A Simple Model of Money and Banking
A Simple Model of Money and Banking
A Simple Model of Money and Banking
http://clevelandfed.org/research/review/
Economic Review 2001 Q3
A Simple Model of
Money and Banking
by David Andolfatto and Ed Nosal
David Andolfatto is an associate professor
of economics at Simon Fraser University.
Ed Nosal is an economist at the Federal
Reserve Bank of Cleveland. This article
represents a simplified version of their
work in progress, Money and Banking
with Costly State Verification. The
authors thank Peter Rupert and
Bruce Smith for their helpful comments
and discussion.
Introduction
This article presents a simple environment
giving rise to banks that create and lend out
money. We define money to be any object
that circulates widely as a means of payment.
In our model, this object takes the form of a
fully secured and redeemable bearer bond.
This monetary instrument is issued by an agent
that can credibly commit to monitoring a pool
of real investments; that is, this capital forms
the requisite backing for a circulating private
debt instrument. While direct trade in securities
is feasible without money, we find that money
can economize on monitoring costs, which
enhances the efficiency of the exchange process.
We define a bank as an agency that simultaneously issues money and monitors investments. In reality, banks also accept deposits of
money, which are then redirected to borrowers.
In our model, banks do not accept deposits;
we do not view this function as a defining
characteristic of a bank.1 In particular, financial
markets also accept deposits of money in
exchange for marketable liabilities (equity and
debt instruments). We think that banks differ
from financial markets in two ways. First, bank
liabilities are designed to be high-velocity
1 Needless to say, there are those who would disagree with this point
of view. To our knowledge, Bullard and Smith (2001) have the only model
featuring intermediaries that simultaneously take deposits, make loans, and
issue circulating liabilities.
21
http://clevelandfed.org/research/review/
Economic Review 2001 Q3
Good 1
Good 2
Good 3
B
y1
c2
y2
y3
c3
IOUy
I. A Simple Model
of Money
y3
The Physical
Environment
IOUy
IOU
B
y1
y1
IOUy
IOUy
y2
IOUy
Figure 1
22
http://clevelandfed.org/research/review/
Economic Review 2001 Q3
Limited
Commitment
and Monetary
Exchange
Consider now an environment in which not
all individuals can commit to keeping their
promises. In particular, suppose that only type
A agents can credibly commit to honoring
claims against their anticipated earnings
stream, y3. In this case, the market value of
both Bs and C s securities as of period 0
equals zero (since these securities represent
unenforceable claims against y 1 and y2, respectively). At first blush, one might be inclined to
think that financial markets could break down
completely. After all, B (C ) can acquire claims
to y2 (y3) only by selling his claims to y1 (y2);
but if these latter claims are worthless, then
B (C ) will be unable to purchase any claims
to y2 (y3).
In fact, if spot markets open up after period
0, then the Pareto optimal allocation can be
implemented with the following sequence of
A
IOUy
y3
y1
y2
A
IOUy
IOUy
Figure 2
23
http://clevelandfed.org/research/review/
Economic Review 2001 Q3
ArrowDebreu
Securities
The type of securities that will be exchanged
on this market are contracts that promise
delivery of a good in the event that returns are
reported to be positive. Since it will always be
in the interest of the person who issues a
security to report zero output (we assume
that people cannot commit to tell the truth),
it has to be understood that the holder of any
such security will, in equilibrium, monitor
project returns.
Clearly, for any kind of trade to occur,
agents must have an incentive to purchase
claims from other agents and then to monitor
them. If the marginal utility of state-contingent
consumption is constant (and equal to unity),
then the parameter restriction (1 )y > is
sufficient to guarantee that trade and monitoring will occur. As in the previous section,
period 0 trades are as follows: A sells a contingent claim to y 3 and purchases one on y 1;
B sells a contingent claim on y 1 and purchases
one on y 2; and C sells a contingent claim on y2
and purchases one on y 3. In period 1, agent A
will monitor agent B; in period 2, agent B will
monitor agent C; and in period 3, agent C will
monitor agent A. Note that the total (economywide) monitoring costs are 3N . Figure 3
summarizes the various trades and monitoring.
y with probability 1
0 with probability .
