Nobel Lecture: The Process and Progress of Economics

Author(s): George J. Stigler

Source: Journal of Political Economy , Aug., 1983, Vol. 91, No. 4 (Aug., 1983), pp. 529-
545
Published by: The University of Chicago Press

Stable URL: http://www.jstor.com/stable/1831067

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend
access to Journal of Political Economy

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
Nobel Lecture: The Process
and Progress of Economics

George J. Stigler
University (ft( Chicago

The lecture focuses on the reasons that new ideas are accepted or
rejected by a science. A distinction is drawn between prescientific
and scientific stages of a discipline. The diverse fates of new ideas are
illustrated by a variety of episodes in the history of economics, in-
cluding the economics of information and the theory of economic
regulation.

In the work on the economics of information which I began 20 some

years ago, I started with an example: How does one find the seller of
automobiles who is offering a given model at the lowest price? Does it
pay to search more, the more frequently one purchases an auto-
mobile, and does it ever pay to search out a large number of potential
sellers? The study of the search for trading partners and prices and
qualities has now been deepened and widened by the work of scores
of skilled economic theorists.
I propose on this occasion to address the same kinds of questions to
an entirely different market: the market for new ideas in economic
science. Most economists enter this market in new ideas, let me em-
phasize, in order to obtain ideas and methods for the applications
they are making of economics to the thousand problems with which
they are occupied: these economists are not the suppliers of new ideas
but only demanders. Their problem is comparable to that of the
automobile buyer: to find a reliable vehicle. Indeed, they usually end
up by buying a used, and therefore tested, idea.

Nobel Lecture presented December 8, 1982, Stockholm. I wish to thank Gary Becker,
Aaron Director, Milton Friedman, and Stephen Stigler for helpful comments.

[Journal of Politicanl EconoinX, 1983, vol. 91, no. 41

? 1982 bv TIhe Nob~el Fo(nldatioOI.

529

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
530 JOURNAL OF POLITICAL ECONOMY

Those economists who seek to engage in research on the new ideas

of the science to refute or confirm or develop or displace them are
in a sense both buyers and sellers of new ideas. They seek to develop
new ideas and persuade the science to accept them, but they also are
following clues and promises and explorations in the current or pre-
ceding ideas of the science. It is very costly to enter this market: it
takes a good deal of time and thought to explore a new idea far
enough to discover its promise or its lack of promise. The history of
economics, and I assume of every science, is strewn with costly errors:
of ideas, so to speak, that wouldn't run far or carry many passengers.
How have economists dealt with this problem? That is my subject.
I begin by distinguishing the prescientific stage of a discipline from
its scientific stage. A science is an integrated body of knowledge, and
it is pursued and developed by a group of interacting practitioners
called scientists. The validation and extension of that body of knowl-
edge is the intellectual goal of the scientists, although of course the
pursuit of that goal in turn serves whatever personal goals such as
prestige, reputation, and income-the scientists seek. These are only
definitions, but I hope they are not strained or unnatural ones.
The prescientific stage is characterized in part by the incom-
pleteness of the body of knowledge, but that is only a relative matter
since no science is ever complete. This prescientific stage is also char-
acterized by absence of a set of interacting practitioners who are de-
voting a large part of their lives to the accumulation of knowledge,
and hence it is characterized by the absence of cumulative progress.

I. Prescientific Economics: Mercantilism

We will find it useful to spend a short time with the large body of
writing called mercantilism. This literature ranges over several cen-
turies, and over England and western Europe. The literature com-
prises hundreds of pamphlets and books and includes participants of
the stature of John Locke and William Petty. I must confess at once
that I have little direct knowledge of that literature, for I have concen-
trated my historical work upon the period which followed. However,
three major studies of mercantilism are reassuringly agreed upon the
characteristics I wish to discuss. The studies are Edgar Furniss's book,
The Position of the Laborer in a System of Nationalism (1920), Jacob Vin
er's famous essay, "English Theories of Foreign Trade before Adam
Smith" (1930), and Eli Heckscher's masterly treatise, Mercantilism
(1935).
A first characteristic of all three surveys of mercantilism is that they
almost totally lack a time dimension. Furniss will document a state-
ment by references to two tracts written more than a century apart.

