The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Vol. 73, No. 5
September/October 1991
3 T h e U .S . B a la n c e S h e e t: W h a t Is It
a n d W h a t D o e s It T e l l Us?
19 D i v i s i a M o n e t a r y S e r v ic e s I n d e x e s
f o r S w itz e r la n d : A r e T h e y U s e fu l
f o r M o n e t a r y T a rg e tin g ?
3 4 T h e U .S . C u r r e n c y S y s te m :
A H is to ric a l P e rs p e c tiv e
THE
FEDERAL
A RESERVE
J Z \ l \NKof
M TSTA O VIS
1
F e d e r a l R e s e r v e B a n k o f St. L o u is
R e v ie w
September/October 1991
In T h is Issu e . . .
Business accounting relies greatly on the balance sheet as a tool for
analyzing a firm ’s financial health. Similar practices do not prevail in na
tional econom ic accounting. Econom ic analysts instead rely mainly on
the GNP accounts as a m easure of national econom ic perform ance.
In the first article in this Review, "T h e U.S. Balance Sheet: W hat Is It
and W hat Does It Tell Us?” Keith M. Carlson provides an overview of
the U.S. balance sheet. He explains its basic stru ctu re and gives some
examples of its usefulness in examining econom ic questions. One of the
most im portant uses of the balance sheet, says Carlson, is to analyze the
role of financial stru ctu re in the process o f econom ic grow th. The ch ief
benefit of the U.S. balance sheet, as currently prepared, is that it forces
the u ser to take a long-term perspective to detect changing econom ic
trends.
* * *
In the second article in this Review, “Divisia M onetary Services Indexes
for Switzerland: Are They Useful for M onetary Targeting?” Piyu Yue
and Robert Fluri derive alternative m onetary aggregates fo r Switzerland
that are based on econom ic aggregation theory, Divisia M l and Divisia
M2. Noting that the historical relationship betw een the Swiss National
Bank’s principal policy instrum ent, the m onetary base, and inflation ap
pears to have broken down in the 1980s, they com pare the p erfo r
m ance of their two new aggregates and simple-sum M l and M2 in ex
plaining Swiss inflation. They find that Divisia and simple-sum M2 p er
form somewhat b etter than the M l aggregates; however, th eir evidence
suggests that simple-sum M2 cannot be controlled. They conclude that
simple-sum M2 is unlikely to b e an adequate substitute for the m onetary
base for policy purposes, but that the oth er aggregates are potentially
w orthy of fu rth er investigation.
* * *
One of the principal roles of the Federal Reserve System is to provide
the United States w ith currency. This role places the Fed at the cen ter
of the U.S. cu rren cy system, a system that provides the foundation for
our m onetary and financial systems. In the third article in this issue,
“The U.S. Currency System: A Historical Perspective,” Steven Russell
uses historical examples to define cu rren cy and cu rren cy systems,
describe the various form s that curren cy can take and identify the
distinctive features of the U.S. cu rren cy system. He also provides a
short history of the U.S. curren cy system from its origins through the
end of the Civil W ar.
In recen t years, Russell notes, there has
alternative curren cy systems. Many of the
proposed are patterned after systems that
of the m odern U.S. curren cy system, says
been considerable interest in
alternatives that have been
existed in the past. D efenders
the author, frequently
SEPTEMBER/OCTOBER 1991
FEDERAL RESERVE
2
p ortray it as the outgrow th of a process of natural selection: as the U.S.
econom y evolved, less efficient systems w ere rejected by the public in
favor of m ore efficient ones. This portrayal interprets the very fact of
the m odern system ’s existence as evidence of its superior efficiency.
Russell’s account suggests that econom ic Darwinism has not played a
dominant role in the development of the m odern U.S. cu rren cy system.
Historically, U.S. cu rren cy systems seem to have risen and fallen for
reasons th at had m ore to do w ith political crises—particularly, m ajor
w ars—than with efficiency considerations.
* * *
BANK OF ST. LOUIS
3
Keith M. Carlson
Keith M. Carlson is an assistant vice president at the Federal
Reserve Bank of St. Louis. Thomas A. Pollmann provided
research assistance.
The U.S. Balance Sheet: What
Is It and What Does It Tell Us?
1 5 USINESS ANNUAL REPORTS provide two
basic accounting statem ents—a balance sheet,
which is also term ed a statem ent of condition,
and an incom e statem ent. A firm 's balance sheet
lists the dollar value of its assets and liabilities
as of a specific date. A firm ’s incom e statem ent
lists its revenues and expenses (the difference
being profit) for a year. Similar statem ents are
prepared on a national level in the United States.
Analogous to a firm ’s incom e statem ent, a na
tion's production of goods and services for a
year (as well as its spending and saving deci
sions) are sum m arized in its gross national pro
duct (GNP) accounts. Analogous to a firm ’s bal
ance sheet, the U.S. balance sheet lists the dollar
value of assets and liabilities fo r U.S. residents.
The flows that are identified in the GNP ac
counts and elsew here are linked to changes in
the levels of assets and liabilities reported in
this balance sheet.
The GNP accounts receive the m ost attention
simply because they focus on cu rren t produc
tion and incom e, w hich in turn, affects and is
affected by, the level of employment. These ac
counts provide vital inform ation on the shortrun perform ance of the economy. On the other
hand, the U.S. balance sheet generally receives
little attention. This might be because it is in
complete, including only nonhum an wealth (see
shaded insert on page 5), and seem s to be m ore
appropriate fo r long-term analysis.
The purpose of this article is to provide an
overview of the U.S. balance sheet. Its stru ctu re
is explained and its usefulness is illustrated by
examining trends in some individual balance
sheet items. Fu rth er examples of its usefulness
are given by examining balance sheet ratios
such as the financial interrelations ratio, the net
foreign balance ratio, the ratio of business
capital to household capital and the relation of
net w orth to inflation.
TH E STANDARD U.S. BALANCE
SHEET
A balance sheet shows the position that a busi
ness or household, or the econom y as a whole,
has reached as a result of its past activity. It
reflects flows o f real and financial activity plus
any revaluations of stocks because of price
changes. Table 1 summarizes the U.S. balance
sheet fo r 1990 as currently prepared by the
Board of Governors of the Federal Reserve
System .1
G en era l D efin itio n s
A balance sheet usually shows all assets and
all liabilities, with the difference called net worth.
1Board of Governors (1991). This is called the C.9 release.
SEPTEMBER/OCTOBER 1991
4
Table 1
Standard U.S. Balance Sheet: 1990 (billions of dollars)1
Beginning-ofyear value
End-ofyear value
Tangible assets
Reproducible assets
Residential structures
Nonresidential structures
Inventories
Consumer durables
Land at market value
Net foreign assets
Foreign assets owned by U.S. residents
Minus: U.S. assets owned by foreigners
U.S. monetary gold and SDRs
$16,017.2
12,163.6
4,546.0
4,633.0
1,050.8
1,933.8
3,853.6
-670.1
852.5
1,522.6
21.0
$16,241.7
12,501.3
4,603.9
4,811.0
1,062.7
2,023.7
3,740.4
- 694.5
941.0
1,635.5
22.0
1.4%
2.8
1.3
3.8
1.1
4.6
-2 .9
NATIONAL NET ASSETS (consolidated)
$15,368.1
$15,569.2
1.3%
Private net worth (consolidated)
Household net assets
Nonfarm noncorporate business
Farm business
Nonfinancial corporate business
Private financial institutions net assets
U.S. holdings of foreign corporate stock2
Minus: Foreign holdings of U.S. corporate stock
Public sector net assets
State and local governments
U.S. government3
Unallocated financial assets
$18,104.5
9,899.3
1,949.2
720.4
3,778.4
1,757.2
95.8
257.4
-2,323.4
-99.9
-2,223.5
-251.3
$18,573.0
10,019.6
1,919.0
758.2
3,870.8
2,005.4
90.5
218.4
-2,616.5
-137.8
-2,478.7
-259.4
2.6%
1.2
-1 .5
5.2
2.4
14.1
-5 .5
-15.2
NATIONAL NET WORTH
$15,368.2
$15,569.2
1.3%
National Net Assets
Percent
change
—
10.4
7.4
4.8
National Net Worth, by sector
—
—
—
—
'Source: Board of Governors of the Federal Reserve System (1991).
2AII corporate equities were subtracted in calculating net assets of households and financial institutions,
includes sponsored credit agencies and monetary authority.
T he U.S. balance sheet is unusual, however,
because the largest category of assets, human
wealth, is not included. National net w orth, as
currently estimated, is all nonhum an wealth.
For the national balance sheet, assets are
divided into two types—tangible and financial.
Tangible assets are econom ic goods that yield a
stream of services in kind. Plant and equipment,
housing, consum er durables and land are all ex
amples of tangible assets. Financial assets are
claims or rights to amounts of money now or in
the future. For every financial asset owned by
FEDERAL RESERVE
BANK OF ST. LOUIS
an econom ic unit there is a corresponding lia
bility owed by another econom ic unit. Financial
assets can be categorized as fixed claims or vari
able claims; variable claims are called equities.
C urrency and deposits, notes, bonds and m ort
gages are all fixed claims. Corporate stock and
the net w orth of noncorporate business are
equities. For a closed economy, financial assets
equal financial liabilities. For an open economy,
national net assets, or national net w orth, is the
total of all tangible assets, U.S. m onetary gold
and SDRs (special drawing rights created and
distributed by the International M onetary Fund)
5
National W ealth and the U.S. B alance Sheet
A nation’s w ealth is defined as the value of
all resources that contribute to the produc
tion of goods and services. The broadest defi
nition would include the goods and services
resulting from both m arket and nonm arket
activities. To b etter understand the process
of w ealth accum ulation, econom ists have
classified w ealth as hum an vs. nonhum an and
tangible vs. intangible. The following classifi
cation is typical:1
Tangible wealth
Human
Nonhuman
Intangible wealth
Human
Nonhuman
Tangible w ealth is m aterial in form , that is,
it is "touchable.” Its quality, or productivity,
depends greatly on intangible factors. Human
tangible w ealth is essentially a population
count, w hich might also take into account the
average age of the population. Nonhuman
tangible wealth is most fam iliar and consists
of structures, durable goods, inventories and
natural resources. This type of w ealth is some
tim es classified as reproducible (man-made)
and nonreproducible (natural).
