler derivatives, which had been standard con-
tracts traded on exchanges, the new derivatives tended to be custom
tailored to the needs of each individual purchaser and were thus not
traded on exchanges. But they served the same hedging purpose as the
older derivatives, by allowing parties to transfer risk, for a price, to
those willing to bear it.
And, as with the older derivatives, the hedgers were only part of the
market. It would be quite a coincidence if a firm wishing to hedge a
particular risk could find another firm that simultaneously wished to
hedge an equal and opposite risk, so the hedger’s counterparty in these
transactions was normally not a hedger but a speculator—​someone
willing to take on risk in the expectation of making a profit. And many
transactions involved no hedgers at all, but rather speculators on both
sides, who simply had different expectations of what the future would
bring.15 The new derivatives, like the old, thus opened up new methods
of speculation.
As always, with new methods of speculation came new ways of
making or losing large sums of money very quickly. Startling stories
of massive losses from trading in derivatives, sometimes incurred by

15 Sergey Chernenko and Michael Faulkender, “The Two Sides of Derivatives


Usage:  Hedging and Speculating with Interest Rate Swaps,” Journal of
Financial and Quantitative Analysis 46 (2011): 1727–​54.
318

318 Speculation

seemingly sophisticated institutions, were recurring features in news


reports throughout the 1990s.16
In 1994, for example, Procter & Gamble announced that it had
lost $157  million in interest rate swaps. The New  York Times won-
dered: “What is a soap company doing in the swap market speculat-
ing with hundreds of millions of dollars?” For many years, Procter &
Gamble, like most large international companies, had used swaps to
hedge against changes in exchange rates and interest rates. But in late
1993 and early 1994, Procter & Gamble agreed to two atypically large
and complex contracts with Bankers Trust, in which Procter & Gamble
would pay a floating interest rate to Bankers Trust in exchange for re-
ceiving a fixed interest rate from Bankers Trust. These contracts proved
disastrous to Procter & Gamble when interest rates rose. The Procter
& Gamble officers who were responsible for these deals apparently did
not understand their full consequences, and indeed Procter & Gamble
sued Bankers Trust on the theory that Bankers Trust had misled Procter
& Gamble about the transactions. “Derivatives like these are danger-
ous and we were badly burned,” Procter & Gamble’s chairman told the
press. “We won’t let that happen again.”17
Later that year, Orange County, California, declared bankruptcy
after losing $1.65 billion from leveraged investments in derivatives
with payoffs that varied inversely with interest rates. County Treasurer
Robert Citron had been re-​elected for more than twenty years, because
he consistently earned higher returns than his counterparts in other
municipalities. As it turned out, the secret of his ostensible success was
that he was taking more risks than his counterparts, and when interest
rates rose, the Orange County Investment Pool lost big. The elected
officials who could have inquired into Citron’s strategy, before it was
too late, appear to have lacked the financial knowledge to understand
it. It is not entirely clear whether Citron himself fully understood the

16 Laurent L.  Jacque, Global Derivative Debacles:  From Theory to Malpractice


(Singapore: World Scientific, 2010).
17 Lawrence Malkin, “Procter & Gamble’s Tale of Derivatives Woe,” New York
Times, 14 Apr. 1994; Procter & Gamble Co. v.  Bankers Trust Co., 925
F. Supp. 1270, 1276–​77 (S.D. Ohio 1996).
  319

Deregulation and Crisis? 319

risk of his investments—​it was later discovered that he relied on an


astrologer and a psychic to foretell the course of interest rates.18
Even more astonishing was the demise of Barings Bank the follow-
ing year, when one of its traders in Singapore managed to lose $1.3
billion before his supervisors noticed. Barings was as well established
as a bank could be. It was founded in 1762. It was the underwriter for
the Louisiana Purchase. But a single employee in his twenties brought
the bank to an end by speculating in derivatives linked to the Nikkei
225 stock index. When the Kobe earthquake of January 1995 caused
Japanese stock prices to plummet, Barings was wiped out.19
Perhaps most surprising of all was the fate of Long Term Capital
Management, a hedge fund run by some of the most sophisticated
investors in the world. LTCM’s partners included two Nobel Prize–​
winning economists, some former faculty members at Harvard Business
School, and several veterans of the investment bank Salomon Brothers.
LTCM began by engaging in nearly riskless arbitrage, but when such
opportunities grew scarce the fund shifted to riskier transactions. After
the Asian financial crisis of 1997 and the Russian financial crisis of
1998, LTCM lost more than $4 billion. The Federal Reserve Bank of
New York organized a bailout by the fund’s main creditors, to prevent
other firms from failing as well.20
These losses seemed enormous at the time, but much larger losses
would accompany the financial crisis of 2007–​08. The government’s
bailout of the insurance company AIG, for example, totaled $182 bil-
lion, much of which AIG lost by investing indirectly but heavily in real
estate, through the purchase of mortgage-​backed securities and the sale
of credit default swaps (effectively insurance against defaults) on securi-
ties backed largely by mortgages.21

18 Mark Baldassare, When Government Fails:  The Orange County Bankruptcy


(Berkeley: University of California Press, 1998).
19 Report of the Board of Banking Supervision Inquiry Into the Circumstances of the
Collapse of Barings (London: HMSO, 1995).
20 Roger Lowenstein, When Genius Failed: The Rise and Fall of Long Term Capital
Management (New York: Random House, 2000).
21 Robert McDonald and Anna Paulson, “AIG in Hindsight,” Journal of Economic
Perspectives 29, no. 2 (2015): 81–​106.
320

320 Speculation

The new derivatives were extraordinarily useful, because they al-


lowed parties to hedge against risks that had previously been unhedge-
able. But they were also extraordinarily dangerous, because, just like the
old derivatives, they opened up forms of speculation in which even the
most sophisticated investors could lose everything.
This was not a new dilemma. For two centuries Americans had
argued about this tension between the positive and negative aspects of
speculation. The law had shifted this way and that, in a collective effort
to limit the dangers of speculation without destroying its benefits. The
new derivatives markets of the late twentieth and early twenty-╉first cen-
turies gave rise to the most recent installment of this debate, particularly
after the market crashed in 2007–╉08. Had deregulation contributed to
the crisis by permitting methods of speculation that would have been
prohibited in an earlier era?