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
 NOBEL LECTURE 531

With the very first doctrine of mercantilism-that it was vitally impor-

tant to have an excess of exports over imports-Viner begins a se-
quence of illustrative quotations with Richard Leicester, who wrote in
1381. (Of course, if one were allowed to go out of economics it would
be easy to continue the sequence of praises of an export balance a full
6 centuries through 1981 and probably another 6 centuries through
the year 2581!) Heckscher also seldom finds it necessary to notice the
temporal sequence of two writers.
A second characteristic is that most mercantilists propose their own
views without any attempt to utilize or improve upon the work of
other mercantilists. There were sharp controversies, of course, but no
regular pattern of sequences of criticisms and responses. These writ-
ings, one may note, were almost always briefs for special interests.
The third characteristic is almost a corollary of the first two: there
was no cumulative improvement in the doctrines being propounded.
I quote Viner:

In many respects, indeed, as the mercantilist argument

became more elaborate and involved, it became more objec-
tionable from the point of view of modern doctrine, and,
except with reference to the bullionist doctrines, a strong
argument could be presented in defense of the thesis that
the mass of ordinary tracts on trade of the first half of the
eighteenth century showed a more extreme and confused
adherence to the fallacies of mercantilism than did the writ-
ings of the sixteenth and early seventeenth centuries.... In
so far as trade theory was concerned, such progress as oc-
curred was due almost solely to a small group of capable
writers, able to analyze economic problems more acutely and
logically than their predecessors, but not able to make a
marked impression upon their contemporaries or even to
attract their attention. [Viner 1937, p. 109]

The process of analysis simply was not cumulative: there was little
advantage in studying foreign trade if one were born in 1680 instead
of 1580.
I am now prepared to come to the rescue of an economist who
needs little rescuing: Adam Smith. A considerable number of econo-
mists, and a few considerable economists, have emphasized the fact
that Smith had many gifted predecessors and almost all or perhaps
exactly all of his ideas are to be found expressed, and sometimes well
expressed, by these predecessors. Some economists therefore wish to
give the title of founder of economics to earlier writers such as Cantil-
lon. This line of argument, in my view, misses the point.

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
532 JOURNAL OF POLITICAL ECONOMY

It was Smith who provided so broad and authoritative an account of

the known economic doctrine that henceforth it was no longer per-
missible for any subsequent writer on economics to advance his own
ideas while ignoring the state of general knowledge. A science consists
of' interacting practitioners, and henceforth no one could decently
ignore Smith's own work and in due time the work of Malthus,
Ricardo, and the galaxy of economists who populated the first half' of
the nineteenth century.
The change came fast. Smith himself did not interact with any
writers on economics after 1776, and of course even in his treatise he
coolly ignored his leading rival, Sir james Steuart. Five years after the
first edition of the Essay on Population (1798), by contrast, Malthus was
making fundamental concessions in response to Godwin and other
critics. The age of economic science had begun.
How complete the transformation of economics has become may be
illustrated by an episode earlier in this century. A. C. Pigou, holding
the chair in economics that his predecessor, Alfred Marshall, had
made the most prestigious in .the world, committed an error in stating
the theory of external diseconomies. He asserted that when a firm
contemplated entry into a competitive industry which is subject to
rising supply prices of its inputs, that firm would make a socially
inefficient decision because it would ignore the effect of its entrance
into the industry in raising the prices other firms would have to pay
for inputs. The error involves a confusion of transfer payments with
social costs. The error appeared in his famous treatise, Wealth and
Welfare, in 1912.
Allowing for the distractions created by the First World War, major
economists soon devoted themselves to the problem. The two most
famous refutations, by Dennis Robertson and Frank Knight, came in
1924, but the essential point had been made earlier by J. M. Clark
(1913) and Allyn A. Young (1913). Under these attacks even Pigou,
the most remote of scholars, capitulated. The era had already begun
when only the detected errors of unimportant economists are spared
a prompt reputation.

II. Economic Science: The Environmental View

The politics and economics of mercantilistic policy were the determi-

nants of the issues in the mercantilist literature. Indeed the prescien-
tific age of any discipline is dominated by the practical concerns of the
society in which it is cultivated. It is an easy step to the view that the
main problems of a discipline, even after it becomes an organized
science, are posed directly by the paramount problems and policies of
the society in which it is pursued.