For all tangible form s of wealth, it is simply
not possible to estim ate value w ithout con
sidering the effect of intangible factors. The
most fundam ental intangible embodied in
human wealth is a person’s stock of knowl
edge. Likewise, health conditions or the state
of medical knowledge, have an im portant
bearing on the value of society’s human
wealth. Still another example of intangible
human w ealth is the mobility of the popula
tion which reflects socio-political factors af
fecting the job m arket. T h ere are also other
form s of intangible wealth that are difficult
to classify as human or nonhum an. These in
clude personal and national security, freedom ,
equity, privacy and a system o f property
rights.2
Nonhuman tangible wealth is easy to under
stand, although estimating its value is difficult
'Kendrick (1976).
2Juster (19/3), pp. 40-48.
because of interaction with the state of tech
nology w hich is an intangible factor. The
value o f stru ctu res and equipm ent reflects
the state o f productive know-how. Similarly,
the value of physical resources like land, sub
soil assets, forests and w ater depend on the
state of technology.
Because so many form s of w ealth are im
m easurable (or at least appear to be), wealth
estimation has generally been limited to its
nonhum an com ponent. Furtherm ore, such
estim ates have been generally restricted to
m arket-oriented activities. The im portance of
human w ealth to national w ealth has been
established clearly, but th ere is little agree
m ent on how to m easure it. Human w ealth is
not traded in a m arket in the same sense as
property, so its value must be determ ined in
directly. Tw o fundam ental approaches have
been used: (1) accum ulate past investm ents in
education and training and allow fo r deprecia
tion, or (2) estim ate the discounted value of
future income. Tw o of the best-known
studies of human wealth in the United States
are by Kendrick (1976) and Jorgenson and
Fraum eni (1989).3 T he table below com pares
their estim ates of the distribution of wealth
for 1969 (the latest y ear for w hich they both
provide estimates).
Distribution of Private National
Wealth: 1969
JorgensonKendrick
Human wealth
Tangible
Intangible
Nonhuman wealth
Tangible
Intangible
Total wealth per capita
(1990 dollars)
69.2%
21.5
47.7
30.8
29.0
1.8
$8,572
Fraumeni
93.9%
N.A.
N.A.
6.1
N.A.
N.A.
$92,540
Kendrick estim ates that U.S. human wealth
(including both tangible and intangible) is
3Also see Eisner (1989), whose estimates of human
wealth as a proportion of the total lie between those of
Kendrick and Jorgenson and Fraumeni.
SEPTEMBER/OCTOBER 1991
6
about 2.3 tim es that of total nonhum an wealth.
Jorgenson and Fraum eni, on the other hand,
estim ate that hum an w ealth is 15.4 tim es that
of nonhum an wealth. The methodologies un
derlying the estim ates are very different with
the Jorgenson-Fraum eni study including esti
mates of the value of nonm arket activities
such as household production and leisure
time.
Despite considerable advances in the under
standing of w ealth, m easurem ent and estim a
and net foreign assets—a balancing item (the
difference betw een foreign assets owned by
U.S. residents and U.S. assets owned by
foreigners).
T h e S ta n d a rd U.S. B a la n ce S h e e t in
1990
Table 1 shows 1990 beginning-of-year and
end-of-year values in cu rren t dollars. The dif
feren ce reflects saving and investm ent flows
(net of depreciation) during the year plus revalu
ations of existing assets.2
For the econom y as a whole, the bottom line
is national net assets, or national net w orth. It
is clear that national net assets are dominated
by tangible assets.3 In fact, w ith the recen t rapid
grow th in foreign ownership of U.S. assets, the
value of tangible assets on U.S. soil has exceeded
the value of the nation’s net assets (owned by
U.S. residents) since 1983.
On the net w orth side of the standard balance
sheet, no liabilities are shown. National net
w orth is defined as the sum of private net
w orth, public sector net assets and unallocated
financial assets (plus a balancing item for foreign
and U.S. holdings of each oth er’s corporate
stock).
Private n et w orth is b rok en down fu rth er by
sector. Household “n et assets” and private finan
2The Board's C.9 release also contains sector balance
sheets for households, farm, nonfarm, noncorporate and
corporate business, private financial institutions and the
rest of the world. For definitions of sectors, see Board of
Governors (1980).
3Tangible assets for federal, state and local governments
are not included in the C.9 release. For further discussion
of the government balance sheet, see Boskin, Robinson
and Huber (1989).
FEDERAL RESERVE
BANK OF ST. LOUIS
tion of many form s o f wealth are still in an
early stage of development. As a result, a
com plete accounting of U.S. wealth is not yet
available on a regular basis; published na
tional balance sheets are generally limited to
the m arket value of nonhum an wealth. De
pending on the estim ates, this portion rep re
sents betw een 6 p ercen t and 31 percent of
total national wealth. In this sense, the U.S.
balance sheet is far from complete.
cial institutions sector "n et assets” are presented
along w ith the net w orth of each type of busi
ness.4 This is useful fo r analysis, but one must
keep in mind that households are the ultimate
ow ners of businesses.
Our focus below is on long-term trends in the
United States, but one development in 1990 is
w orthy of m ention. As shown in table 1, na
tional net w orth grew only 1.3 percent or, given
the increase in GNP prices (GNP implicit price
deflator) of 3.9 percent, it declined 2.6 percent
in real term s. This w eak perform ance is attribu t
able, in part to, declines in real estate prices.
T r e n d s in N ational N et W orth a n d
GNP
One possible use of balance sheet inform ation
is to view national n et w orth as a m easure of
m acroeconom ic perform ance over time. Does it
yield inform ation that differs from that o f GNP?
Many years ago, Raymond Goldsmith and Robert
Lipsey concluded that "National balance sheets
are not intended as a device to m easure econom
ic grow th over time, but they are essential to
study the relations betw een the financial super
stru ctu re and the real in frastru ctu re, w hich
constitute an im portant aspect o f econom ic
grow th.’’5 W ith 30 years of new data, does this
conclusion still hold?
4Equities and pension fund reserves are subtracted from
household net worth to obtain net assets. Private financial
institutions’ net assets are obtained by subtracting cor
porate equities and adding pension fund reserves to their
net worth.
5Goldsmith and Lipsey (1963), p. 25. Also, see Goldsmith
(1985, 1982, 1969, 1966).
7
Figure 1
GNP and National Net Worth
Ratio scale
Billions of 1982 dollars
Ratio scale
Billions of 1982 dollars
2500
2000
1500
1000
1950
1955
1960
1965
Figure 1 shows real national net w orth (NNW)
and real GNP from 1948 to 1990.6 Over the full
period, the grow th rates are the same—3.2 p er
cent. Real NNW is a relatively sm ooth series,
while real GNP displays considerable volatility.
Real NNW shows little cyclical movement, in
dicating that tangible assets are valued over a
long horizon, even if they are utilized m ore or
less intensively during the business cycle.
Table 2 sum m arizes and com pares GNP and
NNW during the 1948-90 period. The referen ce
periods w ere chosen to conform with different
inflation experiences. The 1948-64 period was
one of relatively low inflation, averaging 2.2
percent per year. From 1964 to 1981, inflation
accelerated; it was 3.1 percent in 1965 and 9.4
percent in 1980. The 1981-90 period is more
difficult to define—disinflation from 1981 to
1985 and then m oderately accelerating inflation
over the past five years.
1970
1975
1980
1985
1990
lar rates in both nom inal and real term s. Periods
of accelerating or decelerating inflation, how ever,
produced differing grow th rates fo r the two
m easures. From 1964 to 1981, both nom inal and
real NNW grew faster than GNP. During the
1980s, the opposite occurred; NNW grow th
slowed relative to that of GNP in both nominal
and real term s.
Based on the post W orld W ar II experience,
NNW shows the same long-term trends as GNP.
W hile the grow th rates of NNW and GNP can
deviate fo r several years, fo r purposes of long
term analysis they appear to give the same
answ ers. Goldsmith's and Lipsey’s conclusion
from almost 30 years ago appears to be
confirm ed.
AN ALTERNATIVE VERSION OF
TH E U.S. BALANCE SHEET
During the period of relatively low inflation
from 1948 to 1964, GNP and NNW grew at simi
Since GNP accounts are limited to data on the
cu rren t production of goods and services and
generally omit financial transactions, they are of
6The C.9 release also includes data for 1945-47, but these
years are omitted because of distortions caused by the
transition to peace from World War II. Real NNW was
calculated by using the Department of Commerce
constant-cost net stock of fixed private capital, assuming
that the real value of land changed at the same rate as
real GNP and deflating all other components with the GNP
deflator.
SEPTEMBER/OCTOBER 1991
8
Table 2
A Comparison of GNP and NNW as Measures of Macroeconomic
Performance (compounded annual rates of change)
Measure
1948-64
1964-81
1981-90
1948-90
GNP
GNP deflator
Real GNP
Population
Real GNP per capita
NNW
NNW deflator
Real NNW
Population
Real NNW per capita
6.1%
2.2
3.9
1.7
2.1
5.8
2.1
3.6
1.7
1.9
little use in studying financial superstructure
and the relation betw een real and financial
assets.7 To achieve the purpose of understanding
relations betw een financial and real assets, the
national balance sheet has to be viewed m ore
broadly than NNW.
Financial assets play a key role in the econom ic
development of m arket econom ies, enabling the
tran sfer of lenders’ excess purchasing pow er to
finance spending for borrow ers. If an econom ic
unit consum es less than its income, it accum u
lates assets or retires debt, thereby adding to its
net w orth. If an econom ic unit chooses to ac
cum ulate tangible assets less than its saving, it
must accum ulate financial assets or reduce lia
bilities. Similarly, if an econom ic unit chooses to
accum ulate tangible assets in excess of saving, it
m ust borrow or sell o ff financial assets.