A Wide Range of€Financial Activities


The most conspicuous example of financial deregulation, and the one
perhaps most often cited as a cause of the crisis, was the Gramm-╉Leach-╉
Bliley Act of 1999.22 The very first sentence of the law declared that it
was repealing the Glass-╉Steagall Act of 1933, which barred commercial
banks from entering the securities business. Back in 1933, the purpose
of separating commercial banks from investment banks had been to
prevent banks from speculating with their depositors’ funds. But the
wall between commercial and investment banking was knocked down
with bipartisan support in Congress. “The Gramm-╉Leach-╉Bliley Act
makes the most important legislative changes to the structure of the
U.S. financial system since the 1930s,” declared President Bill Clinton
in his signing statement. “Financial services firms will be authorized to
conduct a wide range of financial activities, allowing them freedom to
innovate in the new economy.”23

22 See, e.g., Joseph Karl Grant, “What the Financial Services Industry Puts
Together Let No Person Put Asunder: How the Gramm-╉Leach-╉Bliley Act
Contributed to the 2008–2009 American Capital Markets Crisis,” Albany Law
Review 73 (2010): 371–╉420.
23 Clinton’s signing statement at http://╉www.presidency.ucsb.edu/╉ws/╉?pid=56922.
  321

Deregulation and Crisis? 321

The Gramm-​Leach-​Bliley Act was a product of long-​term intellec-


tual and institutional changes. For many years, the economists who
studied the issue most closely had concluded that, contrary to what was
widely believed in the 1930s, the commercial banks’ securities affiliates
had in fact not contributed to the stock market crash of 1929 or the
long depression that followed. Using statistical tools that had not been
available in 1933, the economists argued that separating commercial
from investment banks had been a costly mistake all along.24
Meanwhile, after decades of regulatory changes, the supposed bar-
rier between commercial and investment banking had already become
quite flimsy. On one side of the wall, the higher interest rates of the
mid-​twentieth century made commercial bank deposits unattractive
from the customer’s point of view, which led institutions that were not
formally commercial banks to offer higher-​return substitutes for bank
accounts. On the other side, commercial banks repeatedly sought to re-​
enter the lucrative securities business. As early as 1957, the Comptroller
of the Currency ruled that commercial banks could offer profit-​making
brokerage services for their customers. By the 1980s, the Federal Deposit
Insurance Corporation and the Federal Reserve allowed commercial
banks, with some restrictions, to underwrite securities. Such regulatory
interpretations of the Glass-​Steagall Act were by and large upheld by the
courts when they were challenged. By 1999, the Gramm-​Leach-​Bliley

24 George J. Benston, The Separation of Commercial and Investment Banking: The


Glass-​Steagall Act Revisited and Reconsidered (New  York:  Oxford University
Press, 1990); Eugene Nelson White, “Before the Glass-​ Steagall Act:  An
Analysis of the Investment Banking Activities of National Banks,”
Explorations in Economic History 23 (1986):  33–​55; Randall S.  Kroszner and
Raghuram G. Rajan, “Is the Glass-​Steagall Act Justified? A Study of the U.S.
Experience with Universal Banking Before 1933,” American Economic Review
84 (1994): 810–​32; Carlos D. Ramirez, “Did Glass-​Steagall Increase the Cost
of External Finance for Corporate Investment? Evidence from Bank and
Insurance Company Affiliations,” Journal of Economic History 59 (1999): 372–​
96; Carlos D. Ramirez and J. Bradford DeLong, “Understanding America’s
Hesitant Steps Toward Financial Capitalism:  Politics, the Depression, and
the Separation of Commercial and Investment Banking,” Public Choice 106
(2001): 93–​116; Carlos D. Ramirez, “Did Banks’ Security Affiliates Add Value?
Evidence From the Commercial Banking Industry During the 1920s,” Journal
of Money, Credit and Banking 34 (2002): 393–​411.
322

322 Speculation

Act was ratifying changes that had already taken place more than it
was actually making any change. “The Glass-​Steagall Act was already a
dead letter when Gramm-​Leach-​Bliley was passed,” concluded the law
professor Jonathan Macey. “All the Act did was formalize the death.”25
Back in the early 1930s, much of the motivation for the Glass-​Steagall
Act had been the cascade of bank failures, many of which contempo-
raries attributed to speculation. By the late twentieth century, banks
still failed, but bank failures had lost their sharp edge because of the
advent of deposit insurance. Losses from bank failures were now borne
by the banking system generally and by the government rather than by
the particular depositors unfortunate enough to have entrusted their
funds to a now-​insolvent bank. In 1933, bank depositors were keen to
keep their money from being sent in risky directions. In 1999, they were
no longer as worried.
Because the Gramm-​Leach-​Bliley Act was the culmination of incre-
mental changes rather than a major change in its own right, it makes
sense to ask, not whether the act itself contributed to the financial
crisis, but whether the crisis was facilitated by the gradual reblending
of commercial and investment banking. The answer is almost certainly
no. The Glass-​Steagall Act would not have prevented any banks from
making subprime mortgage loans, investing in mortgage-​backed se-
curities, or being so highly leveraged that a small drop in the value
of their assets would wipe them out. The activities that commercial
banks were newly allowed to enter were not the ones that lost them
vast sums of money. And many of the financial institutions that lost the

25 Philip A. Wallach, “Competing Institutional Perspectives in the Life of Glass-​


Steagall,” Studies in American Political Development 28 (2014): 26–​48; Robert
E. Lucas Jr., “Glass-​Steagall: A Requiem,” American Economic Review: Papers
and Proceedings 103, no.  3 (2013):  45–​46; George G.  Kaufman and Larry
R. Mote, “Glass-​Steagall: Repeal by Regulatory and Judicial Reinterpretation,”
Banking Law Journal 107 (1990): 388–​421; Donald C. Langevoort, “Statutory
Obsolescence and the Judicial Process: The Revisionist Role of the Courts in
Federal Banking Regulation,” Michigan Law Review 85 (1987): 672–​733; James
R. Barth, R. Dan Brumbaugh Jr., and James A. Wilcox, “The Repeal of Glass-​
Steagall and the Advent of Broad Banking,” Journal of Economic Perspectives 14
(2000): 191–​204; Jonathan R. Macey, “The Business of Banking: Before and
After Gramm-​Leach-​Bliley,” Journal of Corporation Law 25 (2000): 692.
╇ 323