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
 NOBEL LECTURE 533

Wesley Clair Mitchell went so far as to attempt to present a system-

atic history of economic thought in terms of the responses of each
generation to its environment: "One of the results of any survey of
the development of economic doctrines is to show that in large mea-
sure the important departures in economic theory have been intellec-
tual responses to changing current problems; that is, the economic
theorists who have counted most in the development of thought have
been men who have been deeply concerned with problems that trou-
bled their generation" (Mitchell 1967, 1:13). As examples, he told us:
"Malthus's problem of population was as obviously an intellectual
reflection of current events as was Adam Smith's 'obvious and simple
system of natural liberty' " (1:235). "The description of the course of
English politics in Parliament shows that Ricardo got this problem
[how to determine the way in which the produce of the country was
divided]-his appreciation of its importance-not in his study, but by
following current events. It should also be noticed that Ricardo got his
peculiar conception of what the problem of distribution is directly
from the Parliamentary struggle" (1:286). Yet when Mitchell reached
the 1870s and the rise of the marginal utility theory, he abandoned
the attempt to find environmental changes to which economic theory
was responding. He attributed the abandonment to the difficulty of
achieving understanding of and detachment toward more recent
work, not the failure of his hypothesis (Mitchell 1967, 2:2).
The central task of an empirical science such as economics is to
provide general understanding of events in the real world, and ulti-
mately all of its theories and techniques must be instrumental to that
task. That is very different from saying, however, that it must be
responsive to the contemporaneous conditions and problems of the
society in which it is situated.
If the problems of economic life changed frequently and radically
and lacked a large measure of continuity in their essential nature,
there could not be a science of economics. An essential element of a
science is the cumulative growth of knowledge, and that cumulative
character could not arise if each generation of economists faced fun-
damentally new problems calling for entirely new methods of analy-
sis. The change of problems and methods would also undermine the
training of economists: if the young studied under the old, the young
could be confident that they were learning things that were rapidly
becoming obsolete. A science requires for its very existence a set of
fundamental and durable problems.
In economics the most fundamental of these central problems is the
theory of value. The theory of value must explain how the compara-
tive values of different goods and services are established. Until that
problem is solved, it is not possible to analyze for scientific purposes

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
534 TOURNAL OF POLITICAL ECONOMY

what will be produced and in what quantities, how the resources will
be employed in producing the menu of outputs, and how the re-
sources will be valued. Without a theory of value the economist can
have no theory of international trade nor, possibly, a theory of
money. This central problem of value does not change in its essential
content if one seeks to explain values in rural or urban societies, or in
agricultural or industrial societies. Indeed, if the problem of value
were so chameleon-like as to alter its nature whenever the economic
or political system altered, each epoch in economic life would require
its own theory, and short epochs would get short-lived theories.
If an empirical science requires for its very existence a set of funda-
mental and persistent phenomena, that is not the only kind of phe-
nomena with which it will deal. It will continuously be confronted
with new circumstances which call for more than a routine application
of standard knowledge. Thus the energy crisis of the 1970s has pro-
vided much employment to economists, but it has not called for im-
portant changes in economic science.
An empirical science has a second, and vastly more important, in-
terest in and responsiveness to contemporary problems: its received
theory will at times be incapable of dealing with these problems.
When England began the long-term importation of grain at the time
of the Napoleonic wars and pressed hard upon its domestic produc-
tion capacities, the economists introduced the law of diminishing re-
turns in dealing with the price of grain. It would be difficult to deny a
role to the environment in the appearance of this law. So much for
the origin of that theory: it would not help us one whit in understand-
ing Edgeworth's famous analysis of this law in 1911 to look at his
economic environment. The important place that diminishing re-
turns has achieved in economics is due precisely to the fact that its
usefulness was not limited to Ricardo's analysis of agriculture in Great
Britain.
The responsiveness of economics to environmental problems will
naturally be more complete and more prompt, the more urgent the
problems of the day. The response will also be more complete, the less
developed the relevant body of economic analysis. The respon-
siveness of macroeconomics to contemporary events is notorious.
Keynes's conquest in the 1930s was due to the fact that the neoclas-
sical theory could not account for the persistent unemployment of
that decade. A generation later, persistent inflation even with less
than full employment was equally decisive in ending Keynes's su-
premacy. If and when macroeconomics produces a good theory of the
business cycle, its responsiveness to environmental changes will di-
minish sharply.
A viable and healthy science requires both the persistent and almost

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
 NOBEL LECTURE 535

timeless theories that naturally ignore the changing conditions of

their society and the unsettled theories that encounter much difficulty
in attempting to explain current events. Without the base of persis-
tent theory, there would be no body of slowly evolving knowledge to
constitute the science. Without the challenges of unsolved, important
problems, the science would become sterile.
One final observation: there is no simple or known relationship
between environmental changes and changes in economic analysis.
During the Industrial Revolution, economists adopted the law of di-
minishing returns but ignored the most sustained and widespread
growth of output that the world had yet observed. The vast govern-
mental income redistribution programs of the last hundred years
have only recently attracted the attention of economic theorists. The
scholars who create economic theory do not read the newspapers
regularly or carefully during working hours.