Because of this fundam ental role fo r financial
m arkets, it is useful to focus on national balance
sheets that include the financial asset-liability
stru ctu re.8 A full understanding of the forces
that are driving NNW requires a supplem entary
analysis of financial assets and liabilities.
A G o ld sm ith -T y p e U.S. B a la n ce
S h ee t in 1 9 9 0
Table 3 sum m arizes the U.S. balance sheet in
a Goldsmith-type form at. It uses the same data
that are in the Board of Governors report, but
H'he term “ financial superstructure” is attributable mainly
to Goldsmith. It refers to all aspects of the system of finan
cial markets that channels the funds of savers into invest
ment. For more detailed discussion, see Board of Gover
nors (1980).
Digitized forFEDERAL
FRASER RESERVE BANK
OF ST. LOUIS
9.4%
6.5
2.7
1.1
1.6
10.9
7.2
3.4
1.1
2.3
6.7%
3.6
2.9
1.0
1.9
4.2
2.0
2.2
1.0
1.2
7.6%
4.2
3.2
1.3
1.9
7.5
4.1
3.2
1.3
1.9
sums the assets and liabilities for the separate
sectors—households, businesses and private fi
nancial institutions. Even though there is double
counting, this procedure preserves detailed in
form ation about financial assets and liabilities.
Although the approach used h ere follows
Goldsmith in principle, it is incom plete because
the Board of Governors rep ort does not provide
estim ates of tangible assets held by governm ents.
By adding privately held financial assets to tangi
ble assets, the value of tangible assets held by
business and the equity claims on business held
by households are both included.
The assets and liabilities are fo r the private
sector, so the balancing item is really a m ean
ingless residual that includes governm ent
liabilities, rest-of-w orld liabilities, as well as
equities and net w orth of the private sector.
Consequently, this balance sheet is not offered
as a substitute for the standard Board of Gover
nors version, but as a supplementary summary
of the Board’s rep ort with a focus on private sec
tor assets and liabilities.
T he 1990 values in table 3 indicate that private
sector holdings of financial assets w ere 1.68
tim es the value of tangible assets at the beginning
of the year and 1.70 times at the end of the
year. The sector breakdow n of this ratio sum
marizes the stru ctu re of the U.S. economy. The
household sector’s holdings of financial assets
are about tw ice th eir holdings of tangible assets.
8For a discussion of the importance of keeping financial
assets and liabilities on the national balance sheet, see
Goldsmith and Lipsey (1963).
9
Table 3
Goldsmith-Type U.S. Balance Sheet: 1990 (billions of dollars)1
Beginning-ofyear value
End-ofyear value
Percent
change
Tangible assets
Reproducible assets
Households
Nonfinancial business
Private financial institutions
Land
Households
Nonfinancial business
Financial institutions
Financial assets
Households
Nonfinancial business
Private financial institutions
$16,000.8
12,147.2
5,598.7
6,271.3
277.2
3,853.6
1,458.6
2,376.5
18.5
26,813.6
13,854.5
2,434.5
10,524.6
$16,223.2
12,482.9
5,760.2
6,417.9
304.8
3,740.3
1,315.8
2,404.4
20.1
27,571.8
14,091.4
2,566.1
10,914.3
1.4%
2.8
2.9
2.3
10.0
-2 .9
-9 .8
1.2
8.6
2.8
1.7
5.4
3.7
TOTAL ASSETS
$42,814.4
$43,795.0
2.3%
Liabilities, Equities and Net Worth
Liabilities (private sector)
Households
Nonfinancial business
Private financial institutions
Other liabilities, equities and net worth2
TOTAL LIABILITIES, EQUITIES AND NET WORTH
$18,510.0
3,620.7
4,634.3
10,255.0
24,304.4
$42,814.4
$19,390.3
3,961.3
4,840.4
10,588.6
24,404.7
$43,795.0
4.8%
9.4
4.4
3.3
0.4
2.3%
Assets
'Source: Board of Governors of the Federal Reserve System (1991). Follows Raymond Goldsmith’s procedure of combin
ing (adding) sector balance sheet items to derive a ‘total’ balance sheet. This balance sheet is incomplete, however,
because it omits assets held by government and foreigners.
Calculated as a residual. Reflects considerable double counting and also includes liabilities of government and
foreigners.
The nonfinancial business sector, on the other
hand, holds financial assets equal to about 30
percent of its tangible assets. Financial institu
tions, almost by definition, hold very few tangi
ble assets.
Financial assets can be com pared with liabilities
to show the net m onetary cred itor status of the
different sectors. Households hold financial
assets about 3.7 times as large as their liabilities,
w ith fixed claims almost 1.5 tim es as large as
liabilities. Nonfinancial businesses have liabilities
almost tw ice as large as their financial assets,
w ith the ratio about the same w hen variable
claims and liabilities are subtracted. Private fi
9“ Credit market instruments is a core group of debt claims
that is the principal medium used by nonfinancial sectors
in raising funds through formal credit channels.” [Board of
Governors (1980), pp. 42-43.] It includes all government
nancial institutions are net m onetary creditors,
although not to the extent that households are.
Reproducible assets are divided into the same
categories as in table 1 and in the GNP ac
counts—residential structures, nonresidential
plant and equipment, inventories and consum er
durables. Financial assets can be divided into
many types, but h ere they are grouped into five
categories—cu rren cy and deposits, credit m arket
instrum ents, equities (both corporate equities
and the n et w orth of noncorporate business),
reserves (pension fund and life insurance) and
other (which includes security and trade
credit).9
securities, corporate and foreign bonds, mortgages, con
sumer credit, bank (not elsewhere classified) and other
loans.
SEPTEMBER/OCTOBER 1991
10
Figure 2
Distribution of Tangible Assets
Percent
T r e n d s in T otal A ssets
Postw ar trends in total assets are summarized
by charting the com ponents of both tangible
assets and financial assets relative to their respec
tive totals.
T a n g i b l e a s s e t s — Figure 2 sum m arizes the
m ajor com ponents of tangible assets in 1982
dollars relative to the total of tangible assets, but
broken down by type of asset as in table 1 (the
same as in the GNP accounts). Vertical lines
(1964 and 1981) correspond to the inflation
episodes in table 2. For inventories and land,
the trend generally has been downward
throughout the postw ar period. In the case of
consum er durables, the trend is upward
throughout the period. For fixed residential in
vestm ent, the trend was upward b efore 1964
but has been downward since then. Fixed
nonresidential investm ent has trended upward,
although there appears to be a flattening in the
1980s.
10Land was not included because it is fixed and
nonreproducible. Its real value can change but not its
quantity.
"Higher rates of inflation increase the uncertainty
associated with rates of return on financial assets, making
tangible assets more attractive to investors. See Cagan
and Lipsey (1978).
FEDERAL RESERVE
BANK OF ST. LOUIS
Percent
The relationship betw een the grow th of real
reproducible tangible assets and inflation was
exam ined.10 The conventional wisdom is that in
vestors view tangible assets as a good hedge
against inflation.11 Table 4 shows the correlation
coefficients for the year-to-year percent change
in reproducible tangible assets and inflation.
None of the coefficients is significant and posi
tive fo r each of the subperiods. Explaining trend
movements in the com ponents of tangible assets
is apparently m uch m ore complex than indi
cated by a simple inflation model.
F i n a n c i a l a s s e t s — Privately held financial
assets w ere collected into categories as shown
in figure 3. C urrency and deposits (broadly
defined) drifted downward from the end of
World W ar II until the early 1960s, stabilized
until 1972, shifted to a higher level and then
fell from 1984 to 1990. This category reflects a
num ber of financial innovations throughout the
period, notably certificates of deposit in the
11
Table 4
Correlation Coefficients: Percent Change in Reproducible
Tangible Assets (1982 dollars) and Inflation1____________
1948-64
1948-90
1965-81
1982-90
Consumer durables
Residential structures
Nonresidential plant and equipment
Inventories
-.1 6
-.4 2 *
.08
.23
.30
.60*
.52*
.66*
-.6 4 *
-.2 8
-.5 0 ’
.04
-.8 6 *
-.8 8 *
-.5 7
-.3 5
'Significant at the 5 percent level
'Inflation is year-to-year percent change in GNP deflator.
Figure 3
Distribution of Financial Assets
Percent
Percent
45
1960
1965
1960s, m oney m arket mutual funds in the 1970s
and the paym ent of interest on checkable
deposits in the 1980s.12
The trend of credit m arket instrum ents was
quite flat until the mid-1970s, shifted to a higher
level until the early 1980s and rose sharply from
1982 to 1990. Because federal securities are an
1975
1980
im portant com ponent of credit m arket instru
ments, about half of the increase in the 1980s
can be traced to the rapid grow th in the federal
debt.
The equity portion of financial assets shows a
pattern generally the opposite o f that for credit
m arket instrum ents. The downward trend in
12For a brief financial history of the United States, see
Council of Economic Advisers (1991), chapter 5.
SEPTEMBER/OCTOBER 1991
12
Table 5
Correlation Coefficients: Percent Change in Financial Assets
(1982 dollars) and Inflation1_______________________________
1948-90
1948-64
1965-81
1982-90
Currency and deposits
Credit market instruments
Life insurance and pension
fund reserves
Equities
Other financial assets
-.1 3
-.4 2 *
-.41
-.6 9 *
-.2 6
-.31
-.1 9
-.8 4
-.3 8 *
-.2 0
-.1 5
-.5 6 *
.15
.57*
-.1 8
-.1 5
-.3 7
-.0 3
-.2 8
-.8 3
‘ Significant at the 5 percent level
’ Inflation is year-to-year percent change in GNP deflator.
equities began in the high-inflation 1970s, but
continued through the disinflation of the 1980s.13
These trends suggest a com plem entary relation
betw een equities and credit m arket instrum ents
(including governm ent securities). The total of
these two categories has varied betw een 60 and
70 percent of all financial assets since W orld
W ar II.