Deregulation and Crisis? 323

most were investment banks and insurance companies, not commercial


banks, so the Glass-╉Steagall Act would not have constrained them. The
financial crisis would most likely have looked very similar even if the
Glass-╉Steagall Act had still been in full force.26
But if separating the functions of commercial and investment banks
would not have prevented the financial crisis, there were many who nev-
ertheless thought it wise for a different reason. The federal government
insures the deposits in commercial banks, but there is no analogous
insurance for the assets of investment banks. Deposit insurance raises
two concerns: that taxpayers will be required to bail out a commercial
bank that speculates unsuccessfully with its depositors’ money, and that
the managers of commercial banks, aware of this safety net, will take
undue risks with the depositors’ money. Paul Volcker, the former chair
of the Federal Reserve, accordingly suggested after the crisis that com-
mercial banks should be barred from the most speculative transactions.
The idea quickly became known as the “Volcker Rule” when President
Barack Obama referred to it by that name. The Volcker Rule was en-
acted into law as part of the Dodd-╉Frank Act of 2010. Commercial
banks are now barred from “proprietary trading” and from owning an
interest in “a hedge fund or private equity fund.”27 The transactions
now off-╉limits to commercial banks are not exactly congruent with
those once proscribed by the Glass-╉Steagall Act. But the purpose of
the two laws is the same—╉to prevent banks from speculating with the
funds entrusted to them by depositors.

A Giant, Global, Electronic€Ponzi


A less well-╉known set of changes in the law, culminating in the Commodity
Futures Modernization Act of 2000, has a more plausible link to the

26 Paul G. Mahoney, Wasting a Crisis: Why Securities Regulation Fails (Chicago:


University of Chicago Press, 2015), 155–╉56; Jerry W. Markham, “The Subprime
Crisis—╉A Test Match for the Bankers: Glass-╉Steagall vs. Gramm-╉Leach-╉
Bliley,” University of Pennsylvania Journal of Business Law 12 (2010): 1081–╉134;
Lawrence J. White, “The Gramm-╉Leach-╉Bliley Act of 1999: A Bridge Too Far?
Or Not Far Enough?,” Suffolk University Law Review 43 (2010): 937–╉56.
27 Paul Volcker, “How to Reform Our Financial System,” New  York Times, 31
Jan. 2010; New York Times, 23 Jan. 2010, B1; 12 U.S.C. § 1851(a)(1).
324

324 Speculation

financial crisis, although it is not clear that “deregulation” is the right word
for these changes, and even here the connection to the financial crisis is
uncertain.
In the 1980s, when the new derivatives became commercially signifi-
cant, transactions in derivatives were still largely governed by the legal
framework established in the 1920s, when the main derivatives were
options and futures relating to agricultural commodities. The Grain
Futures Act of 1922 had been replaced by the Commodity Exchange
Act of 1936, but both statutes essentially required derivatives to be
traded on organized exchanges approved by the federal government.
Many states still had their anti–​bucket shop laws on the books, and
in principle the common law of every state still rendered unenforce-
able all off-​exchange transactions in which the parties did not actually
intend the delivery of a physical commodity. In practice, though, state
law largely fell by the wayside after the enactment of the Commodity
Exchange Act, which created an administrative agency, the Commodity
Exchange Authority (later replaced by the Commodity Futures Trading
Commission), charged with overseeing derivative trading nationwide.
At first, the new derivatives of the 1980s occupied an uncertain
position within this legal framework.28 They tended to be tailored to
individual circumstances rather than being standardized products, so
they were not traded on exchanges. Did that make them unlawful? In
swap transactions, for example, parties exchanged future cash flows.
“The economic reality of swaps,” the CFTC observed, “resembles that
of futures contracts.” But were they futures contracts? If so, they were
illegal if they were not traded on an exchange, and they were subject
to the full array of CFTC regulation. In 1989, the CFTC determined
that swaps were not futures contracts so long as, among other things,
their terms were individually tailored and they were entered into “in
conjunction with a line of business”—​that is, they were limited to
hedgers and financial institutions. The CFTC declared that it would

28 This discussion compresses and simplifies an intricate story; for more details,
see Philip McBride Johnson and Thomas Lee Hazen, Derivatives Regulation
(New York: Aspen, 2004), 1:50–​85.
  325

Deregulation and Crisis? 325

not regulate such transactions.29 In later years the CFTC would issue
similar determinations for other new kinds of derivatives.
But that only raised a new question. If the new derivatives were not
subject to CFTC regulation, did that mean they were governed by state
law instead? Could a state court condemn them as gambling transac-
tions in violation of state anti–​bucket shop statutes or the common-​law
prohibition on contracts without intent to deliver? Congress ended this
uncertainty three years later, by declaring that federal law in this area
preempted state law. The Futures Trading Practices Act of 1992 pro-
vided that derivatives were not subject to “any State or local law that
prohibits or regulates gaming or the operation of ‘bucket shops.’ ”30 The
new derivatives had cleared the last legal hurdle.
Some government officials were uncomfortable with allowing un-
fettered derivatives trading. In 1994, on the same day that Procter
& Gamble announced it had lost $157  million in swaps, the House
Banking Committee held hearings on the topic. The purchaser of a
derivative is “entering into a gamble,” worried Representative Henry
Gonzalez of Texas, the committee chair. “I picture it as an inverted
pyramid where the apex is a basic nominal value and then derived up
to this tremendous base are offshoots of all these formidable definitions
and forms of derivatives.” As Gonzalez saw the new market, participants
were not profiting “in the usual financial term of things, but by out and
out gambling.” He foresaw “the possibility of a giant, global, electronic
Ponzi” that one day would come crashing down. The financier George
Soros shared Gonzalez’s concern. “Some of those instruments appear to
be specifically designed to enable institutional investors to take gambles
which they would not otherwise be permitted to take,” he explained.
“For example, some bond funds have invested in synthetic bond issues
which carry a tenfold or twentyfold multiple of the normal risk within
defined limits. And some other instruments offer exceptional returns
because they carry the seeds of a total wipeout.” If a meltdown were
to occur, Soros noted, “the regulatory authorities may find themselves
obliged to step in to preserve the integrity of the system. It is in that