III. The Omniscient Scholar?

Once a science becomes well populated, has achieved a secure aca-

demic base, and is equipped with the machinery of intellectual ex-
change-journals, learned societies, and conferences-it is presented
with a stream of proposals for new directions or new methods for
research. Indeed, the science itself carefully fosters the output of new
ideas. Robert K. Merton has shown in his fundamental studies of the
reward structure of science that immense value is attached to priority
in the development of successful new ideas (Merton 1973, esp. chap.
14).
And yet ideas will be proposed which are ignored at the time but at
some later date are accepted (almost invariably after an independent
rediscovery) as important to the science. This phenomenon re-
peatedly called forth the rebukes of Schumpeter in his great History of
Economic Analysis. Here are examples of men who, Schumpeter be-
lieved, quite correctly, were "writing above their time": "Longfield's
merits may be summed up by saying that he overhauled the whole of
economic theory and produced a system that would have stood up
well in 1890" (Schumpeter 1954, p. 465). John Stuart Mill "even went
so far as to compare [John] Rae's performance on accumulation with
Malthus' performance on population. And all this, written in what
was to be for forty years the most influential textbook of economics,
was insufficient to introduce Rae to the profession or to rouse any
curiosity concerning the rest of his book!" (p. 469). Of course
Schumpeter, than whom no economist was more sophisticated, gave
some sensible reasons for these acts of neglect of genius, but he failed
to give the most important reason of all.

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
536 JOURNAL OF POLITICAL ECONOMY

In every period of the active pursuit of a science, new ideas are

continually being proposed. Any new idea a new conceptualization
of an existing problem, a new methodology, or the investigation of a
new area-cannot be fully mastered, developed into the stage of a
tentatively acceptable hypothesis, and possibly exposed to some em-
pirical tests without a large expenditure of time, intelligence, and
research resources. That is fact 1. Fact 2 is that the overwhelming
majority of these new ideas will prove to be sterile-in fact, quite
possibly all the new ideas of a period of years will prove to be sterile.
Only afterward, with the fullness of knowledge that history some-
times provides, can we identify the truly fertile ideas of a period.
Some men have superb instincts as to which of the new ideas of the
time will repay intensive exploration, but no one is infallible. Even the
greatest of economists pursue some problems that take them no-
where. In the last months of his life, Ricardo was still attempting to
fashion a precise measure of value and not advancing one inch. John
Stuart Mill and Leon Walras devoted much energy to the propagation
of the proposal of nationalization of unanticipated future increments
of land values-not the first time or the last that someone proposed
nationalizing a sum with an expected value no larger than zero. je-
vons could not get over the idea that cycles in sunspots left their tracks
on commercial cycles. The great Pareto took a detour through the
question of the order in which people consumed various products,
out of a belief that this was related to the order of integration of a
partial differential equation.
Not only great economists, but all economists who pursue anything,
pursue will-o'-the-wisps for periods of time that are painful to con-
sider in retrospect. In the 1930s, the area variously known as indus-
trial organization and microeconomics-with-evidence was offered the
following major research hypotheses:
1. The ownership and the effective control of large corporations
have become separated.
2. The phenomenon of product differentiation calls for funda-
mental changes in the theory of the firm and the industry (the theory
of monopolistic competition).
3. Prices do not respond downward to changes in supply and de-
mand, perhaps because a particular expectation with respect to rivals'
behavior creates a kink in the firm's demand curve.
4. The economist is able to construct criteria of the satisfactory or,
alternatively, the unsatisfactory performance of an industry, where
the satisfaction of the economist should be shared by society (the
theory of workable competition).
These were not the only new research proposals: the annual output of'
new theories of oligopoly was supplemented by searches for truth
through the feeding and wining of business leaders.

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
 NOBEL LECTURE 537

Each of the four research proposals I have listed received a good

deal of attention: none lost- its fashionable appeal to at least some
highly competent economists for at least 5 or 10 years, and indeed not
one is a cold corpse today. But it is also true that not one of them has
been absorbed into the mainstream of price theory as a regular and
significant part of the analysis of the workings of markets and indus-
tries. Quite possibly one could find that Schumpeter followed several
of these detours for at least a short distance. Of course at the same
time some important new ideas (such as that of Hotelling on exhaust-
ible resources and Ramsey on optimal pricing) were being neglected.
To err is not only human but also scientific.

