374
PREVIEW
In Chapter 15, we developed a simple model of multiple deposit creation that showed
how the Fed can control the level of checkable deposits by setting the required reserve
ratio and the level of reserves. Unfortunately for the Fed, life isn’t that simple; control
of the money supply is far more complicated. Our critique of this model indicated
that decisions by depositors about their holdings of currency and by banks about
their holdings of excess reserves also affect the money supply. To deal with this cri-
tique, in this chapter we develop a money supply model in which depositors and
banks assume their important roles. The resulting framework provides an in-depth
description of the money supply process to help you understand the complexity of
the Fed’s role.
To simplify the analysis, we separate the development of our model into several
steps. First, because the Fed can exert more precise control over the monetary base
(currency in circulation plus total reserves in the banking system) than it can over
total reserves alone, our model links changes in the money supply to changes in the
monetary base. This link is achieved by deriving a money multiplier (a ratio that
relates the change in the money supply to a given change in the monetary base).
Finally, we examine the determinants of the money multiplier.
Study Guide One reason for breaking the money supply model into its component parts is to help
you answer questions using intuitive step-by-step logic rather than memorizing how
changes in the behavior of the Fed, depositors, or banks will affect the money supply.
In deriving a model of the money supply process, we focus here on a simple def-
inition of money (currency plus checkable deposits), which corresponds to M1.
Although broader definitions of money—particularly, M2—are frequently used in
policymaking, we conduct the analysis with an M1 definition because it is less com-
plicated and yet provides a basic understanding of the money supply process.
Furthermore, all analyses and results using the M1 definition apply equally well to the
M2 definition. A somewhat more complicated money supply model for the M2 defi-
nition is developed in an appendix to this chapter, which can be viewed online at
www

We will now derive a formula that describes how the currency ratio desired by
depositors, the excess reserves ratio desired by banks, and the required reserve ratio
set by the Fed affect the multiplier m. We begin the derivation of the model of the
money supply with the equation:
R ϭ RR ϩ ER
which states that the total amount of reserves in the banking system R equals the sum
of required reserves RR and excess reserves ER. (Note that this equation corresponds
to the equilibrium condition RR ϭ R in Chapter 15, where excess reserves were
assumed to be zero.)
The total amount of required reserves equals the required reserve ratio r times the
amount of checkable deposits D:
RR ϭ r ϫ D
Deriving the
Money Multiplier
CHAPTER 16
Determinants of the Money Supply
375
Substituting r ϫ D for RR in the first equation yields an equation that links reserves
in the banking system to the amount of checkable deposits and excess reserves they
can support:
R ϭ (r ϫ D) ϩ ER
A key point here is that the Fed sets the required reserve ratio r to less than 1. Thus
$1 of reserves can support more than $1 of deposits, and the multiple expansion of
deposits can occur.
Let’s see how this works in practice. If excess reserves are held at zero (ER ϭ 0),
the required reserve ratio is set at r ϭ 0.10, and the level of checkable deposits in the
banking system is $800 billion, the amount of reserves needed to support these
deposits is $80 billion (ϭ 0.10 ϫ $800 billion). The $80 billion of reserves can sup-
port ten times this amount in checkable deposits, just as in Chapter 15, because mul-
tiple deposit creation will occur.