29 “Policy Statement Concerning Swap Transactions,” Federal Register 54


(1989): 30694.
30 106 Stat. 3590, 3631–​32 (1992).
326

326 Speculation

light that the authorities have both the right and an obligation to su-
pervise and regulate derivative instruments.”31
Within the Clinton administration, the leading voice in favor of reg-
ulation was the Washington lawyer Brooksley Born, who was chair of
the CFTC between 1996 and 1999. Born repeatedly urged greater gov-
ernment oversight of the ballooning derivatives market, particularly the
idea of requiring buyers and sellers of derivatives to make disclosures of
their financial positions so their counterparties would be able to evalu-
ate their creditworthiness. “No reporting requirements are imposed
on most OTC derivatives market participants,” she observed shortly
after the collapse of Long Term Capital Management, in a speech that
would look quite prescient a decade later. “This lack of basic informa-
tion about the positions held by OTC derivatives users and about the
nature and extent of their exposures potentially allows OTC derivatives
market participants to take positions that may threaten our regulatory
markets or, indeed, our economy without the knowledge of any fed-
eral regulatory authority.”32 But Born’s view was not shared by senior
government economic officials, including Treasury Secretary Robert
Rubin and Federal Reserve Chair Alan Greenspan, who believed that
market participants already had every incentive to evaluate their coun-
terparties’ creditworthiness. “Individuals who lend money to others,”
Greenspan responded, “have a very important interest in getting that
money back.”33
Instead, the Clinton administration went in the opposite direction.
The President’s Working Group on Financial Markets, made up of
Greenspan, Lawrence Summers (Rubin’s successor as Treasury secretary),
William Rainer (Born’s successor as CFTC chair), and Arthur Levitt
(the chair of the Securities and Exchange Commission), concluded

31 Risks That Hedge Funds Pose to the Banking System: Hearing Before the Committee
on Banking, Finance and Urban Affairs (Washington, D.C.: Government
Printing Office, 1994), 14, 38.
32 Richard B. Schmitt, “The Born Prophecy,” ABA Journal, May 2009, 50–​55;
Brooksley Born, “The Lessons of Long-​Term Capital Management L.P.” (Oct.
15, 1998), http://​www.cftc.gov/​opa/​speeches/​opaborn-​37.htm. “OTC” stands
for “over the counter”—​that is, not traded on an exchange.
33 Greenspan quoted in Sebastian Mallaby, More Money Than God: Hedge Funds
and the Making of a New Elite (New York: Penguin, 2010), 245.
  327

Deregulation and Crisis? 327

in 1999 that all trading of financial derivatives by sophisticated partici-


pants like banks and pension funds should be exempt from the regulatory
regime that had long applied to the older commodity-​based derivatives.
Congress implemented this proposal the following year with over-
whelming bipartisan support. The Commodity Futures Modernization
Act of 2000 provided that financial derivatives were not subject to the
requirement of being traded on an exchange and were not otherwise
subject to the rules governing the older derivatives.34
Whether to call this “deregulation” depends on one’s opinion of what
the law had been before. There are two plausible ways to tell the story.
On one view, ever since the 1920s, the organized exchanges had held a
government-​mandated monopoly on the trading of derivatives, so that
market participants would have some assurance that their counterpar-
ties would not default. That monopoly was broken by the Commodity
Futures Modernization Act, which allowed derivatives to be traded out-
side of exchanges, thus removing the protections against default that
had wisely been in place for three-​quarters of a century. If we tell the
story this way, it is a tale of deregulation. On the other view, the system
of regulation adopted in the 1920s was for the older agricultural deriva-
tives. When the new financial derivatives began to be traded, there was
genuine uncertainty as to whether the old rules should apply to the new
circumstances. The Commodity Futures Modernization Act removed
this uncertainty by clarifying that only the older derivatives had to be
traded on exchanges, thus freeing up sophisticated institutional traders
to buy and sell financial derivatives among one another. This was the
view taken by the President’s Working Group on Financial Markets,
who argued that exempting the new derivatives was necessary because
“uncertainty arises from concerns under current law as to whether some
of these contracts could be construed to be subject to the” Commodity
Exchange Act. And it was the view taken by Congress:  the title of
the relevant section of the Commodity Futures Modernization Act is
“Legal Certainty for Excluded Derivative Transactions.”35 If we tell the

34 Over-​the-​Counter Derivatives Markets and the Commodity Exchange Act: Report


of the President’s Working Group on Financial Markets (Nov. 1999), http://​www
.treasury.gov/​resource-​center/​fin-​mkts/​Documents/​otcact.pdf; 114 Stat. 2763A
(2000), at 377, 404.
35 Over-​the-​Counter Derivatives Markets, 6; 114 Stat. 2763A (2000), at 377.
328

328 Speculation

story this way, “nonregulation” would be a better word than “deregula-


tion,” because it was not clear that there was any preexisting law that
had been removed.
But it makes little difference what we call it. Either way, there can
be little doubt that speculation in the new derivatives was facilitated
by exempting them from the rules governing the old derivatives, es-
pecially the requirement of being traded on an exchange. Some of the
new derivatives were custom tailored to the hedging needs of particular
firms and were thus impractical to trade on an exchange. To require
such contracts to be traded on an exchange would have been equivalent
to prohibiting them. Other derivatives were more amenable to stan-
dardization. They could have been traded on an exchange, and indeed
after the crisis the Dodd-​Frank Act of 2010 directed the CFTC to es-
tablish such an exchange, along with a method of clearing transactions
modeled on the clearinghouses that had long been used in commodity
trading. It is likely that the absence of such requirements before 2010
helped the market grow so quickly.
The resulting increase in speculation may not have caused the finan-
cial crisis, but it almost certainly contributed to the size of the crisis.
There is, of course, no shortage of competing theories as to what caused
the crisis. But any plausible explanation of the magnitude of the crisis
has to take some account of the massive amount of mortgage-​backed
derivatives held by financial institutions, and the likewise massive
amount of credit default swaps linked to the health of those institu-
tions. The root cause of the problem may have been imprudent lending
to homebuyers who would be unable to pay back their loans unless
housing prices kept rising. No doubt all this lending would have led
to a crisis even without speculation in derivatives. There have been
similar boom-​and-​bust cycles for centuries, including in real estate,
without the benefit of exotic new financial products.36 But much of
the money available to lend was provided by investors in derivatives
backed by these mortgages. Without all this capital coming into the