IV. The Continuity of Scientific Change

"Nature does not move in jumps," says the proverb, and a science also
progresses through time without making large jumps. This continuity
is often illustrated by two kinds of evidence.
One evidence of scientific continuity that has been adduced by
Robert Merton is the existence of multiple and nearly simultaneous
independent discoveries of a theory by several scientists. The popular
examples in economics are the discovery of the theory of rent by
Edward West and Thomas Robert Malthus in 1815 and the publica-
tion of the theory of utility in the early 1870s by Jevons, Menger, and
Walras. In each case, the new idea was presumably appropriate to the
development of economics at the time: the rent theory allowed the
construction of a theory of the distribution of income; and the utility
theory led naturally to the marginal productivity theory and the
generalization of the theory of utility-maximizing behavior.'
This continuity is also used to explain the not uncommon phenom-
enon of the failure of a man of genius to get acceptance of his ideas
from his contemporaries, even though later generations will applaud
the performance. Augustin Cournot, for example, was an important
scholar in one of the leading intellectual centers of Europe, but he
could not persuade economists in 1838 that the mathematical theory
of maxima and minima was a useful tool for economic analysis.
I would find it more persuasive to establish the continuity of
scientific development by a close examination of the evolution of im-
portant concepts in economics, but that route does not seem appro-
priate to the occasion.2 Candor compels me to note that the route of
close historical study would not be easy to follow because it would

l I have presented elsewhere an alternative interpretation of Merton's theory of

multiple discoveries which emphasized even more than he does how essential it is that
science be "ready" for a new idea (see Stigler 1980).
2 For a fascinating case study in another discipline, see Fisher (1982).

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
538 JOURNAL OF POLITICAL ECONOMY

require definite answers to the questions: What is a large change in a

science? What is a rapid change in a science?
Gary Becker has suggested that a substantial resistance to the accep-
tance of new ideas by scientists can be explained by two familiar
economic concepts. One is the concept of specific human capital: the
established scholar possesses a valuable capital asset in his command
over a particular body of knowledge. That capital would be reduced if
his knowledge were made obsolete by the general acceptance of a new
theory. Hence, established scholars should, in their own self-interest,
attack new theories, possibly even more than they do in the absence of
joint action. The second concept is risk aversion, which leads young
scholars to prefer mastery of established theories to seeking radically
different theories. Scientific innovators, like adventurers in general,
are probably not averse to risk, but for the mass of scholars in a
discipline, risk aversion is a strong basis for scientific conservatism.
We will find the specific human capital theory illustrated in the epi-
sodes to which I shall soon turn.
No one can describe the precise characteristics or content of a new
piece of scientific work that will find ready and eager reception from
the scientists of a period. Indeed, if knowledge sufficient to identify
the theories that will succeed were possessed, it would be of immense
value in finding and developing those theories and therefore would
be the key to scientific fame. To the scientist such knowledge would be
much more valuable than an accurate method of predicting stock
prices! Even without such a priceless key to the understanding of
scientific innovation, it is interesting to examine several routes by
which a scientific idea makes its way into the work of economists. I
illustrate two of these routes by subjects on which I have worked.

Acceptance without Struggle: The Economics of Information

Economists have always known that the extent and accuracy of the
knowledge of the economic actor had influence, and often a decisive
influence, on his behavior and therefore on the behavior of markets.
One striking example of this critical role of information is provided
by the theory of oligopoly. The first formulation of the problem of
oligopoly as a specific problem in economic theory was made by Cour-
not, whose long failure to get acceptance I have already mentioned. It
was essential, in explaining how each of two rivals in a market would
behave, to attribute to each some belief about the behavioral pattern
of the other. Cournot made the assumption that each assumed that
the rival did nothing in response to his own actions. The later theories
of oligopoly all rest upon different assumptions concerning patterns

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
 NOBEL LECTURE 539

of behavior which each seller attributes to his rivals. A dozen other

areas of economic analysis, such as the workings of the labor market
and the role of advertising, also rest squarely on assumptions about
information of the economic actors. In this tradition, the amount of
information possessed by individuals in any market was arbitrarily
postulated rather than derived from economic principles. The con-
sensus was that consumers knew little, traders on organized ex-
changes a great deal; investors were either gullible or omniscient.
Even the powerful and luminous essay by Friedrich von Hayek on
"The Use of Knowledge in Society" (1945) had not addressed the
principles of acquisition of knowledge.
I proposed (in 1961) the use of the standard economic theory of
utility-maximizing behavior to determine how much information peo-
ple would acquire with special attention to the prices at which they
would buy and sell, and a year later made an application of the analy-
sis to labor markets. There is one interesting feature of the subse-
quent history of the reception of this work by economists to which I
wish to call attention.
The proposal to study the economics of information was promptly
and widely accepted, and without even a respectable minimum of
controversy. Within a decade and a half, the literature had become so
extensive and the theorists working in the field so prominent that the
subject was given a separate classification in the Index of Economic
Articles, and more than a hundred articles a year are now devoted to
this subject.
The absence of controversy certainly was no tribute to the defin-
itiveness of my exposition. I had chosen fixed sample rather than
sequential analysis, which a majority of later economists prefer. I had
not presented a general equilibrium solution in which the behavior of
both sides of a market is analyzed, and that step proved difficult to
take. I had done little with information on quality and other variables,
in contrast to price, although I soon extended the approach to a
different kind of information in the theory of oligopoly. I had not
applied the theory to the problem of unemployment, a literature
initiated by an important paper by Armen Alchian (1970). All I had
done was to open a door to a room that contained many fascinating
and important problems.
The absence of controversy was due instead to the fact that no
established scientific theory was being challenged by this work: in fact,
all I was challenging was the neglect of a promising subject. More-
over, the economics of information was susceptible to study by quite
standard techniques of economic analysis. The theory immediately
yielded results which were intuitively or observationally plausible.
Here was a Chicago theory that didn't even annoy socialists!