an expression linking checkable deposits D to the monetary base MB:
(2)D ϭ
1
r ϩ e ϩ c
ϫ MB
376 PART IV
Central Banking and the Conduct of Monetary Policy
Using the definition of the money supply as currency plus checkable deposits (M ϭ
D ϩ C ) and again specifying C as c ϫ D,
M ϭ D ϩ (c ϫ D) ϭ (1 ϩ c) ϫ D
Substituting in this equation the expression for D from Equation 2, we have:
(3)
Finally, we have achieved our objective of deriving an expression in the form of our ear-
lier Equation 1. As you can see, the ratio that multiplies MB is the money multiplier
that tells how much the money supply changes in response to a given change in the
monetary base (high-powered money). The money multiplier m is thus:
(4)
and it is a function of the currency ratio set by depositors c, the excess reserves ratio
set by banks e, and the required reserve ratio set by the Fed r.
Although the algebraic derivation we have just completed shows you how the
money multiplier is constructed, you need to understand the basic intuition behind it
to understand and apply the money multiplier concept without having to memorize it.
In order to get a feel for what the money multiplier means, let us again construct a
numerical example with realistic numbers for the following variables:
r ϭ required reserve ratio ϭ 0.10
C ϭ currency in circulation ϭ $400 billion
D ϭ checkable deposits ϭ $800 billion
ER ϭ excess reserves ϭ $0.8 billion
M ϭ money supply (M1) ϭ C ϩ D ϭ $1,200 billion
From these numbers we can calculate the values for the currency ratio c and the

1 ϩ c
r ϩ e ϩ c
ϫ MB
CHAPTER 16
Determinants of the Money Supply
377
of our money supply model is to realize that although there is multiple expansion of
deposits, there is no such expansion for currency. Thus if some portion of the
increase in high-powered money finds its way into currency, this portion does not
undergo multiple deposit expansion. In our analysis in Chapter 15, we did not allow
for this possibility, and so the increase in reserves led to the maximum amount of mul-
tiple deposit creation. However, in our current model of the money multiplier, the
level of currency does increase when the monetary base MB and checkable deposits
D increase because c is greater than zero. As previously stated, any increase in MB that
goes into an increase in currency is not multiplied, so only part of the increase in MB
is available to support checkable deposits that undergo multiple expansion. The over-
all level of multiple deposit expansion must be lower, meaning that the increase in M,
given an increase in MB, is smaller than the simple model in Chapter 15 indicated.
1
Factors That Determine the Money Multiplier
To develop our intuition of the money multiplier even further, let us look at how this
multiplier changes in response to changes in the variables in our model: c, e, and r.
The “game” we are playing is a familiar one in economics: We ask what happens when
one of these variables changes, leaving all other variables the same (ceteris paribus).
If the required reserve ratio on checkable deposits increases while all the other vari-
ables stay the same, the same level of reserves cannot support as large an amount of
checkable deposits; more reserves are needed because required reserves for these
checkable deposits have risen. The resulting deficiency in reserves then means that
banks must contract their loans, causing a decline in deposits and hence in the
money supply. The reduced money supply relative to the level of MB, which has

This result can be demonstrated from the Equation 4 formula as follows: When r rises, the denominator of the
money multiplier rises, and therefore the money multiplier must fall.
The analysis just conducted can also be applied to the case in which the required
reserve ratio falls. In this case, there will be more multiple expansion for checkable
deposits because the same level of reserves can now support more checkable deposits,
and the money multiplier will rise. For example, if r falls from 10% to 5%, plugging
this value into our money multiplier formula (leaving all the other variables
unchanged) yields a money multiplier of:
which is above the initial value of 2.5.
We can now state the following result: The money multiplier and the money sup-
ply are negatively related to the required reserve ratio r.
Next, what happens to the money multiplier when depositor behavior causes c to
increase with all other variables unchanged? An increase in c means that depositors
are converting some of their checkable deposits into currency. As shown before,
checkable deposits undergo multiple expansion while currency does not. Hence when
checkable deposits are being converted into currency, there is a switch from a com-
ponent of the money supply that undergoes multiple expansion to one that does not.
The overall level of multiple expansion declines, and so must the multiplier.
3
This reasoning is confirmed by our numerical example, where c rises from 0.50
to 0.75. The money multiplier then falls from 2.5 to:
We have now demonstrated another result: The money multiplier and the money
supply are negatively related to the currency ratio c.
When banks increase their holdings of excess reserves relative to checkable deposits,
the banking system in effect has fewer reserves to support checkable deposits. This
means that given the same level of MB, banks will contract their loans, causing a
decline in the level of checkable deposits and a decline in the money supply, and the
money multiplier will fall.
4
This reasoning is supported in our numerical example when e rises from 0.001