Life insurance and pension fund reserves rose
gently until 1980, and then accelerated in the
1980s. This recen t acceleration is consistent
with a num ber of explanations. One would be
that it represented a favorable long-term plan
ning response to the deceleration of inflation.
Another would be the demographics of the de
cade which included a rise in the average age
of the population.14
The residual com ponent of financial assets,
called “oth er,” reflects mainly trade and security
credit. This category moved upward slowly but
steadily until the mid-1970s and then stabilized.
As with reproducible tangible assets, coeffi
cients w ere calculated for the correlation b e
tw een the percent change in financial assets in
1982 dollars and inflation. These results are
summarized in table 5. Most of the coefficients
are negative, although most are insignificant.
Even though nominal financial assets tend to in
crease with inflation, their grow th is generally
outpaced by inflation so that in real term s there
is an inverse relationship.
13lt was formerly believed that corporate stocks were a
hedge against inflation. Fischer and Modigliani (1978) suggest that this changed when investors realized that higher
inflation carries with it a higher real tax burden.
Digitized for FEDERAL
FRASER RESERVE BANK OF ST. LOUIS
SOME USES OF THE BALANCE
SHEET
The U.S. balance sheet covers a relatively small
portion of the nation’s wealth. However, it can
yield insights into particular relationships that
cannot be fully analyzed using inform ation only
from GNP accounts. T h e accumulation of flows
into stocks provides a built-in long-term perspec
tive that is generally missing w ith GNP accounts.
By lengthening the tim e perspective, balance
sheet inform ation can shed new light on some
commonly held perceptions about economic
trends.
F in a n cia l In terrela tio n s Ratio
One of the most im portant applications of
balance sheet inform ation is the calculation and
analysis of the financial interrelations ratio.15
This ratio m easures the size of the financial
superstructure relative to the real in frastru c
ture. Specifically, it is the ratio of the value of
financial assets to the value of tangible assets.
The financial interrelations ratio provides a
fram ew ork for the analysis of the relationship
betw een financial development and economic
growth. However, as Goldsmith points out, “Eco
nomic grow th is so complex a phenomenon, ob
viously determ ined or influenced by basic fac
tors of a physical, technological, and masspsychological nature, that an attempt to isolate
the effects of apparently secondary forces such
' “Carlson (1990).
'sGoldsmith (1966)
13
Figure 4
Financial Interrelations Ratio1
fin a n c ia l assets relative to tangible assets (both in 1982 dollars)
as the ch aracter of financial institutions and the
nature of credit practices does not promise suc
cess.”16 Generally, the argum ent is that a rise in
the interrelations ratio indicates a broadening of
the range of financial assets and institutions.
This prom otes the flow of saving into its most
productive uses w hich stimulates econom ic
grow th and increases productivity.17
Figure 4 shows the interrelations ratio for the
1948-90 period. The factors influencing its move
m ent are num erous and complicated, although
inflation appears to have played a role. Prices of
tangible assets, the denom inator in the ratio,
tend to increase m ore than other prices during
periods of accelerating inflation, and by a lesser
amount w hen it decelerates. The ratio fell to its
postw ar low during the high-inflation period of
the 1970s before rising during the disinflation
of the 1980s. Such an explanation is simplistic
because a full analysis of the interrelations ratio
would consider all other factors entering into its
16Goldsmith (1969), p. xi.
17For more detailed discussion of this theory, see Goldsmith
(1969), pp. 390-401. Also see Shaw (1973).
18Goldsmith (1969), p. 97.
19Recently an argument has been offered challenging the
determ ination. Nonetheless, inflation is a factor
influencing the ratio.18 On the other hand, real
GNP grow th does not appear to be related
systematically to the ratio, especially since the
mid-1970s. Thus, even though the financial in
terrelations ratio shows interesting movements
in the postw ar period, it is only a starting point
in the analysis of financial stru ctu re and eco
nomic grow th.19
One facet of the interrelations ratio that has
produced con cern in the 1980s is the rapid
grow th of credit m arket debt in the private sec
tor. Expansion of debt permits m ore spending
than otherw ise, but adds to the severity of a
recession w hen the pace of econom ic activity
slows. To maintain debt payments, households
and businesses have to restrain their spending
or default on their loans. W idespread loan de
faults could endanger the econom ic health of the
financial system.
notion that growth in financial structure is always
beneficial. See Fingleton (1991).
SEPTEMBER/OCTOBER 1991
14
Figure 5
Credit Market Debt Relative to Tangible Assets
Percent
Figure 5 puts the 1980s expansion of debt into
perspective. All credit m arket debt of households,
private nonfinancial business and private finan
cial institutions is included and m easured
against total privately held tangible assets. Debt
expanded at an 11 percent annual rate from
1983 to 1990, pushing the ratio of debt to tangi
ble assets to a historical high of 0.52. The ex
tent of the increase is dampened somewhat
w hen the ratio is calculated w ith 1982 dollars,
but 1990 is still at a historical high; it appears
to be leveling off, however. W hether this debt
burden is “too high” will probably not be
answ ered until the strength and duration of the
recovery from the recen t recession is clear.
In fla tio n a n d th e D istrib u tio n o f
N et W orth
An additional use of balance sheet inform ation
is to analyze the effect of inflation on the net
w orth of various sectors. The standard theory
of such effects is outlined in the shaded insert
20No attempt is made here to measure anticipated inflation
However, based on current procedures, the variances of
change in inflation and the unanticipated change have
been found to be similar. See Ball and Cecchetti (1990),
p. 242.
FEDERAL RESERVE
BANK OF ST. LOUIS
Percent
at right.20 Figure 6 shows the distribution of
private net w orth among sectors.
The proportion of private net w orth held by
households gradually increased from 1948 until
inflation accelerated sharply in the mid-1970s; it
then declined until 1981. Since then, households'
share of net w orth has risen as the disinflation
continued through most of the decade. These
responses are typical of a sector that is a net
m onetary creditor.
The nonfinancial business sector has shown a
variety of long-term trends, but the response to
the acceleration and deceleration of inflation is
similar for the three subsectors because they
are all m onetary debtors. Farm business has
been in a long-term decline throughout the post
w ar period, interrupted by a slight increase
from 1971 to 1980 w hich prim arily reflected a
rise in farm real estate values. Nonfarm non cor
porate business declined as a share of private
net w orth until 1976, increased until 1980, and
15
Inflation and the B alance Sheet
units are net m onetary creditors. Net m one
tary creditors are harm ed by unanticipated
inflation because the purchasing pow er of
their m onetary assets declines m ore than the
real value of their m onetary liabilities. Simi
larly, net m onetary debtors benefit from un
anticipated inflation. These effects take place
without any action on the part of the eco
nomic unit and rep resen t a passive redistribu
tion of wealth.
Inflation has many effects, but most dis
cussed are its effects on the distribution of
income and w ealth.1 If inflation w ere fully
and correctly anticipated, nominal interest
rates on fixed claim assets would adjust to
com pensate for the declining purchasing pow
er of the principal. Generally, however, in
terest rates have not com pletely com pensated
for inflation. As a result, periods of inflation
have been accom panied by arbitrary tran s
fers of real net w orth from net m onetary
creditors to net m onetary debtors w hen infla
tion is accelerating, o r vice versa w hen it is
decelerating.
Even if inflation is anticipated and reflected
in nominal interest rates, added uncertainty
about future prices can affect econom ic deci
sions.2 As a result, econom ic units will at
tempt to redistribute their assets for protec
tion from inflation. This will elevate prices of
tangible assets relative to financial assets.
The key factors in determ ining w hether
econom ic units will benefit from , or be harm
ed by, inflation are (1) w hether the inflation
is anticipated and (2) w hether the econom ic
2See Cagan and Lipsey (1978).
1See Bach and Stephenson (1974), Cagan and Lipsey
(1978), Conard (1964), Fischer and Modigliani (1978)
and Kessel and Alchian (1962).
Figure 6
Distribution of Private Net Worth
Percent
Percent
6 0 1---------
---------60
Households1
Nonfinancial
corporate business
Nonfarm noncorporate
business
Farm business
Private financial institutions'
1950
1955
1960
1965
1970
1975
1980
1985
1990
’ Net assets
SEPTEMBER/OCTOBER 1991
16
Figure 7
Net Foreign Assets Relative to Total Assets Held
by the Private Sector
since then has moved back to its 1976 level.
Nonfinancial corporate business is difficult to
characterize for the full period. Its response to
the acceleration and deceleration of inflation
seems quite clear, with its net w orth proportion
increasing during the acceleration and then fall
ing back during the disinflation.
The net w orth of private financial institutions
does not seem to be affected m uch by swings of
inflation, con trary to the well-known problem s
of the savings and loan industry. Private finan
cial institutions are net m onetary creditors, but
the difference betw een their financial assets
and their liabilities is very small.21
F o r e ig n O w n ersh ip o f U.S. A ssets
Another application of the U.S. balance sheet
is to exam ine con cern about the accumulation
21It is difficult to relate to the private financial sector
because of its heterogeneity. It consists of commercial
banking, savings institutions, insurance (including private
pension funds and state and local government retirement
funds) and other (including finance companies and mutual
funds).
22This concern and several others are examined in Ott
(1989).
23These trends in foreign ownership are unlikely to continue.
For example, assets of U.S. business acquired or
FEDERAL RESERVE
BANK OF ST. LOUIS
of U.S. assets by foreigners during the 1980s. A
common perception is that foreigners could
eventually own m ore than 50 percent of business
capital leading to the potential for foreign con
trol of the U.S. economy. This would threaten
U.S. econom ic sovereignty and national securi
ty.23 Balance sheet data can b e used to examine
this concern.
Figure 7 shows n et foreign assets as a percen
tage of total assets in the United States. It is
clear that the proportion of U.S. assets held by
foreigners has increased sharply since the early
1980s. Foreign direct investment, how ever, has
increased from only 0.4 percent in 1980 to about
1 percent of U.S. total assets in 1990. W ith con
tinued grow th at this pace, foreign ow nership
would not exceed 50 p ercen t fo r 800 years.23
Rather than "signaling an econom y in decline,
established by foreign investors dropped by 25 percent
from 1989 to 1990. See Fahim-Nader (1991).