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
540 JOURNAL OF POLITICAL ECONOMY

Acceptance by Necessity: The Economics of Regulation

The work on the economics of regulation has entered economics by a

different route.
The modern era of economists' interest in the economic workings
of the state may be dated from the influential work of Anthony
Downs, An Economic Theory of Democracy (1957), and James Buchanan
and Gordon Tullock, The Calculus of Consent (1962). Although I had
read these works with deep interest and admiration, my own work on
regulation at first followed a different, more empirical route.
An examination of the economic literature had revealed no serious
professional attempt to measure the impact of public regulation in
areas with long histories: the regulation of rates of electrical utilities;
the review of new issues by the Securities and Exchange Commission;
and the antitrust policy of the United States. The investigations of
these problems (with the indispensable assistance of Claire Fried-
land), strongly reinforced by related work of colleagues and students,
gradually forced me to confront a question that should have been
obtrusively obvious at once: Why does the state engage in its regula-
tory activities?
The answer (at least for an economist) seemed to lie much less in
the theorems of welfare economics or the prescriptions of traditional
political science than in the systematic examination of the self-interest
of the various participants in political life. These participants, to be
sure, operated under different rules and constraints than the traders
in markets, but that did not argue against using that powerful tool of
economic analysis, the theory of utility-maximizing behavior. Once
the economist can identify the costs and returns from various actions,
this theory allows him to make predictions of behavior that have been
reasonably successful.
This approach proved to be highly uncongenial to many econo-
mists. My teacher, Frank Knight, had often expressed the belief that
many economists still share, that the actors (and especially the voters)
in political life are ignorant, emotional, and usually irrational. In a
famous, unpublished speech he ended a parable with the words:
"Truth in society is like strychnine in the individual body, medicinal
in special conditions and minute doses; otherwise and in general, a
deadly poison." These economists believe that voters are myopic and
forgetful, and that political institutions are designed or perverted to
allow the public servants to pursue chiefly their own interests. An-
other and perhaps larger group of economists are critical of the util-
ity-maximizing approach for the opposite reason: that it appears to be
an attack on the chief instrument for purposive social improvement
that a society possesses: the state.
Nevertheless the economic theory of regulation is achieving a sub-

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
 NOBEL LECTURE 541

stantial scientific prosperity. Its findings with respect to both the

operation and the origins of regulatory policies directed to particular
industries (such as the securities markets, transportation, and occupa-
tional licensing) command a substantial support. To be sure, the ex-
planatory triumphs have not been overwhelming, and indeed the
theory itself is still relatively primitive. The main reason for the con-
siderable acceptance of the approach is that fundamental rule of
scientific combat: it takes a theory to beat a theory. No amount of
skepticism about the fertility of a theory can deter its use unless the
skeptic can point to another route by which the scientific problem of
regulation can be studied successfully.
There is an interesting asymmetry in the success of this literature in
dealing with the two problems into which the theory is commonly
divided: Why are regulatory policies adopted and abandoned? What
are their effects? Economists have been much more successful in
measuring effects of policies than in explaining their adoption. The
explanation is that one can choose the effects of a policy to study, and
usually more easily measured effects are chosen for study. One has no
such options when addressed with the question, Why did the United
States adopt an antitrust policy in 1890?
Thus studies of effects of regulatory policies have usually been
concerned with the effects upon prices and outputs, although the
effects desired by the supporters of these policies have probably been
upon the distribution of income. The panoply of regulatory measures
can be used to effect vast income redistributions, and these redistribu-
tions of income do not appear explicitly in the budget of the state.
The frequent exclusion of new entrants from a field, for example,
leads to smaller outputs, higher prices, and higher profits for the
protected enterprises and allows these benefits to increase with the
growth of the protected area. If these income transfers are as large as
fragmentary evidence suggests, the theory of regulation may well
become a full partner of tax and expenditure theory in public finance.

Acceptance by Trial by Combat?