24
http://clevelandfed.org/research/review/
Economic Review 2001 Q3
IOUy + monitor
3
y3
IOUy
IOUy
IOUy
B
y1
IOU
+
monitor
y2
y1
IOUy
+
monitor
2
Figure 3
Monetary Exchange
Instead of trades in ArrowDebreu securities,
imagine that trades occur in a sequence of spot
markets with the help of a circulating private
debt instrument issued by type A agents. The
private debt instrument issued by A is a contingent claim, as described earlier. Assume that
project returns are realized at the beginning
of each period and that a spot trade of money
for goods occurs after the periods risk has
been resolved.
The sequence of trades is as follows: In
period 1, just after project returns are realized,
type A agents offer their security to anyone
who actually has y 1 to sell (as opposed to a
contingent claim against y 1 that they would
have purchased before realizing project
returns). In the context of this environment,
the people who are in a position to approach
type A agents are the successful type B
agents. It is important to note here that under
this scenario, successful type B agents can
costlessly reveal their success by the very act
of displaying the goods they have to trade; in
other words, there is no need for type A agents
to monitor.
Successful type B agents are willing to
accept a type A security as payment because
they anticipate being able to use this security
as payment for future goods that they desire.
In particular, following the resolution of risk
in period 2, a type B agent can purchase y 2
directly from a successful type C agent. Again,
there is no need for monitoring. Type C agents
willingly accept type A securities as payment
because they represent direct claims against
* =
(1 )y
25
http://clevelandfed.org/research/review/
Economic Review 2001 Q3
A
IOUy
y3
C
monitor
y1
IOU
*
y2
A
y3
IOUy
B
P 1* =
Figure 4
Banking
The type of monitoring activity described
above is commonly regarded as an important
part of the business of banking. But banks are
also associated with the business of creating
liquidity (nowadays in the form of transaction
deposits, but historically also in the form of
paper money), which is injected into the economy by way of money loans (as well as wage
and dividend payments). In the model
described above, the money creation and
monitoring activities are undertaken by separate sets of agents; in reality, these activities
appear to be bundled. What might account for
this bundling?
The first thing to note is that our model is
not necessarily inconsistent with the fact that
money creation and monitoring activities are
bundled (although the model does not necessarily point to bundling as the unique organizational form either). We could, for example,
imagine that the monitoring agent also decides
to take on the responsibility of creating the
economys monetary instrument. In this case,
such an agent would more closely resemble
what is commonly called a bank. So let us
consider what happens when the monitoring
agent also issues money (for example,
banknotes).
1
M
.
(1 )y N
26
http://clevelandfed.org/research/review/
Economic Review 2001 Q3
CB
y3
y3
(1+R) $
$
y2
Figure 5
1
(1 )y
M
> P *1 = P *2 .
N
y1
P *3 =
Transaction Costs
(1 +R *)M
.
(1 )Ny
R $ + monitor
>0 ;
(1 )y
R *M = P *3 N .
P *3 =
R* =
27
http://clevelandfed.org/research/review/
Economic Review 2001 Q3
Conclusions
We have constructed an environment in which
something that looks like a bank emerges as
an efficient exchange mechanism. A bank
monitors projects and issues money that
circulates as a medium of exchange. When
individual transactions are costly, the banking
institution turns out to be an efficient trading
mechanism because goods are only exchanged
between the initial seller and the final consumer. An economy that has private (nonbank)
securities circulating will have some fraction of
final output exchanging hands more than
once; as a result, its transaction costs will be
higher than those of a banking economy.
28
http://clevelandfed.org/research/review/
Economic Review 2001 Q3
References
Bullard, J., and B. Smith. Intermediaries
and Payments Instruments, unpublished
manuscript, 2001.
Cavalcanti, R. A Monetary Mechanism for
Sharing Capital: Diamond and Dybvig Meet
Kiyotaki and Wright, unpublished manuscript, 2001.
Diamond, D. Financial Intermediation and
Delegated Monitoring, Review of Economic
Studies, vol. 51 (1984), pp. 393414.
Kiyotaki, N., and J. Moore. Inside Money
and Liquidity, unpublished manuscript,
London School of Economics, 2000.
Smith, V.C. The Rationale of Central Banking.
Westminster, England: P.S. King and Son,
Ltd., 1936.