Determinants of the Money Supply
379
3
As long as r ϩ e is less than 1 (as is the case using the realistic numbers we have used), an increase in c raises
the denominator of the money multiplier proportionally by more than it raises the numerator. The increase in c
causes the multiplier to fall. If you would like to know more about what explains movements in the currency
ratio c, take a look at an appendix to this chapter on this topic, which can be found on this book’s web site at
www.aw.com/mishkin. Another appendix to this chapter, also found on the web site, discusses how the money
multiplier for M2 is determined.
4
This result can be demonstrated from the Equation 4 formula as follows: When e rises, the denominator of the
money multiplier rises, and so the money multiplier must fall.
has been extremely small, so changes in it have only a small impact on the money
multiplier. However, there have been times, particularly during the Great Depression,
when this ratio was far higher, and its movements had a substantial effect on the
money supply and the money multiplier. Thus our final result is still an important
one: The money multiplier and the money supply are negatively related to the excess
reserves ratio e.
To understand the factors that determine the level of e in the banking system, we
must look at the costs and benefits to banks of holding excess reserves. When the
costs of holding excess reserves rise, we would expect the level of excess reserves and
hence e to fall; when the benefits of holding excess reserves rise, we would expect the
level of excess reserves and e to rise. Two primary factors affect these costs and bene-
fits and hence affect the excess reserves ratio: market interest rates and expected
deposit outflows.
Market Interest Rates. As you may recall from our analysis of bank management in
Chapter 9, the cost to a bank of holding excess reserves is its opportunity cost, the
interest that could have been earned on loans or securities if they had been held
instead of excess reserves. For the sake of simplicity, we assume that loans and secu-
rities earn the same interest rate i, which we call the market interest rate. If i increases,

of excess reserves, and their level should fall. We have the following result: The excess
reserves ratio e is positively related to expected deposit outflows.
Additional Factors That Determine the Money Supply
So far we have been assuming that the Fed has accurate control over the monetary
base. However, whereas the amount of open market purchases or sales is completely
controlled by the Fed’s placing orders with dealers in bond markets, the central bank
cannot unilaterally determine, and therefore cannot perfectly predict, the amount of
borrowing by banks from the Fed. The Federal Reserve sets the discount rate (inter-
est rate on discount loans), and then banks make decisions about whether to borrow.
The amount of discount loans, though influenced by the Fed’s setting of the discount
rate, is not completely controlled by the Fed; banks’ decisions play a role, too.
Therefore, we might want to split the monetary base into two components: one
that the Fed can control completely and another that is less tightly controlled. The less
tightly controlled component is the amount of the base that is created by discount
loans from the Fed. The remainder of the base (called the nonborrowed monetary
CHAPTER 16
Determinants of the Money Supply
381
FIGURE 1 The Excess Reserves Ratio e and the Interest Rate (Federal Funds Rate)
Source: Federal Reserve: www.federalreserve.gov/releases/h3/hist/h3hist2.txt.
0.0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
5
10

monetary base MB is that the nonborrowed monetary base, which is tied to open mar-
ket operations, is directly under the control of the Fed, whereas the monetary base,
which is also influenced by discount loans from the Fed, is not.
To complete the money supply model, we use MB ϭ MB
n
ϩ DL and rewrite the
money supply model as:
M ϭ m ϫ (MB
n
ϩ DL) (5)
where the money multiplier m is defined as in Equation 4. Thus in addition to the
effects on the money supply of the required reserve ratio, currency ratio, and excess
reserves ratio, the expanded model stipulates that the money supply is also affected
by changes in MB
n
and DL. Because the money multiplier is positive, Equation 5
immediately tells us that the money supply is positively related to both the nonbor-
rowed monetary base and discount loans. However, it is still worth developing the
intuition for these results.
As shown in Chapter 15, the Fed’s open market purchases increase the nonborrowed
monetary base, and its open market sales decrease it. Holding all other variables con-
stant, an increase in MB
n
arising from an open market purchase increases the amount
of the monetary base that is available to support currency and deposits, so the money
supply will increase. Similarly, an open market sale that decreases MB
n
shrinks the
amount of the monetary base available to support currency and deposits, thereby
causing the money supply to decrease.