Is it exceptional of the theories I have been discussing that neither

was subjected to direct trial by combat with an alternative theory? We
speak so often of the competition of ideas. How is that competition
conducted?
The direct confrontation of two alternative theories, each seeking
to explain the same body of observable phenomena, is not common in
economics.3 (It is perhaps encountered more often in macroeconom-

' I once made such a direct confrontation of the theory of the kinked oligopoly
demand curve and more traditional theories, finding no evidence to support the exis-

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
;42 IOURNAL OF POLITICAL ECONOMY

ics than in microeconomics.) Two modern examples from microeco-

nomics will illustrate the proposition that economists seldom choose
between directly rival theories on the basis of critical empirical tests.
1. The doctrine of limit pricing by oligopolists asserted that the
firms in an industry would set prices at such a level as to discourage or
prevent entry of additional firms into an industry. The theory had a
long prehistory under the name of potential competition, but it was
given an explicit formulation by Sylos-Labini (1962), Joe Bain (1949),
and Franco Modigliani (1958). This version gave rise to a substantial
literature, but at no time was a direct empirical test made of this
theory as against explicit alternative theories of oligopoly behavior.
2. The Pigovian theory of external economies was challenged di-
rectly by Ronald Coase (1960), who in effect argued that the Pigovian
theory had assumed noneconomic behavior on the part of the eco-
nomic actors in a wide class of phenomena. This challenge was met
for a time by a considerable number of counterarguments, but these
arguments were addressed to the logic of what has come to be known
as the Coase theorem. No explicit comparison of the explanatory
powers of the Coasian and Pigovian approaches has been undertaken.
Why did not the profession seek directly to test these theories and,
for that matter, the four theories of the 1930s that I characterized as
largely unsuccessful innovations? Some part of the answer may lie in
the fact that formal empirical tests of economic theories have histori-
cally been scarce, although they are increasing in frequency, but I
would not press this answer. Instead, the testing procedure-the trial
by combat-takes a different form.
It is seldom that a theory in economics has a well-defined domain of
applicability. It may have been created to explain a specific class of
events-the pricing by oligopolists when entry is possible, in the first
illustration above-but it always has a wider domain of possible ap-
plicability. The specification of a critical test which, if conducted cor-
rectly on a sufficient scale, will decide the combat between two alter-
native theories is seldom possible over the whole range of the domains
of the two theories.
Economists have therefore generally chosen to decide between the
alternative theories by the process of using each to explore a variety of
problems. How does the limit theory of oligopoly pricing, for ex-
ample, handle the process of growth of an industry or the phenome-
non of vertical integration? How does the Coasian theory illuminate
the structure of the law of torts or the economics of professional
sports? These explorations are a form of testing of the theories: they

tence of a kink. The theory has disappeared from professional work but lives in every
textbook (see Stigler 1978).

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
 NOBEL LECTURE 543

test the fertility of the theories (or at least the intellectual fertility of
economists), and the varied applications are partial empirical tests of
the theories. Gradually a consensus emerges among the economists
working on the subject: the theory becomes a part of the standard
analytical corpus or it dies of neglect.

V. Conclusion

Our list of factors which influence the receptivity of a science to new

ideas could easily be extended.
In particular, it would be useful to examine the question of whether
the attractiveness of the public policy positions associated with a
theory has an effect upon the acceptability of the theory. The text-
books on methodology lecture us on the need to separate positive and
normative theories. The study of economics tells us that few if any
theories lead unequivocally to one set of policy implications. So sci-
ence and policy should be separated. Are they? I believe that the
separation has been far from complete, especially in the short run,
but this is not the occasion to undertake the substantial study neces-
sary to support the belief.
Again, the institutional organization of economic research is a po-
tential influence upon the receptiveness of a science to new ideas. The
powerful institutional position of Schmoller and the German Histor-
ical School no doubt played a role in the slow development of eco-
nomic science in Germany after 1870. The dominant role of Cam-
bridge University in economics from Marshall to Keynes surely was
not favorable to the receptiveness of new ideas from outsiders. I
believe that the shift of the center of economics to the United States
was due in some part to the failure of the English economists to share
fully in the quantitative empirical study of economics.
Even if I extended this list of potential determinants of scientific
choice and documented each more fully than I have, I would still have
kept my promise not to tell you the detailed characteristics of the
successful new theories in economic science. I do not lament this
failure.
The fascination of scientific work does not lie in the craftsman-like
utilization of the tools of a science. It is admirable for the gymnast to
put his splendidly disciplined body through intricate maneuvers, and
it is no doubt equally admirable for the scientist to put his disciplined
mind through a sequence of complex analytical or experimental ma-
neuvers. The great fascination of scientific endeavor, however, is pre-
cisely in the speculative pursuit of new ideas that will widen the hori-
zon of our understanding of the world. This endeavor is not that of a
graceful, intellectual gymnast: on the contrary, the scientist is stum-