17
Figure 8
Ratio of Business Capital to Household C apital 1
1950
1955
1960
1965
1970
1975
1980
1985
1990
1Nonresidential plant and equipment divided by residential structures plus consumer durables
(all in 1982 dollars)
such investm ent by foreigners is a m easure of
the econom y’s vigor.”24 On a worldwide basis,
foreign direct investm ent increases econom ic
w elfare by moving resources from less to m ore
productive uses.
R atio o f B u s in e ss Capital to
H o u s e h o ld Capital
Another concern that developed in the 1980s
was that the United States was not channeling
its saving into the "right” kind of investm ent.25
Mainly because of a com bination of inflation
and tax shelters, savings w ere directed tow ard
household capital rath er than productive busi
ness capital.26 Business investm ent in plant and
equipm ent is a m ajor vehicle fo r increasing eco
nom ic growth. It boosts productivity by provid
ing m ore capital per w orker and also embodies
technological improvements. Household capital,
on the oth er hand, provides services to con
sum ers but does not add directly to productive
capacity. Figure 8 shows the ratio of nonresi-
^O tt (1989), p. 63.
25This view is developed in Rutledge and Allen (1989).
26For a discussion of the effects of the Tax Reform Act of
1986, see Fazzari (1987) and Slemrod (1990).
dential plant and equipment to household capi
tal—the sum of consum er durables and residen
tial structures.
Following W orld W ar II, households enlarged
their stock of capital until 1964. From 1964 to
1970, grow th of business capital stock exceeded
that of household capital stock. For the next 12
years, business and household capital grew at
roughly the same rate. Since 1982, how ever, the
grow th of household capital has exceeded that
of business capital. Boosting the overall level of
saving is the prim ary vehicle for stimulating
econom ic growth. T h ere is also potential for
faster grow th by designing policies that direct
the flow o f saving away from household capital
into business capital.27
SUMMARY
Economic analysts rely mainly on the nation's
GNP accounts as a source of inform ation on
econom ic perform ance. For purposes of under-
27For a more complete discussion of saving and its role in
economic growth, see Cullison (1990) and Harris and
Steindel (1991).
SEPTEMBER/OCTOBER 1991
18
standing the forces at w ork in the determ ina
tion of cu rren t production, GNP accounts are
indispensable. Generally overlooked, how ever, is
another source of inform ation—the nation’s
balance sheet. Business accounting relies greatly
on the balance sheet as a tool for analyzing a
firm ’s financial health. Similar practices do not
prevail in national econom ic accounting.
W hat would appear to be one of the m ost im
portant items in the U.S. balance sheet is the
m easure of national n et w orth. W hen com pared
with GNP as a m easure of long-term econom ic
perform ance, how ever, it does not seem to of
fer m uch added inform ation.
Probably the most im portant use of balance
sheet data is to analyze the role of financial
stru ctu re in the process of econom ic grow th. A
variety of other questions, how ever, can also be
exam ined by developing ratios of particular
balance sheet items. The chief benefit of the
U.S. balance sheet, as it is currently prepared,
seems to be that it forces the u ser to take a
long-term perspective to detect changing trends.
W hat appeared to be m ajor concerns during the
1980s sometimes took on a different in terp reta
tion w hen viewed from the perspective of the
U.S. balance sheet over the entire post-W orld
W ar II period.
REFEREN CES
Bach, G. L., and James B. Stephenson. “ Inflation and the
Redistribution of Wealth,” Review of Economics and
Statistics (February 1974), pp. 1-13.
Ball, Laurence, and Stephen G. Cecchetti. “ Inflation and
Uncertainty at Short and Long Horizons," Brookings Papers
on Economic Activity, 1 (1990), pp. 215-45.
Board of Governors of the Federal Reserve System. Balance
Sheets for the U.S. Economy 1945-90 (March 1991).
_______ . Introduction to Flow of Funds (June 1980).
Boskin, Michael J., Marc S. Robinson, and Alan M. Huber.
“Government Saving, Capital Formation, and Wealth in the
United States, 1947-85,” in Robert E. Lipsey and Helen
Stone Tice, eds., The Measurement of Saving, Investment,
and Wealth (University of Chicago Press, 1989), pp.
287-356.
Cagan, Phillip, and Robert E. Lipsey. The Financial Effects of
Inflation (Ballinger Publishing Company, 1978).
Carlson, Keith M. “ On Maintaining a Rising U.S. Standard of
Living Into the Mid-21st Century,” this Review (March/April
1990), pp. 3-16.
Conard, Joseph W. “The Causes and Consequences of Infla
tion,” in Inflation, Growth and Employment, A Series of
Research Studies prepared for the Commission on Money
and Credit (Prentice-Hall, 1964), pp. 1-144.
Council of Economic Advisers. 1991 Annual Report (GPO,
February 1991).
FEDERAL RESERVE
BANK OF ST. LOUIS
_______ . 1987 Annual Report (GPO, January 1987).
Cullison, William E. “ Is Saving Too Low in the United
States?” Federal Reserve Bank of Richmond Economic
Review (May/June 1990), pp. 20-35.
Eisner, Robert. The Total Incomes System of Accounts
(University of Chicago Press, 1989).
Fahim-Nader, Mahnaz. “ U.S. Business Enterprises Acquired
or Established by Foreign Direct Investors in 1990,” Survey
of Current Business (May 1991), pp. 30-39.
Fazzari, Steven M. “Tax Reform and Investment: How Big
an Impact?” this Review (January 1987), pp. 15-27.
Fingleton, Eamonn. “ Highly Speculative,” Atlantic (June
1991), pp. 22-25 .
Fischer, Stanley, and Franco Modigliani. “Towards an
Understanding of the Real Effects and Costs of Inflation,”
Weltwirtschaftliches Archiv (1978), pp. 810-33.
Goldsmith, Raymond W. Comparative National Balance
Sheets: A Study of Twenty Countries, 1688-1978 (University
of Chicago Press, 1985).
_______ . The National Balance Sheet of the United States,
1953-1980 (University of Chicago Press, 1982).
_______ . Financial Structure and Development (Yale Univer
sity Press, 1969).
_______ .“The Uses of National Balance Sheets,” Review of
Income and Wealth (June 1966), pp. 95-133.
Goldsmith, Raymond W., and Robert E. Lipsey. Studies in
the National Balance Sheet of the United States, Volume I
(Princeton University Press, 1963).
Harris, Ethan S., and Charles Steindel. “ The Decline in U.S.
Saving and Its Implications for Economic Growth,” Federal
Reserve Bank of New York Quarterly Review (Winter 1991),
pp. 1-19.
Jorgenson, Dale W., and Barbara M. Fraumeni. “ The Ac
cumulation of Human and Nonhuman Capital, 1948-84,” in
Robert E. Lipsey and Helen Stone Tice, eds., The
Measurement of Saving, Investment, and Wealth (University
of Chicago Press, 1989), pp. 227-86.
Juster, F. Thomas. “A Framework for the Measurement of
Economic and Social Performance,” in Milton Moss, ed.,
The Measurement of Economic and Social Performance (Na
tional Bureau of Economic Research, 1973), pp. 25-109.
Kendrick, John W. The Formation and Stocks of Total Capital
(National Bureau of Economic Research, 1976).
Kessel, Reuben A., and Armen A. Alchian. “ Effects of Infla
tion,” Journal of Political Economy (December 1962), pp.
521-37.
Ott, Mack. “ Is America Being Sold Out?” this Review
(March/April 1989), pp. 47-64.
Rutledge, John, and Deborah Allen. Rust to Riches (Harper
and Row, 1989).
Shaw, Edward S. Financial Deepening in Economic Develop
ment (Oxford University Press, 1973).
Slemrod, Joel, ed. Do Taxes Matter? The Impact of the
Tax Reform Act of 1986 (MIT Press, 1990).
19
Piyu Yue and R obert Fluri
Piyu Yue is a research associate scholar at the 1C2 Institute,
University of Texas at Austin. Robert Fluri is an economist at
the Swiss National Bank. This article was written while both
authors were visiting scholars at the Federal Reserve Bank of
St. Louis. Lynn Dietrich provided research assistance.
Divisia M onetary Services In
dexes for Sw itzerland: Are
They Useful fo r M onetary
Targeting?
F r o m 1973 TO 1989, INFLATION in Switzerland was roughly one-half that in the United
States. For example, consum er prices in Sw itzer
land rose 3.5 percent per year during this
16-year period com pared with the 6.6 percent
average annual rise in U.S. consum er prices.
Similarly, Swiss wholesale prices rose at an an
nual rate of 2 p ercen t during this period in con
trast to the nearly 6 percent annual average
increase in the U.S. producer price index. In
deed, the Swiss inflation experience, along with
that of W est Germany, is often cited as an ex
cellent example of the gains that accrue to a
nation whose central bank conducts m onetary
policy by announcing—and achieving—a target
fo r the grow th of a m onetary aggregate.
The Swiss central bank has used m onetary
base grow th rate targets since 1980. Because
the historical relationships betw een m onetary
base grow th and econom ic activity have changed
m arkedly since the end of the 1980s, the Swiss
have begun to reconsider the use of annual
m onetary aggregate targets, and are considering
the potential usefulness of broad er m onetary
aggregates as indicators of m onetary policy.
This paper develops two alternative broader
monetary aggregate m easures, Divisia M l and
M2, for Switzerland and com pares their poten
tial usefulness as m onetary indicators w ith the
Swiss M l and M2 aggregates as usually defined.
First, however, we show why questions have
been raised about the continued usefulness of
the annual m onetary base grow th rate target in
Switzerland. W e then discuss the methodology
underlying the Divisia approach to constructing
m onetary aggregates and use this methodology
to derive Swiss Divisia M l and M2 m easures.