2003) keeps the interest rate on discount loans (the discount rate) above market inter-
est rates at which banks can borrow from each other, banks usually have little incen-
tive to take out discount loans. Discount lending, DL, is thus very small except under
exceptional circumstances that will be discussed in the next chapter.
Overview of the Money Supply Process
We now have a model of the money supply process in which all four of the players—
the Federal Reserve System, depositors, banks, and borrowers from banks—directly
influence the money supply. As a study aid, Table 1 charts the money supply (M1)
response to the six variables discussed and gives a brief synopsis of the reasoning
behind each result.
Study Guide To improve your understanding of the money supply process, slowly work through
the logic behind the results in Table 1 rather than just memorizing the results. Then
see if you can construct your own table in which all the variables decrease rather than
increase.
CHAPTER 16
Determinants of the Money Supply
383
Table 1 Money Supply (M1) Response
SUMMARY
Change in Money Supply
Player Variable Variable Response Reason
Federal Reserve r ↑↓Less multiple deposit
System expansion
MB
n
↑↑More MB to support
D and C
DL ↑↑More MB to support
D and C
Depositors c ↑↓Less multiple deposit

Central Banking and the Conduct of Monetary Policy
Explaining Movements in the Money Supply, 1980–2002
Application
To make the theoretical analysis of this chapter more concrete, we need to see
whether the model of the money supply process developed here helps us
understand recent movements of the money supply. We look at money sup-
ply movements from 1980 to 2002—a particularly interesting period, because
the growth rate of the money supply displayed unusually high variability.
Figure 2 shows the movements of the money supply (M1) from 1980 to
2002, with the percentage next to each bracket representing the annual growth
rate for the bracketed period: From January 1980 to October 1984, for exam-
ple, the money supply grew at a 7.2% annual rate. The variability of money
growth in the 1980–2002 period is quite apparent, swinging from 7.2% to
13.1%, down to 3.3%, then up to 11.1% and finally back down to 2.3%.
What explains these sharp swings in the growth rate of the money supply?
Our money supply model, as represented by Equation 5, suggests that the
movements in the money supply that we see in Figure 2 are explained by either
changes in MB
n
ϩ DL (the nonborrowed monetary base plus discount loans)
or by changes in m (the money multiplier). Figure 3 plots these variables and
shows their growth rates for the same bracketed periods as in Figure 2.
Over the whole period, the average growth rate of the money supply
(5.3%) is reasonably well explained by the average growth rate of the non-
borrowed monetary base MB
n
(7.4%). In addition, we see that DL is rarely an
important source of fluctuations in the money supply since MB
n
ϩ DL is

385
that have a major impact on the growth rate of the money supply. The cur-
rency ratio c, which is also plotted in Figure 3, explains most of these move-
ments in the money multiplier.
From January 1980 until October 1984, c is relatively constant.
Unsurprisingly, there is almost no trend in the money multiplier m, so the
growth rates of the money supply and the nonborrowed monetary base have
similar magnitudes. The upward movement in the money multiplier from
October 1984 to January 1987 is explained by the downward trend in the
currency ratio. The decline in c meant that there was a shift from one com-
ponent of the money supply with less multiple expansion (currency) to one
with more (checkable deposits), so the money multiplier rose. In the period
from January 1987 to April 1991, c underwent a substantial rise. As our
money supply model predicts, the rise in c led to a fall in the money multi-
plier, because there was a shift from checkable deposits, with more multiple
expansion, to currency, which had less. From April 1991 to December 1993,
c fell somewhat. The decline in c led to a rise in the money multiplier,
because there was again a shift from the currency component of the money
supply with less multiple expansion to the checkable deposits component
FIGURE 2 Money Supply (M1), 1980–2002
Percentage for each bracket indicates the annual growth rate of the money supply over the bracketed period.
Source: Federal Reserve: www.federalreserve.gov/releases.
1200
1000
400
500
600
800
700
900