36 Andrew W.  Lo, “Reading About the Financial Crisis:  A  Twenty-​One-​Book


Review,” Journal of Economic Literature 50 (2012): 151–​78; Carmen M. Reinhart
and Kenneth S.  Rogoff, This Time Is Different:  Eight Centuries of Financial
Folly (Princeton, N.J.: Princeton University Press, 2009).
  329

Deregulation and Crisis? 329

mortgage market, the lending might well have tapered off sooner, so
the crisis might not have been as severe. And when the housing bubble
burst, the firms facing insolvency were not just the banks that held
mortgage loans. Equally imperiled were a host of other institutions that
held derivatives linked to those loans, derivatives that had not existed
in previous housing bubbles. Speculation in derivatives thus may have
made the crisis more pervasive than it would otherwise have been.
After the crisis, some urged the prohibition of derivatives that are
used primarily for speculation (in the sense that they are bought and
sold because market participants have different expectations about the
future) rather than for some other purpose such as hedging. As the title
of one such effort put it, trading in such derivatives was simply “gam-
bling by another name.”37 “Financial products are socially beneficial
when they help people insure risks,” reasoned one pair of authors, “but
when those same products are used for gambling they can instead be
socially detrimental.” After earlier market downturns, critics had like-
wise urged that new financial products should be banned. Defenders
of the market had always responded by pointing out how difficult it
would be to distinguish the harmful transactions from the beneficial
ones. The most recent crisis once again elicited these arguments. Would
it even be possible for government officials to distinguish trading moti-
vated by hedging from trading motivated by differences in beliefs about
the future? Were purely speculative traders necessary to make a market
for traders wishing to hedge, in the same way that insurance would
be impossible to buy without the existence of insurance companies?
Were hedgers actually distinct entities from speculators, or did hedgers

37 Timothy E. Lynch, “Gambling by Another Name: The Challenge of Purely


Speculative Derivatives,” Stanford Journal of Law, Business & Finance 17
(2011):  67–​130; Saule T.  Omarova, “License to Deal:  Mandatory Approval
of Complex Financial Products,” Washington University Law Review 90
(2012):  63–​140; Eric A.  Posner and E.  Glen Weyl, “An FDA for Financial
Innovation: Applying the Insurable Interest Doctrine to Twenty-​First-​Century
Financial Markets,” Northwestern University Law Review 107 (2013): 1307–​57.
For an earlier suggestion along similar lines, see Lynn A. Stout, “Why the Law
Hates Speculators: Regulation and Private Ordering in the Market for OTC
Derivatives,” Duke Law Journal 48 (1999): 701–​86.
330

330 Speculation

speculate too?38 Such questions were no easier to answer in the early


twenty-​first century than they had been in earlier eras.
Among the questions raised by the financial crisis were thus some
that Americans had debated for more than two centuries. In the twenty-​
first century, as in preceding centuries, virtually everyone agreed that
investment was necessary and should be encouraged. While gambling
had shed some of its moral stigma, virtually everyone agreed that gam-
bling was often harmful and thus that it should be discouraged or even
prohibited. As always, speculation lay somewhere in the middle. Before
one could form an opinion about how best to regulate speculators, one
needed to distinguish between two kinds of risky commercial transac-
tions, the good and the bad. Drawing that line has never been easy, and
it has not become any easier today.

38 Posner and Weyl, “An FDA for Financial Innovation,” 1308; Darrell Duffie,
“Challenges to a Policy Treatment of Speculative Trading Motivated by
Differences in Beliefs,” Journal of Legal Studies 43 (2014): S173–​82; Ing-​Haw
Cheng and Wei Xiong, “Why Do Hedgers Trade So Much?,” Journal of Legal
Studies 43 (2014): S183–​207.
  331

q i n de x

Abbott, Charles, 64–​5 Banking Act of 1933, 164, Brissot de Warville, J.P., 10


Adams, Charles Francis, 10 205, 208 British Companies Act of
Adams, George, 57 Barings Bank, 319 1908, 151, 187
Adams, John, 5 Barnard, John, 61–​2 Brooks, Rosa, 308
Agassiz, Rodolphe, 241–​45 Batchelder, Charles, 143 Brooks, T.J., 119
Agnew, Daniel, 78–​9 Bates, James, 247 Brown, Henry
AIG, 319 Bates, John, 247 Gunnison, 226
Airline Deregulation Act of Beecher, Henry Ward, 14 Brown, W.C., 117
1978, 310 Berle, Adolf, 179, 182, 189, 200 Buck, William, 175
Akerson, George, 175 Berlin, Irving, 218, 231 bucket shops, 92–​104
American Bankers Association, Bernanke, Ben, 313 Bulkley, Robert, 180
150, 214 Berry, Nicholas, 81–​2 Burleson, Albert, 126
American Bar Association, Birkbeck, Morris, 22 Burmeister, George, 98
266, 302 Blackstone, William, 15, 47, 63 Bush, George W., 308
Ames, Fisher, 45 Blodgett, Henry, 59 Butterworth, Benjamin, 113
Amory, Francis, 280–​81 blue sky laws, 138–​63, 182–​3
Amory, John, 280–​81 Boesky, Ivan, 277 Cahill, Michael, 179
Amory, Thomas, 48 Bonsall, Sterling, 79 Caines, George, 41, 48
Angell, James Born, Brooksley, 326 Canright, Garfield, 157
Waterhouse, 159 Bowditch, Nathaniel capital gains tax, 280, 293–​302
Anson, Austin, 231 Ingersoll, 281 Capper, Arthur, 173–​4, 179
Armstrong, William, 30 Boyle, James, 133 Caraway, Thaddeus, 168
Aroni, Julius, 80 Brackenridge, Hugh Henry, 10 Carey, Mathew, 23
Atkeson, T.C., 135 Bradenbaugh, John, 89–​90, 95 Carr, Dabney, 44
Brandeis, Louis, 187–​8, 199 Carson, Robert, 210
Baldwin, Henry, 250 Bravman, M. Francis, 302 Carter, Jimmy, 310–​1
Baldwin, Veril, 231 Breed, William, 183 Cary, William, 271, 273
Bank of the United States, 11, Breyer, Stephen, 310 Cassard, George, 68
32–​3, 283 Brigham, J.H., 126 Cassel, Ernest, 1