Next, we examine the relationship betw een
Swiss inflation and the grow th rates of Swiss
M l and M2 and the Swiss Divisia M l and M2
aggregates to determ ine their relative usefulness
as m onetary policy indicators. Finally, we exa
mine the relationships betw een these various
m onetary aggregates and the m onetary base to
assess the extent to which the Swiss central
bank could control their grow th.
SOME BACKGROUND ON THE
SWISS MONETARY BASE
Prior to 1986, the relationship betw een the
m onetary base and econom ic activity in Sw itzer
land was quite close; this link, how ever, has
broken down since then. T he sudden change
SEPTEMBER/OCTOBER 1991
20
■I
Figure 1
Inflation and Swiss Monetary Base Growth
Percent
Percent
can be illustrated, in part, by looking at the
relationship betw een inflation and the grow th
rate of the m onetary base as depicted in fig
ure 1. Until 1986, Swiss inflation movements
lagged about th ree years behind corresponding
variations in the base m oney grow th rate. Since
1986, how ever, this pattern no longer holds. For
example, the sharp drop in Swiss inflation in
1986 was attributable to substantial reductions
in the prices of im ported goods. Consequently,
it appears that m onetary base grow th neither
contributed to this decline nor provided any
w arning that it would occur; indeed, the grow th
of the Swiss m onetary base was virtually cons
tant from 1984-1987.
Similarly, movem ent in the Swiss m onetary
base from 1987-1989 (in particular, the sharp
drop in 1988) yielded neither w arning nor ex
planation of the sharp rise in Swiss inflation in
1989. This change followed two significant insti
FEDERAL RESERVE
BANK OF ST. LOUIS
tutional innovations in the Swiss banking sys
tem. First, a new electronic interbank payment
system (Swiss Interbank Clearing System, SIC)
was introduced in the sum m er of 1987. Then,
on Jan uary 1, 1988, reduced reserve req u ire
m ents on Swiss bank deposits w ent into effect.
In response to these changes, Swiss banks
have sharply reduced their reserve balances at
the Swiss National Bank (SNB). Swiss bank
reserves dropped from m ore than SF8 billion
(about $5.5 billion) at the end of 1987 to SF3 bil
lion (about $2.1 billion) by the end of 1989.
As a result of changes in the relationship b e
tw een the m onetary base and inflation, the con
tinued usefulness of the monetary base as a
monetary policy indicator has been questioned.
One suggestion is to rely more on broader mone
tary aggregates as monetary policy indicators.
21
MONETARY AGGREGATION
Generally, central banks worldwide use essen
tially identical procedures to con stru ct th eir na
tions’ m onetary aggregates. They first define the
specific aggregate—that is, they determ ine which
financial com ponents it will include—and then
they simply “add” its selected com ponents to
gether. Not too surprisingly, these m onetary ag
gregates are called "simple-sum” aggregates.
Simple-sum aggregation has been criticized for
failing to distinguish betw een the differing de
grees of m onetary (transaction) services and
store-of-value services provided by the com
ponents in the m onetary aggregate. Presumably,
only the form er (that is, m onetary or tran sac
tion) services should be included w hen a m one
tary aggregate is considered. Friedm an and
Schw artz (1970) have described this problem:
This [summation] procedure is a very special case
of the more general approach discussed earlier. In
brief, the general approach consists of regarding
each asset as a joint product having different
degrees of "moneyness,” and defining the quantity
of money as the weighted sum of the aggregate
value of all assets, the weights for individual assets
varying from zero to unity with a weight of unity
assigned to that asset or assets regarded as having
the largest quantity of "moneyness” per dollar of
aggregate value. The procedure we have followed
implies that all weights are either zero or unity.
The more general approach has been suggested
frequently but experimented with only occasional
ly. We conjecture that this approach deserves and
will get much more attention than it has so far
received, (pp. 151-52)
As Friedman and Schwartz surmised, economic
aggregation theory and statistical index num ber
theory have been used to provide both theoreti
cal and em pirical solutions to the problem of
m onetary aggregation.1 This research has led to
the development of alternative m onetary aggre
gates, in particular, the Divisia m onetary service
m easure.2
T h e Divisia In d e x N u m b e r
Index num bers are widely used to provide a
single broad m easure for a disparate collection
of items. Well-known examples of index numbers
are the industrial production index, the consum
er price index and the producer price index.
These index num bers depend upon both the
prices and quantities of items included in the in
dex because the values of commodities involved
are determ ined by their physical quantities and
corresponding prices.
Because quantities of financial assets are
m easured in term s of "dollars,” simply adding
the balances of various m onetary com ponents
would appear to be a natural approach to m ea
suring m onetary aggregates. Consequently, it is
not surprising that simple-sum m onetary aggre
gates have been used extensively throughout
the world. However, econom ic theory suggests
that various m onetary aggregate com ponents
differ in term s of their “liquidity” and, thus,
may have substantially different effects on eco
nom ic activity. If this is so, the simple-sum
procedure may actually be inappropriate for
m easuring “m onetary service flow s” in the na
tion. Instead, an alternative approach that in
volves calculation of Divisia indices may provide
superior alternatives to m easuring m onetary ag
gregates w hen com pared with the traditional
simple-sum m onetary m easures.
Theoretically, the Divisia index num ber is
derived from the econom ic aggregation theory
and first-order conditions for utility optimiza
tion. An expanded discussion of the Divisia ap
proach appears in appendix 1 of this paper. Em
pirically, the Divisia index num ber is estim ated
from a nonlinear function of the quantities and
the corresponding p rices of individual com
ponents that create the aggregate. M oreover, its
grow th rate is a linear combination of the
grow th rates of its com ponents, w here the
weights (or coefficients) on the com ponents are
their average ex p en d itu re shares. In com parison,
the simple-sum index is the linear sum of the
quantities of the components in which the weight
(coefficient) given to each com ponent is unity
and their prices have no effect on the index.
The grow th rate of the simple-sum index is also
a linear com bination of the grow th rates of its
’ See Barnett (1980).
2Use of Divisia monetary aggregates appears in Barnett
and Spindt (1982), Donovan (1978), Farr and Johnson
(1985), Belongia and Chalfant (1989), among others.
SEPTEMBER/OCTOBER 1991
22
com ponents, w here the coefficients are equal to
the quantity shares.3
D ifferences betw een the behavior of Divisia
and simple-sum aggregates stem from the dif
feren t weights assigned to the grow th rates of
the com ponents, w hich m easure their contribu
tions to the m onetary aggregates. Coefficients
called “shares” are expressed by the notations
S * a n d Sit for various com ponents used to d eter
mine the Divisia and simple-sum indexes, re
spectively, in the discussion that follows.
In calculating the Divisia index, the share of
each com ponent is the ratio of the expenditures
on the m onetary service flows it provides to the
total expenditure on m onetary service from all
com ponents in the aggregate; as such, it
represents an "expenditure share." In contrast,
shares fo r com ponents in the simple-sum index
are equal to the quantities of the balances held
in each com ponent divided by the total balances
of all com ponents in the aggregate. In general,
these tw o types of shares yield d ifferent values
and move diversely over time. For example, the
expenditure share of tim e deposits in Swiss M2
in January 1980 was 0.0856; its quantity share,
in contrast, was 0.3681. In January 1988, how
ever, the expenditure share of time deposits in
Swiss M2 w as 0.2823, while its quantity share
was 0.4527.
For the com ponents of simple-sum m onetary
aggregates, the only data required to compute
their respective shares are quantities of the
com ponents themselves. In contrast, for Divisia
aggregates, th e p rices of the com p onen ts—
w hich involve their interest rates—must also
be obtained in ord er to com pute their expen
diture shares.
T h e U ser Costs o f th e M o n eta ry
A ssets
As noted above, one m ajor problem involved
in computing Divisia index num bers for m one
tary aggregates is in determ ining the relevant
prices of the individual m onetary assets that
3By definition, the simple-sum aggregate is denoted such
that
SIMit = I mcit
dSIM/SIM, = I (mcit/SIMt)(dmcit/mcit)
where me; is the quantity of the i- th component. Thus,
approximately
ln(SIM,)-ln(SIM,_,) = IS it(in(mcit)-ln (m cit_1)),
FEDERAL RESERVE
BANK OF ST. LOUIS
make up the aggregate. In econom ic aggregation
theory, m onetary assets are treated as com mod
ities and their prices are defined similarly to
rental prices of durable goods. In this approach,
it is assumed that people receive m onetary ser
vices from holding m oney to finance their con
sumption. In doing so, they forego higher yields
typically available on other financial assets.
W hile m onetary services are considered con
sumed during some given period, money stock
(like any durable good) is not generally con
sumed during this period. Because m onetary
services are flow variables—n ot stock variables—
they should be evaluated by their rental prices
or u ser costs. T h erefore, Divisia index num bers
can be used to m easure the m onetary service
flows provided by various m onetary assets in
the econom y only if the u ser costs of these
assets can be correctly defined and accu rate
ly measured.
The appropriate u ser costs of m onetary assets
are based on m icroeconom ic theory and are
derived by examining the representative con
sum er’s optimal intertem poral consum ption pat
tern and m onetary asset portfolio allocation.4
These user costs are m easured as the oppor
tunity costs of foregone in terest associated with
holding funds in different types of m onetary
assets. The opportunity cost is obtained by com
paring each asset’s rate of retu rn to that on a
b en c h m a rk asset with the highest rate of return.
Under the relevant consum er theory, the b en
chm ark asset is assumed to provide no liquidity
or other m onetary services. Because it is held
only for accum ulating and tran sferrin g wealth
across time, its in terest rate is the highest in the
economy. Consumer theory, how ever, does not
specify other characteristics of the benchm ark
asset that would enable research ers to identify
the actual benchm ark asset to be used in em
pirical studies.
Barnett and Spindt (1982) have suggested that,
while hum an capital might best fit the theoreti
cal concept of the benchm ark asset, no satisfac
tory em pirical data exists on its rate of retu rn.
In their research , they found that
where Sit = mcit/SIMt is the quantity share of the i- th
component. This equation shows that the growth rate of
the simple-sum aggregate is a linear combination of the
growth rates of its components with weights equal to the
fractions of the quantities of the components to the simplesum aggregate.