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
544 JOURNAL OF POLITICAL ECONOMY

bling about in a jungle of ideas or facts that seem to defy system or

logic, and usually he fails to emerge with anything but scratches. The
dangers of the search include the chance that a gifted rival will reach
the goal, and the danger is not reduced by the fact that the rivalry is
conducted under what for able and ambitious competitors are un-
usually chivalrous rules. Still, learning more about how this search for
new knowledge proceeds is itself a worthy search for new knowledge,
and we shall not abandon it.

References

Alchian, Armen. "Information Costs, Pricing, and Resource Unemploy-

ment." In Microeconomic Foundations of Employment and Inflation
edited by Edmund S. Phelps et al. New York: Norton, 1970.
Bain, Joe S. "Pricing in Monopoly and Oligopoly." A.E.R. 39 (March 1949):
448-64.
Buchanan, James M., and Tullock, Gordon. The Calculus of Consent: Logical
Foundations of Constitutional Democracy. Ann Arbor: Univ. Michigan Press,
1962.
Clark, J. M. Review of Wealth and Welfare. A.E.R. 3 (September 1913): 623-
25.
Coase, Ronald H. "The Problem of Social Cost."J. Law and Econ. 3 (October
1960): 1-44.
Downs, Anthony. An Economic Theory of Democracy. New York: Harper, 1957.
Fisher, Nicholas. "Avogadro, the Chemists, and Historians of Chemistry,
Parts I and II." Hist. Sci. 20 (June 1982): 77-102; 20 (September 1982):
212-31.
Furniss, Edgar S. The Position of the Laborer in a System of Nationalism: A Study in
the Labor Theories of the Later English Mercantilists. Boston: Houghton Mifflin,
1920.
Hayek, Friedrich A. von. "The Use of Knowledge in Society." AE.R. 35
(September 1945): 519-30.
Heckscher, Eli F. Mercantilism. London: Allen & Unwin, 1935.
Knight, Frank H. "Some Fallacies in the Interpretation of Social Cost." QJ.E.
38 (August 1924): 582-606. Reprinted in Readings in Price Theory. Home-
wood, Ill.: Irwin (for American Econ. Assoc.), 1952.
Malthus, Thomas. An Essay on Population. 1798.
Merton, Robert K. The Sociology of Science: Theoretical and Empirical Investiga-
tion. Chicago: Univ. Chicago Press, 1973.
Mitchell, Wesley Clair. Types of Economic Theory: From Mercantilism to In-
stitutionalism. 2 vols. Edited by Joseph Dorfman. New York: Kelley, 1967.
Modigliani, Franco. "New Developments on the Oligopoly Front." J.P.E. 66
(June 1958): 215-32.
Pigou, A. C. Wealth and Welfare. London: Macmillan, 1912.
Robertson, Dennis H. "Those Empty Boxes." Econ. J. 34 (March 1924): 16-
31. Reprinted in Readings in Price Theory. Homewood, Ill.: Irwin (for Amer-
ican Econ. Assoc.), 1952.
Schumpeter, Joseph A. History of Economic Analysis. New York: Oxford Univ.
Press, 1954.
Stigler, George J. "The Literature of Economics: The Case of the Kinked
Oligopoly Demand Curve." Econ. Inquiry 16 (April 1978): 185-204. Re-

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms
 NOBEL LECTURE 545

printed in The Economist as Preacher, and Other Essays. Chicago: Univ.

Chicago Press, 1982.
. "Merton on Multiples, Denied and Affirmed." In Science and Social
Structure. A Festschrift for Robert K. Merton, edited by Thomas L. Gieryn.
New York: Academy of Science, 1980. Reprinted in The Economist as
Preacher, and Other Essays. Chicago: Univ. Chicago Press, 1982.
Sylos-Labini, Paolo. Oligopoly and Technical Progress. Cambridge, Mass.: Har-
vard Univ. Press, 1962.
Viner, Jacob. "English Theories of Foreign Trade before Adam Smith, Parts
I and II." J.P.E. 38 (June 1930): 249-301; 38 (August 1930): 404-57.
Reprinted in Studies in the Theory of International Trade. New
1937.
Young, Allyn A. Review of Wealth and Welfare. Q.J.E. 27 (August 1913): 672-
86.

This content downloaded from

24.179.216.56 on Mon, 22 Jun 2020 22:04:07 UTC
All use subject to https://about.jstor.org/terms