331
332

332 Index

Catchings, Waddill, 196–7 credit default swaps, 235, of the late 1830s, 13,
caveat emptor, 247–​53, 266–​7 319, 328 23, 283
Chambers, Charles, 177 Crèvecoeur, Hector St. of 1907, 110, 144
Chase, Salmon, 7, 54 John de, 30 of 1929–​33, 164–​217
Chevalier, Michel, 7–​8 Crissinger, Daniel, 167 of 2007–​10, 233,
Chiarella v. United States, 275 Curtin, Andrew, 6 235–​40, 307–​30
Chicago Board of Trade Curtis, Ezra, 286 Edwards, J.R., 173
agrarian opposition to, Custer, George, 72 efficient market hypothesis,
83–​5, 105–​37 313–​5
campaign against bucket Dana, Francis, 48 Egan, Martin, 170
shops of, 96–​103 Dane, Nathan, 250 Eliot, Robert, 124
formation of, 66 Danforth, Charles, 81–​2 Ely, Richard, 112, 227
self-​regulation by, 91–​2, 178 Danforth, George, 254–​5 Emergency Home Finance Act
speculation on, 56–​8, Davis, C. Wood, 113, 126 of 1970, 234–​5
76–​7, 81–​2 Dean, Arthur, 189, 205 Emerson, Ralph Waldo, 28–​9
Christie, C.C., 96 Defoe, Daniel, 62 Emery, Henry Crosby, 131–​2
Citron, Robert, 318 Deming, William, 207 Ernst, Daniel, 201
Civil War, speculation during, Denison, Edward, 162–​3 Erskine, Thomas, 16
6–​7, 52–​4, 165 Denslow, Van Buren, 133
Clark, John Bates, 144 Depository Institutions Federal Home Loan Bank Act
Clark, R.C., 120 Deregulation and of 1932, 234
Clews, Henry, 141, 257 Monetary Control Act of Federal Home Loan Mortgage
Clinton, Bill, 308, 320 1980, 237 Corporation, 235
Cobbett, William, 15 depressions. See economic Federal National
Cochran, John, 195 downturns Mortgage Association,
Cohen, Benjamin, 188, 201 deregulation, 307–​30 234–​5
collateralized debt obligations, derivatives, 316–​20, 324–​9 Fessenden, William, 54
235, 238 Dew, Thomas, 13 Fidler, Isaac, 8
Commodity Futures Dewey, T. Henry, 78 Financial Crisis Inquiry
Modernization Act of Dickson, Frederick, 97 Commis­sion, 309
2000, 309, 323, 327 Dirks v. S.E.C., 276 Findley, William, 26
Commodity Futures Trading Disney, John, 40 Fisher, Irving, 166
Commission, 324–​26 Dix, John, 28 Fitch, Walter, 118
Conant, Charles, 112, 144 Dodd-​Frank Act of 2010, Fletcher, Duncan, 180
Cone, Solomon, 125 239–40, 323, 328 Flexner, Bernard, 193
Connely, E.F., 215 Dolley, Joseph, 138–​40, Flint, Walter, 157
Cook, Donald, 145–​9, 153–​4 Florida land boom, 218–​24
265, 269 Dos Passos, John, 73 Flynn, John T., 171, 213
Cook, William, 255 Douglas, Michael, 277 Foote, Ira, 56–​9
Cooley, Thomas, 80 Douglas, William O., 183, 190, Forbes, Allan, 174
Cooper, Thomas, 249 201–​2, 204 Foss, H.A., 123
Corbet, Thomas, 32 Drummond, Thomas, 70 Foster, Benjamin, 13
Corcoran, Thomas, 188, 201 du Pont, Alfred, 184 Foster, George, 14
Cotton Futures Act of Duff, E.A., 118 Frank, Jerome, 190,
1914, 134–​5 Dunlap, M.L., 83–​4 200, 202–​4
Coulson, Barbara, 307 Dwight, Timothy, 23 Frankfurter, Felix, 188–​9,
Counselman, Charles, 121 192, 193
Couzens, James, 168 Eccles, Marriner, 296 Frazer v. Gervais, 247
Cowan, Edgar, 53–​4 Eckles, A.H., 212 Frederick, J. George, 176
Cowdin, J. Cheever, 271–​2 economic downturns Freeman, Milton, 201, 268
Coxe, Tench, 20 of 1792, 13, 23, 35, 43, French, D.E., 225
crashes. See economic 45, 165 Freneau, Philip, 18
downturns of 1819, 13, 23 Friedman, Elisha, 297
  333