4See Barnett (1978).
23
Figure 2
Selected Swiss Interest Rates
Percent
Percent
Rt = m ax[rbaa, r jt, i = 1,2, .. n]
many research ers have used the approach
adopted by Barnett and Spindt.
provided the best available proxy for the theo
retical benchm ark rate, w here r baa is Moody’s
series of seasoned Baa corporate bond rates and
rit is the own rate of retu rn on each of the com
ponents of L (the broadest U.S. m onetary ag
gregate defined by the Federal Reserve Board).5
Although Donovan (1978) used the nominal rate
of retu rn on "bonds” to com pute the rental
price of interest-bearing money fo r Canada,
In this paper, we use the Barnett-Spindt ap
proach to generate a proxy for the Swiss ben
chm ark asset rate (see appendix 2 for fu rth er
details). The benchm ark asset is either the long
term Swiss bond or short-term Euro-Swiss
deposits, depending on w hich yield is higher.
Thus, as shown in figure 2, the benchm ark
asset was long-term Swiss bonds b efore 1980,
Euro-Swiss deposits during 1980-1981 (due to an
5ln the user cost formula, R, is the maximum available yield
in the economy on any monetary asset which is a uniquely
defined theoretical maximum available yield. Empirically,
the proxy variable is defined by a long-term bond yield
relative to the rates of return on all monetary components.
The need for a long-term bond yield in measuring short
term holding-period yields is demonstrated by R. Shiller
(1979). The Baa bond rate is used as a representative
yield for long-term debts of average risk. See Barnett and
Spindt (1982).
SEPTEMBER/OCTOBER 1991
24
inverted term stru ctu re of Swiss interest rates
during this period), long-term Swiss bonds from
1982 to 1987 and Euro-Swiss deposits again
in 1988-1989.6
T he form ula fo r the real user costs of m one
tary assets in period t is expressed as
uit = (Rt - r it)/(l + Rt),
w here uit is the u ser cost of the i-th m onetary
asset, Rt is the nominal interest rate on the ben
chm ark asset and r itis the nominal interest rate
on the i-th m onetary asset in period t.7
COMPARISON O F THE BEHAVIOR
OF SWISS DIVISIA AND SIMPLESUM MONETARY AGGREGATES
Divisia m onetary aggregates and simple-sum
aggregates fo r M l and M2 w ere calculated us
ing the Swiss m onetary data described in appen
dix 2 .8 The monthly simple-sum and Divisia
m onetary aggregates are indexed to equal 100
in June 19 7 5 .9 Figures 3 and 4 show the 12m onth grow th rates of these aggregates and the
Swiss consum er price index from Ju n e 1976 to
D ecem ber 1989. W hile Divisia and simple-sum
M l indices display virtually identical grow th
rates in this period, the grow th rates of Divisia
and simple-sum M2 indices differ substantially
beginning in 1979 (see figure 4).
From 1980-1981, for example, the grow th of
simple-sum M2 rose rapidly while that of Divisia
M2 slowed markedly. From 1982 to 1983, in
contrast, the opposite pattern can be observed
in their respective grow th rates. In 1989, how
ever, the divergent pattern observed from 198081 occurs once again.
These widely divergent grow th rates over ex
tended periods fo r the simple-sum and Divisia
M2 suggest that discussions about the appropri
ateness of alternative procedures used to con
6Needless to say, such shifts in the benchmark asset raise
questions about the validity of this approach and have
resulted in criticisms of Divisia indices by a number of
economists. We do not address this issue in this paper.
H'he nominal user costs of monetary assets usually are ex
pressed as
Ui, = P, (R, —r„)(1 - t,)
1 + R ,(1 - t,)
where uitis the user cost of the i- th monetary asset in
period t, R, is the benchmark rate in period t, rit is the
nominal interest rate on the i- t h monetary asset; xt is the
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
struct m onetary aggregates are not m erely
"academ ic.” In general, the direction and mag
nitude of grow th in m onetary aggregates are
presum ed to provide useful inform ation about
the cu rren t stance of m onetary policy and the
future course of econom ic conditions. Such ex
trem e d ifferences in the grow th of alternative
M2 m easures (as shown in figure 4), however,
may produce considerable difficulty in assessing
that information.
S im p le S u m Vs. Divisia M o n eta ry
A g g re g a te s : W hat’s th e D if f e r e n c e ?
In Switzerland, M l consists of cu rren cy (C),
demand deposits with banks (DB) and demand
deposits with the postal giro system (DP). To
com pare simple-sum and Divisia M l, we calcu
lated the shares (Sit, S*t) fo r each com ponent of
M l. For m ost of the period, the respective
shares of each m onetary com ponent fo r simplesum and Divisia M l moved so uniform ly that
their respective contributions to these aggregates
are roughly equal. T h erefore, the grow th in
simple-sum and Divisia M l was essentially the
same over the sample period (as already noted
in figure 3).
In May 1989, how ever, the explicit interest
rate on demand deposits with the postal giro
system (DP) rose from zero to tw o percent,
reducing the user cost of DP, U3 (as shown in
figure 5). Since expenditure shares depend both
on quantities of the com ponents and their user
costs, DP’s share (S *) fell, reducing its weight in
calculating Divisia M l. T h erefore, since the in
troduction of explicit in terest paym ents on DP
had no effect on its weight in calculating
simple-sum M l, the values S * and S3 diverged
after May 1989. Since the values of S *a n d S3
are quite small, how ever, the difference b e
tw een the grow th of simple-sum and Divisia M l
after May 1989 is trivial.
Figure 5 shows that the u ser costs of the
th ree Divisia M l com ponents (U1-U3) follow
marginal income tax rate, and p , is the true cost-of-living
index used to deflate all nominal quantities to real quan
tities. Since taxes are not considered here, we use the
simplified formula.
W eak separability conditions should be satisfied first to
calculate a meaningful aggregate. However, we did not
conduct weak separability tests; instead, we used the ac
tual Swiss definitions of M1 and M2 and simply assumed
that they are admissible aggregates. For details, see
Belongia and Chalfant (1989).
9See appendix 2.
25
Figure 3
Inflation and M1 and Divisia M1 Growth
Percent
Figure 4
Inflation and M2 and Divisia M2 Growth
SEPTEMBER/OCTOBER 1991
26
Figure 5
User Costs of Currency, Demand
Deposits, PGS, and Time Deposits
wmmmm
similar m ovements from 1975 through 1989,
especially fo r the u ser costs of curren cy and de
mand deposits with banks. Thus, the relative
user costs for the Divisia M l com ponents are
nearly constant, making simple-sum M l as use
ful as Divisia M l over this period.10
Although the expenditure and quantity shares
w ere similar fo r the M l com ponents, low er
u ser costs fo r tim e deposits in M2 explain the
divergent patterns shown earlier betw een
simple-sum and Divisia M2. W e can illustrate
the im portance of changes in the econom ic en
vironm ent on the weights used by examining
the d ifferent behavior of Divisia and simple-sum
M2 over these tim e periods: Jan uary 1979 -
10This result is consistent with Hicks’ (Hicks, 1946) conclu
sion that “ when the relative prices of a group of com
modities can be assumed to remain unchanged, they can
be treated as a single commodity.”
FEDERAL RESERVE
BANK OF ST. LOUIS
D ecem ber 1981; Jan uary 1982 - November 1987;
D ecem ber 1987 - D ecem ber 1989. At the begin
ning of each period, th ere was a significant
change in Swiss m onetary policy as m easured
by sharp movements in the Swiss m onetary
base. During these periods, changes in the
econom ic environm ent w ere reflected in the
levels of short-term and long-term in terest rates.
As noted earlier (in figure 2), we used the
three-m onth Euro-Swiss Franc rate as the short
term rate, the Swiss governm ent bond yield as
the long-term rate and the benchm ark rate was
equal or close to the higher of these two rates
in any specific period. The grow th rate of the
Swiss m onetary base over these periods was
previously shown in figure 1.
27
Figure 6
Simple-Sum Share S4 and Divisia Share
SJ of Time Deposits and User Cost U4
Percent
T h e F irst P e rio d : J a n u a r y 1 9 7 9 to
D e c e m b e r 1981
During this period, the SNB’s response to ris
ing Swiss inflation was sharply slow er grow th
in the Swiss m onetary base (resulting in lower
inflation in 1983-84). The abrupt rise in short
term Swiss interest rates produced an inverted
yield curve for the next two years. The dramatic
increase in interest rates on tim e deposits re
duced their user costs (U4) to nearly zero as their
in terest rates approached the benchm ark rate
(see figures 2 and 5).
How did this interest rate movem ent affect
the m onetary aggregates? Asset holders shifted
from low er yielding securities into time deposits
(TD) causing the quantity of tim e deposits to in
crease dram atically and the quantity of demand
deposits w ith banks (DB) to decrease substantial
ly. Furtherm ore, asset holders also shifted funds
into time deposits from other financial assets
not included in M2. This sharp rise in the quan
tity of time deposits is indicated by the surge in
their share in simple-sum M2, S4 (shown in
figure 6).
These changes produced quite different r e
sults in the Divisia M2 m easure, how ever. As in
terest rates on time deposits increased relative
to other interest rates, time deposits had lower
opportunity costs and the m onetary service
flows from a given quantity of time deposits
naturally fell. Thus, despite the large increase in
time deposits, the expenditure share of the
m onetary service flows from tim e deposits ac
tually declined during this period (see S^in
figure 6), as did the grow th of Divisia M2 (see
figure 4).
T h e S e c o n d P e r io d : J a n u a r y 1 9 8 2
to N o v e m b e r 1 9 8 7
During this period, the rate of inflation de
clined to low er levels and an extended period of
SEPTEMBER/OCTOBER 1991
28
expansionary grow th in the m onetary base
began in Jan uary 1982. The term stru ctu re of
Swiss interest rates resum ed its norm al shape,
w ith short-term in terest rates below long-term
interest rates. Figure 5 shows that the user
costs of currency, demand deposits and demand
deposits with the giro system (U„ U2, U3) began
to fall in 1982. This reflects the fact that the
difference betw een their interest rates and the
benchm ark rate was declining, while the user
cost of time deposits (U4) increased as its in
terest rate fell relative to the benchm ark rate.