2.0%
–4.9%
–3.6%
4.4%
0.1%
9.1%
7.0%
10.0%
6.6%
7.1%
8.7%
6.9%
9.1%
6.6%
6.3%
MB
n
+
DL
MB
n
c
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2005
2.4
2.2
2.0
1.8
2.8
3.2
0.40

we use the model to explain the monetary contraction that occurred during
the Great Depression, the worst economic downturn in U.S. history. In
Chapter 8, we discussed bank panics and saw that they could harm the econ-
omy by making asymmetric information problems more severe in credit mar-
kets, as they did during the Great Depression. Here we can see that another
consequence of bank panics is that they can cause a substantial reduction in
the money supply. As we will see in the chapters on monetary theory later in
the book, such reductions can also cause severe damage to the economy.
Figure 4 traces the bank crisis during the Great Depression by showing
the volume of deposits at failed commercial banks from 1929 to 1933. In
their classic book A Monetary History of the United States, 1867–1960, Milton
Friedman and Anna Schwartz describe the onset of the first banking crisis in
late 1930 as follows:
Before October 1930, deposits of suspended [failed] commercial banks had
been somewhat higher than during most of 1929 but not out of line with
experience during the preceding decade. In November 1930, they were more
than double the highest value recorded since the start of monthly data in
1921. A crop of bank failures, particularly in Missouri, Indiana, Illinois, Iowa,
Arkansas, and North Carolina, led to widespread attempts to convert check-
able and time deposits into currency, and also, to a much lesser extent, into
postal savings deposits. A contagion of fear spread among depositors, starting
from the agricultural areas, which had experienced the heaviest impact of bank
failures in the twenties. But failure of 256 banks with $180 million of deposits
in November 1930 was followed by the failure of 532 with over $370 million
of deposits in December (all figures seasonally unadjusted), the most dramatic
being the failure on December 11 of the Bank of United States with over
388 PART IV
Central Banking and the Conduct of Monetary Policy
$200 million of deposits. That failure was especially important. The Bank of
United States was the largest commercial bank, as measured by volume of

200
400
500
300
1929 1930 19321931 1933
Start of First
Banking
Crisis
End of Final
Banking
Crisis
Deposits
($ millions)
6
Milton Friedman and Anna Jacobson Schwartz, A Monetary History of the United States, 1867–1960 (Princeton,
N.J.: Princeton University Press, 1963), pp. 308–311.
CHAPTER 16
Determinants of the Money Supply
389
accounts, and c would rise. Our earlier analysis of the excess reserves ratio
suggests that the resulting surge in deposit outflows would cause the banks
to protect themselves by substantially increasing their excess reserves ratio e.
Both of these predictions are borne out by the data in Figure 5. During the
first bank panic (October 1930–January 1931) c began to climb. Even more
striking is the behavior of e, which more than doubled from November 1930
to January 1931.
The money supply model predicts that when e and c increase, the
money supply will fall. The rise in c results in a decline in the overall level of
multiple deposit expansion, leading to a smaller money multiplier and a
decline in the money supply, while the rise in e reduces the amount of

End of
Final Banking
Crisis
Excess Reserves
Ratio, e
Currency Ratio,
c
0.0
c
e
0.0
end of the crises in March 1933, the money supply (M1) had declined by
over 25%—by far the largest decline in all of American history—and it coin-
cided with the nation’s worst economic contraction (see Chapter 8). Even
more remarkable is that this decline occurred despite a 20% rise in the level
of the monetary base—which illustrates how important the changes in c and
e during bank panics can be in the determination of the money supply. It also
illustrates that the Fed’s job of conducting monetary policy can be compli-
cated by depositor and bank behavior.
390 PART IV
Central Banking and the Conduct of Monetary Policy
FIGURE 6 M1 and the Monetary Base, 1929–1933
Source: Milton Friedman and Anna Jacobson Schwartz, A Monetary History of the United States, 1867–1960 (Princeton, N.J.: Princeton University Press, 1963), p. 333.
19331932193119301929
0
6
7
8
9
19