Index 333

Friedman, Milton, 308 Hanes, John W., 202 150–​3, 159, 161–​3, 173,


Frost, John, 31 Haney, Lewis, 165 224, 297–​8
Funston, Edward, 111 Hanson, Alexander, 42–​3 Irvine, L.C., 212
Future Trading Act of 1921, 136 Hardwicke, John, 231 Israels, Carlos, 272
Futures Trading Practices Act Hare, Robert, 11 Ives, Brayton, 141
of 1992, 325 Harris, E.H., 124
Hatch, William, 110 Jefferson, Thomas, 4, 13,
Gantt, Thomas, 73 Haugh, Walter, 88, 90 21, 35–​6
Garn-​St. Germain Depository Hayden, Warren, 150 Johnson, Hugh, 188
Institutions Act of Healy, William, 299 Johnson, Jed, 181–2, 304
1982, 237 Hearst, William Johnson, Percy, 172
George, Henry, 225 Randolph, 209
Gerry, Elbridge, 38 hedging, 89–​90, 316–​8, 329 Kahn, Alfred, 310
Gibson, John Bannister, 18 Heflin, James, 168 Kahn, Otto, 193
Gilchreest, Robert, 43 Heimann, Henry, 173 Kelly, D.F., 192
Gintel, Robert, 271–​2 Hemphill, Marie Kelly, Fred C., 289
Girard, Stephen, 282–​3 Antoinette, 282 Kendall, Waldo,
Girault, Francis, 245 high-​frequency trading, 306 192, 214
Glass, Carter, 165, 177, 181, Hildreth, Richard, 40 Kennedy, Joseph, 183, 190, 204
197, 208–​11, 213–​6 Hill, John, 106, 116 Kennedy, Ted, 310
Glass-​Steagall Act, 164, 205–​17 Hinmann, Elisha, 68 Kennedy, W. McNeil, 273
repeal of, 309, 320–​3 Hodges, George, 163 Kent, Fred, 215
Glover, Joseph, 287 Hoffman, Clare, 232 Kent, James, 10, 41, 49,
Glover, Lloyd, 254–​5 Hoffman, Murray, 69 249, 252
Gonzalez, Henry, 325 Holmes, Oliver Wendell, Key, Francis Scott, 246
Goodwin, Homer, 242–​5 102–​3, 201 Keynes, John Maynard, 304
Gordon, Thomas, 62 Holt, Daniel, 158 Keyser, Paul, 162
Gouge, William, 257 Home Owners’ Loan Act of Kimbrough, T.J., 112
Government National 1933, 234 King, Willford, 294
Mortgage Association, Hone, Philip, 23 King, William, 168
234–​5 Hooker, Stephen, 56–​9 Kosuga, Vincent, 229–​30, 232
Grain Futures Act of Hoover, Herbert, 166, 170–​8, Kress, Harry, 99
1922, 119, 137 185–​6, 187 Krock, Arthur, 200
Gramm, Phil, 308 Hopkins, Robert, 211
Gramm-​Leach-​Bliley Act of Housing and Economic La Follette, Robert, 168
1999, 320–​3 Recovery Act of 2008, 239 La Guardia, Fiorello, 181
Grantham, Jeremy, 314 Howard, Earl, 143 Laidlaw v Organ, 245–​7
Gray, Roger, 232 Howland, Harold, 112 Lamar, Joseph, 263
Great Moderation, 312–​3 Hubbard, George, 128 Lamont, Thomas, 166, 177
Greenbaum, Walter, 224 Hudnut, Alexander, 117 land, speculation in, 10–​1,
Greene, Asa, 19 Hughes, Charles Evans, 144 22–​3, 218–​40
Greenleaf, Moses, 10 Hume, J.F., 258 Landis, James, 188–​9, 201, 203
Greenspan, Alan, 326 Hunholz, John, 156 Landis, Kenesaw
Grund, Francis, 7 Hutcheon, Robert, 107 Mountain, 158
Guenther, Louis, 176 Hutchinson, Benjamin, 121 Lawrence, William Beach, 23
Laylin, Clarence, 264
Hagner, Alexander Burton, 72 Ingersoll, Charles, 32 Lee, Henry, 11
Haig, Robert Murray, 298 insider trading, 184, 241–​78 Lefèvre, Edwin, 94, 190
Haines, Charles Glidden, 9 insurable interest, 47–​9, 51 Leffingwell, Russell, 169, 191
Hamill, Charles, 114, 122–​3 intent-​to-​deliver rule, Leggett, William, 14
Hamilton, Alexander, 4, 20, 67–​83, 324–​5 Lenroot, Irvine, 167, 294–​5
29, 35–​7 Investment Bankers Leventritt, David, 144
Handmaker, Herman, 173 Association of America, Levine, Dennis, 277
334