base MB
n
, which is determined by Fed open market
CHAPTER 16
Determinants of the Money Supply
391
operations. The money supply model therefore allows
for the behavior of all four players in the money supply
process: the Fed through its setting of the required
reserve ratio, the discount rate, and open market
operations; depositors through their decisions about the
currency ratio; the banks through their decisions about
the excess reserves ratio and discount loans from the
Fed; and borrowers from banks indirectly through their
effect on market interest rates, which affect bank
decisions regarding the excess reserves ratio and
borrowings from the Fed.
Key Terms
money multiplier, p. 374 nonborrowed monetary base, p. 381
Questions and Problems
Questions marked with an asterisk are answered at the end
of the book in an appendix, “Answers to Selected Questions
and Problems.”
*1. “The money multiplier is necessarily greater than 1.” Is
this statement true, false, or uncertain? Explain your
answer.
2. “If reserve requirements on checkable deposits were
set at zero, the amount of multiple deposit expansion
would go on indefinitely.” Is this statement true, false,
or uncertain? Explain.

10. The Fed has been discussing the possibility of paying
interest on excess reserves. If this occurred, what
would happen to the level of e?
*11. If the Fed sells $1 million of bonds and banks reduce
their discount loans by $1 million, predict what will
happen to the money supply.
12. Predict what will happen to the money supply if there
is a sharp rise in the currency ratio.
*13. What do you predict would happen to the money
supply if expected inflation suddenly increased?
14. If the economy starts to boom and loan demand picks
up, what do you predict will happen to the money
supply?
*15. Milton Friedman once suggested that Federal Reserve
discount lending should be abolished. Predict what
would happen to the money supply if Friedman’s sug-
gestion were put into practice.
QUIZ
392 PART IV
Central Banking and the Conduct of Monetary Policy
Web Exercises
1. An important aspect of the supply of money is reserve
balances. Go to www
.federalreserve.gov/Releases/h41/
and locate the most recent release. This site reports
changes in factors that affect depository reserve balances.
a. What is the current reserve balance?
b. What is the change in reserve balances since a
year ago?
c. Based on Questions a and b, does it appear that the

value of the multiplier for M2 is:
m
2
ϭ
1 ϩ 0.5 ϩ 3 ϩ 0.5
0.10 ϩ 0.001 ϩ 0.5
ϭ
5.0
0.601
ϭ 8.32
M2 ϭ
1 ϩ c ϩ t ϩ mm
r ϩ e ϩ c
ϫ MB
The M2 Money Multiplier
appendix1
to chapter
16
1
From the derivation here it is clear that the quantity of checkable deposits D is unaffected by the depositor ratios
t and mm even though time deposits and money market mutual fund shares are included in M2. This is just a
consequence of the absence of reserve requirements on time deposits and money market mutual fund shares, so
T and MMF do not appear in any of the equations in the derivation of D in the chapter.
1
An important feature of the M2 multiplier is that it is substantially above the M1
multiplier of 2.5 that we found in the chapter. The crucial concept in understanding
this difference is that a lower required reserve ratio for time deposits or money mar-
ket mutual fund shares means that they undergo more multiple expansion because
fewer reserves are needed to support the same amount of them. Time deposits and
MMFs have a lower required reserve ratio than checkable deposits—zero—and they

Response to
Changes in t
and mm
Changes in r,
c, and e
The M2 Money Multiplier
2
Appendix 1 to Chapter 16
Table 1 Response of the M2 Money Supply to Changes in MB
n
, DL, r, e, c, t, and mm
SUMMARY
Change in M2 Money
Variable Variable Supply Response Reason
MB
n
↑↑More MB to support C and D
DL ↑↑More MB to support C and D
r ↑↓Less multiple deposit expansion
e ↑↓Fewer reserves to support C and D
c ↑↓Less overall deposit expansion
t ↑↑More multiple deposit expansion
mm ↑↑More multiple deposit expansion
Note: Only increases ( ↑ ) in the variables are shown; the effects of decreases in the variables on the money multiplier would be the
opposite of those indicated in the “Response” column.
3
The general outline of the movements of the currency ratio c since 1892 is shown in
Figure 1. As you can see, several episodes stand out:
1. The declining trend in the ratio from 1892 until 1917, when the United States
entered World War I

ratio, is to use the theory of asset demand developed in Chapter 5. Recall the theory
states that four categories of factors influence the demand for an asset such as cur-
rency or checkable deposits: (1) the total resources available to individuals, that is,
wealth; (2) the expected return on one asset relative to the expected return on alter-
native assets; (3) the degree of uncertainty or risk associated with the return from this
asset relative to the alternative assets; and (4) the liquidity of one asset relative to alter-
native assets. Because risk and liquidity factors have not changed independently of
wealth and expected returns and lead to similar conclusions on the historical move-
ments of c, we will focus only on how factors affecting wealth and expected returns
influence c.
What is the relative response of currency to checkable deposits when an individual’s
resources change? Currency is a necessity because it is used extensively by people
with low incomes and little wealth, which means that the demand for currency grows
proportionately less with accumulation of wealth. In contrast, checkable deposits are
held by people with greater wealth, so checkable deposits are less of a necessity. Put
another way, as wealth grows, the holdings of checkable deposits relative to the hold-
ings of currency increase, and the amount of currency relative to checkable deposits
falls, causing the currency ratio c to decline. A decrease in income will lead to an
increase in the amount of currency relative to checkable deposits, causing c to
increase. The currency ratio is negatively related to income or wealth.
The second factor that influences the decision to hold currency versus checkable
deposits involves the expected returns on the checkable deposits relative to currency
and other assets. Four primary factors influence expected returns (and hence the cur-
rency ratio): interest rates on checkable deposits, the cost of acquiring currency, bank
panics, and illegal activity.
1
Interest Rates on Checkable Deposits. By its very nature, currency cannot pay inter-
est. Yet banks can and do pay interest on checkable deposits. One measure of the
expected return on checkable deposits relative to currency is the interest rate on
checkable deposits. As this interest rate increases, the theory of asset demand tells us

its use, raising c.
Bank Panics. Our discussion of interest-rate effects suggests that they did not have
a substantial impact on c before 1980. You might conclude that expected returns
have had little importance in determining this ratio for most of its history. Figure 1
provides us with a clue that we are overlooking an important factor when measur-
ing expected returns solely by the interest rates on assets. The steepest rise in c
occurred during the Great Depression years 1930–1933, when the banking system
nearly collapsed. Legend has it that during this period, people stuffed their mat-
tresses with cash rather than keep it in banks, because they had lost confidence in
them as a safe haven for their hard-earned savings. Can the theory of asset demand
explain this phenomenon?
A bank failure occurs when the bank is no longer able to pay back its depositors.
Before creation of the FDIC in 1933, if you had an account at a bank that failed, you
would suffer a substantial loss—you could not withdraw your savings and might
receive only a small fraction of the value of your deposits sometime in the future. The
simultaneous failure of many banks is called a bank panic, and the Great Depression
years 1930–1933 witnessed the worst set of bank panics in U.S. history. From the end
of 1930 to the bank holiday in March 1933, more than one-third of the banks in the
United States failed.
Bank panics can have a devastating effect on the expected returns from holding
deposits. When a bank is likely to fail during a bank panic, depositors know that if
they have deposits in this bank, they are likely to suffer substantial losses, and the
expected return on deposits can be negative. The theory of asset demand predicts that
depositors will shift their holdings from checkable deposits to currency by withdraw-
ing currency from their bank accounts, and c will rise. This is exactly what we see in
Figure 1 during the bank panics of the Great Depression period 1930–1933 and to a
lesser extent 1893 and 1907, when smaller-scale bank panics occurred. The conclusion
is that bank panics lead to a sharp increase in the currency ratio. Bank panics have
been an important source of fluctuations in this ratio in the past and could be impor-
tant in the future.