334 Index

Levitt, Arthur, 326 Miller, A.C., 222 Perkins, W.R., 178


Lewisohn, Adolph, 142 Miller, Peter, 298 Perrin, Lee, 152
Lincoln, Abraham, 5–​7, 52 Moley, Raymond, 188 Perry, Jairus, 284
Lindstrom, Ralph, 210 Moore, Clement Clarke, 19 Phillips, Willard, 27, 38
life insurance, 49–​51 Moore, James, 291 Pike, Sumner, 268
Lippman, Walter, 172 Moorhouse, H.W., 166 Pointer, Joseph, 174
Livermore, Jesse, 94 Morgan, J.P., 166, 174 Pomeroy, Daniel, 182
Long Term Capital Morgenthau, Henry, 203 Populist Party, 105, 109
Management, 319, 326 Morris, Robert, 38, 40, 44 Porter, William, 214
Lord, Nancy, 51 Morrison, C.D., 261 Posner, Richard, 313
Loss, Louis, 267, 273 mortgage securitization, 221–​2, Potter, Alonzo, 39
Love, James, 77–​8 235–​40, 328 Powell, Lewis, 275–​6
Lowenstein, Roger, 314 Mortimer, Thomas, 256 Powell, Nathaniel, 307
Lucas, Robert, 312–​3 Motor Carrier Act of 1980, 311 Prentiss, John W., 193
Lujan, Jerry, 307 Mott, James, 182–​3 Procter & Gamble, 318
Muller, George, 128 “prudent man” rule, 279–​93
Macey, Jonathan, 322 Mulvey, Thomas, 147 Pujo, Arsène, 161
Maclure, William, 19 Purcell, Ganson, 203
Madison, James, 11 National Housing Act of Putnam, Samuel, 281
Maidenberg, H.J., 292 1934, 234
Manne, Henry, 277–​8 Neergaard, W.B., 192 Rainer, William, 326
Mapes, Carl, 195 Nelson, Murry, 121 Rayburn, Samuel, 195
margin trading, 178–​80, New York Stock Exchange Raymond, Daniel, 21, 257
194, 196–​8 campaign against bucket real estate taxation, 224–​6
marine insurance, 47–​9 shops of, 94–​103 recessions. See economic
Marryat, Frederick, 8 self-​regulation by, 91, 178 downturns
Marshall, John, 246 Newcomb, Simon, 53 Rede, Leman
Martin, John, 173 Noah, Mordecai, 34 Thomas, 15
Martin, William Norris, George, 119–​20, 136 Reed, Robert, 151
McChesney, 204 Norton, J.H., 121 Reiling, Herman, 297
Martineau, Harriet, 7 Noyes, Alexander, 171 Repide, Francisco, 261–​2
Marx, Groucho, 218–​9, 223 Rice, E.L., 193
Mason, Cyrus, 37 Obama, Barack, 323 Rickards, James, 309
Mason, George, 26 Oliver, William, 262 Ripley, William, 187–​8
Matthews, W.A., 212 onion futures, 229–​32 Roberts, George, 167
Maule, William Henry, 65 Onion Futures Act of 1958, Rochefoucault, Duke de la, 18
Mayfield, Earle, 181 229, 232 Rogers, Will, 175, 185
McAllister, William, 56, 59 Orange County Ronaldson, James, 14
McCain, Charles, 214 bankruptcy, 318 Roosevelt, Franklin, 179, 181, 186,
McCumber, Porter, 294 Organ, Hector, 245–​6 187–​9, 191, 193–​6, 201–3
McDermott, George, 118 Roosevelt, Theodore, 160
McFadden Act of 1927, 207 panics. See economic Roper, Daniel, 193
McKenna, Joseph, 153–​4 downturns Rossiter, Edgar, 210
McLean, Ann, 280–​1 Pantaleoni, Guido, 183 Rowen, Paul, 268
McLean, John, 280–​1 Park, James, 48 Rubin, Robert, 326
McNaughton, James, 242 Parke, James, 65 Rugg, Arthur Prentice, 243–​4
McVickar, John, 9, 21 Parker, Isaac, 11, 48 Ruml, Beardsley, 299
Melish, John, 13 Parsly, E.G., 191 Rumsey, Israel, 81–​2
Mellon, Andrew, 169–​70 Peck, Jacob, 47 Rush, Benjamin, 20
Merrill, Jonathan, 288 Pell, Herbert Claiborne, 181 Ryan, John, 119
Merritt, E.T., 147 Pelosi, Nancy, 307
Messmore, John, 120–​1 Percy, John, 87, 99 Sabath, Adolph, 176, 182, 195
Milken, Michael, 277 Perkins, Isaac, 42–​3 Sachs, Alexander, 189, 191
  335

Index 335

Sakolski, Aaron, 296 Stewart, Martha, 244 Waksal, Samuel, 244–​5


San Francisco Mining Stiglitz, Joseph, 305, 309 Walcott, Frederic, 176, 211
Exchange, 83 Stimson, Henry, 192 Walker, Roberts, 264
Schieren, Charles, 144 Stockbridge, Frank, 223 Wall, E.C., 124
Scollin, John Jr., 307 Stokes, Edward C., 169 Wallace, Henry, 188
Scott, Austin Wakeman, 288 Stone, Goldie, 301–​2 Walsh, David, 294
Scott, Charles, 175 Story, Joseph, 65–​6, 84, 251–​2 War of 1812, speculation
Secondary Mortgage Market Stout, Lynn, 309 during, 24–​5
Enhancement Act of Stowe, George, 108 Warren, William, 88, 98
1984, 235 Strong, James, 136 Washburn, William, 110,
Securities Act of 1933, Strong v. Repide, 261–​2 124, 129
164, 186–​94 Stubbs, Walter, 138, 147 Washington, George, 12,
Securities and Exchange Studley, Elmer, 195 35–​6
Commission, 183–​4, Sully, Daniel, 112 Waterman, Sterry, 274
186, 194, 197–​205, 239, Summers, Lawrence, 305, 326 Watson, Lester, 204
265, 267–​76 Surrey, Stanley, 297 Wayland, Francis, 28
Securities Exchange Webster, Pelatiah, 20
Act of 1934, Taft, William Howard, 260 Weems, Mason Locke, 16
164, 194–​8, 265–​6 Tague, Thomas, 307 Wells, F.B., 118
securities transaction tax, Tallmadge, S.W., 120 Wells, Frederick, 211
181–2, 280, 303–​6 Taylor, Edward, 162 Wharton, Francis, 76
Seligman, Eustace, 191, 193 Temple, Alan, 165 Whitaker, William, 253–​4
Seligman, Joel, 188 Texas Gulf Sulphur case, 273–​5 White, Compton, 176
Seltzer, Lawrence, 299 Thompson, Huston, 183 White, Horace, 144
Sergeant, John, 33 Thompson, James, 71 White, S.V., 120
Seward, William, 27 Tilghman, William, 11, 12, 22 Whitney, Richard, 170, 175,
Shattuck, Mayo Adams, Tincher, Jasper, 111, 135 177, 201, 204
288, 290 Tobin, James, 304 Whittaker, John, 114
Sherman, John, 7, 52–​3 Tocqueville, Alexis de, 8 Wilgus, H.L., 264
short selling, 113–​4, 174–​8, 194, Treanor, Jim, 268 Wilkes, John, 43
197–​9, 202, 226–​7 Truax, Charles, 172 Williams, Clark, 144
Siegel, Sam, 229–​30, 232 Tucker, R.S., 295 Williams,
Smith, Martin, 180 Tugwell, Rexford, 188 Egerton, 113
Smith, Rollin, 119–​20 Williams, John Skelton, 207
Smith, William Stark, 213 United Nations, 309 Willis, Henry Parker, 172,
Snow, Alviras, 261 Untermyer, Samuel, 188, 209 208
Soros, George, 325 Winter, Keyes, 160
South Sea Bubble, 62, 71, 170 van Antwerp, W.C., 192 Woodin, William, 189
Spain, Louis Jr., 307 Verplanck, Gulian, 29, 247–​9 Wright, Nathan, 42–​3
Spence, Joseph, 253–​4 Vilas, William, 107–​9,
Staggers Rail Act of 1980, 312 120, 122–​4 Yeargen, Charles, 307
Steagall, Henry Bascom, 208 Volcker, Paul, 323 Yeates, Jasper, 48
Stevens, Albert, 132 Vroom, Peter, 282 Youngman, Elmer, 223
336
  337
338
  339
340

You might also like