As noted previously, the contributions o f each
com ponent to the Divisia and simple-sum M2 ag
gregates are determ ined by their shares. W e on
ly display the quantity and expenditure shares
fo r time deposits (S * and S4) in figure 6 fo r illus
tration. In 1982, the quantity share of time de
posits (S4) fell substantially, while its expenditure
share (S*) rose sharply. This resulted in the pos
itive grow th of Divisia M2 and negative grow th
of simple-sum M2 shown in figure 4 for 1982.
Divergent movem ents in Divisia and simple-sum
M2 occu rred again in 1985 w hen the simple-sum
M2 growth was positive, while Divisia M2 growth
w as almost zero. During the rest of this period,
Divisia and simple-sum M2 moved similarly.
These th ree episodes suggest that, at least for
a m onetary aggregate as broad as Swiss M2, dif
feren t aggregation procedures produce m one
tary m easures that can move quite differently
and generate very distinct interpretations for
the stance of policy and the likely cou rse of
econom ic conditions. T h erefore, it is im portant
to know which o f the potential broader m one
tary aggregate m easures are m ore closely re
lated to key econom ic conditions.
COMPARISON O F TH E P E R F O R
MANCE OF TH E DIVISIA AND
SIMPLE-SUM AGGREGATES
To determ ine w hether a m onetary aggregate
can be used as a m onetary policy target, two
questions m ust be answered. First, is th ere a
satisfactory relationship betw een the m onetary
aggregate and some key econom ic variable, such
as inflation or nominal GDP or GNP? Second, is
the m onetary aggregate strongly related to
something that is directly controllable by the
m onetary authority? W e exam ine both these
questions in this section.
In fla tio n a n d M o n eta ry A g g re g a te s
T h e T h ird P e rio d : D e c e m b e r 1 9 8 7
to D e c e m b e r 1 9 8 9
Swiss inflation rose from 2 percent throughout
most of 1988 to nearly 5 percent by the end of
1989. W hile both short- and long-term interest
ra te s ro se over this period, sh o rt-term rates
rose relative to long-term rates. M oreover, the
m ajor institutional changes that took place re
duced demand fo r the m onetary b a se.11 In r e
sponse to these events, u ser costs of the first
two com ponents (U, and U2) rose, while the user
cost of tim e deposits (U4) fell (figure 5); these
movements w ere similar to those in the first
period. However, as m entioned earlier, the user
cost of demand deposits in the postal giro sys
tem (U3) fell in May 1989 w hen the interest rate
jumped from zero to two percent. Consequent
ly, the Divisia and simple-sum M2 m easures
moved in opposite directions; simple-sum M2
rose sharply, while Divisia M2 fell substantially
(figure 4).
"S ee the data description in appendix 2.
,2See Batten and Thornton (1983); Thornton and Batten
(1985).
Digitized for FEDERAL
FRASER RESERVE
BANK OF ST. LOUIS
To evaluate the first question, the relationship
betw een selected Swiss m onetary aggregates
and inflation are compared. Specifically, quarterly
Swiss inflation rates w ere regressed on distri
buted lags of selected Divisia and non-Divisia
m onetary aggregates. Because the sample is rel
atively small and because we would like to in
clude enough lags to capture the significant ef
fect of money grow th on inflation, the Polynom
ial Distributed Lag (PDL) estimation technique
was used.12
Ideally, it is desirable to use one of the com
monly used lag-length selection m ethods for
choosing both the lag length and the degree of
the polynominal. However, for two of the m one
tary aggregates, simple-sum M l and Divisia M l,
the equations exhibited significant serial correla
tion. This com plicates the application o f these
procedures fo r these aggregates. Because of this,
w hen these aggregates w ere used, several speci
fications of both lag length and polynominal
29
Table 1
PPL Regressions of Inflation on Money Growth
Growth Rate of
Growth Rate of
Simple-sum
Divisia
M1
M2
M1
M2
PDL order
1
1
Lag
1
1
Coefficients of Distributed Lags
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
0.00898
0.01232
0.01566
0.01900
0.02233
0.02567
0.02901
0.03235
0.03569
0.03903
0.04237
0.04571
0.04905
0.05238
0.05572
0.05906
0.06240
0.06574
0.06908
0.02740
0.02721
0.02702
0.02683
0.02664
0.02646
0.02627
0.02608
0.02589
0.02570
0.02552
0.02533
0.02514
0.02495
0.02476
0.02458
0.02439
0.02420
0.02401
0.00917
0.01257
0.01597
0.01938
0.02278
0.02618
0.02958
0.03298
0.03639
0.03979
0.04319
0.04659
0.04999
0.05339
0.05680
0.06020
0.06360
0.06700
0.07040
-0.01316
-0.00653
0.00010
0.00672
0.01335
0.01998
0.02661
0.03323
0.03986
0.04649
0.05311
0.05974
0.06637
0.07299
0.07962
0.08625
0.09287
0.09950
0.10613
Sum
t-Stat1
0.74154
(3.1168)
0.48838
(8.1203)
0.75596
(3.1877)
0.88322
(6.0924)
AR(1)
S.E.
Adj-R2
D-W
0.4673
0.00697
0.5751
1.9811
0.0763
0.00611
0.6729
1.9409
0.4596
0.00695
0.5776
1.9773
0.2253
0.00625
0.6579
1.9249
S.E. is the standard error of the regression.
1 The critical t-statistic value at the 5 percent significance level is 1.645.
degree w ere estimated. Specifications with rela
tively long lags and relatively low polynominal
degrees produced the highest adjusted R-square.
To keep the specifications comparable with those
o f simple-sum and Divisia M2 (which w ere cho
sen using the FPE criteria), results with lag
lengths of 18 and first-degree polynominals are
presented.13
To com pare the long-run relationship of the
m onetary aggregates on inflation, we estimated
the selected PDL models and computed the sum
of coefficients of the distributed lags to test
w hether the sum of the lagged money growth
coefficients was significantly different from
z ero . The estim ated coefficients and respective
statistics are shown in table 1.
The results in table 1 show that m onetary ag
gregates influenced Swiss inflation over periods
o f up to fou r or m ore years. W ith the exception
o f simple-sum M2, the m onetary aggregates
have roughly similar values for the sum of the
coefficients on th eir distributed lags. T he hypo
thesis that the sum of the lag coefficients is
13The identical polynominal degree was obtained for M2 and
Divisia M2 by the Pagano-Hartley technique if a T-statistic
of 2.0 is used for the critical value.
SEPTEMBER/OCTOBER 1991
30
Table 2
F-Statistics for PDL Models1
Growth
Divisia M2
Divisia M2 and
rates
and M1
Divisia M1
M2 and
M1
M2 and
Divisia M1
F-Stat.
H,
h2
1.8349
7.3115
1.8349
7.1715
0.0605
4.0645
0.0726
3.9585
'At the 5 percent significance level, the critical F value is 3.29.
zero was rejected in all cases at the 5 percent
significance level.14
Because M l and Divisia M l are included in
th eir respective M2 counterparts, we investi
gated w h eth er the non-M l com ponents of M2
them selves added significant explanatory pow er
in the inflation equation. Thus, we defined the
u n restricted model as regressions of inflation on
both the M2 and M l PDLs; the restricted models
w ere those with regressions of inflation on the
M2 or M l PDLs, respectively. The actual F-statistics and their 5 percent significance level
critical values are shown in table 2. H, is the
hypothesis that inflation can be explained by
Divisia M2 or M2 alone; H2 is the hypothesis
that inflation can be explained by M l aggregates
alone. These hypotheses are tested against the
corresponding u nrestricted PDL model that in
flation is explained jointly by simple-sum and
Divisia M2 PDL an d the M l aggregates PDL.15
The results in table 2 show that the data failed
to reject H„ but did reject H2. This result sug
gests that the broad er aggregates M2 and
Divisia M2 b etter explain Swiss inflation than
does M l or Divisia M l alone.
C ontrollability
As noted earlier, the practical use of a m one
tary aggregate as an intermediate target depends
on its controllability. Even if some m onetary ag
gregate shares a close relationship with inflation
14The hypothesis that the sum of the lag coefficients was
unity failed to be rejected in all cases except for simplesum M2. Thus, except for simple-sum M2, the results did
not reject a one-to-one long-run relationship between the
growth of the monetary aggregate and the rate of inflation.
The results for the monetary base are similar to those of
the Swiss monetary aggregates; its adjusted R2 (0.4880)
was lower than those displayed in table 1. As shown in the
first part of this paper, the relationship between inflation
FEDERAL RESERVE BANK OF ST. LOUIS
or nominal GDP, it would be of little use as a
m onetary target if its grow th can not be con
trolled by m onetary authorities. Since the cen
tral bank controls the m onetary base, the rela
tionships between it and the broader Swiss mone
tary aggregates are examined.
Because the cross correlations betw een the
grow th rates of m onetary aggregates and the
grow th rate of the m onetary base show a longlag pattern, we used the PDL models to estimate
their relationships. Again, because of significant
serial correlation, various specifications of lag
length and polynominal degree w ere estimated.
However, they share the same qualitative pro
perties. Table 3 displays the estim ates of the
16-lag and first-degree PDL models.
Results show that the grow th rates of M l,
Divisia M l and Divisia M2 are statistically signif
icant in relation to the grow th rate of the
m onetary base. However, a significant long-term
relationship betw een the grow th rate of the
m onetary base and simple-sum M2 is rejected at
the 5 percent level of significance.
In addition, the contem poraneous grow th rate
of the m onetary base is positively and signifi
cantly correlated to the grow th rates of both
M l aggregates and the Divisia M2 aggregate.
For simple-sum M2 grow th, how ever, the con
tem poraneous and initial lagged grow th rates of
the m onetary base have negative effects on its
and growth of the monetary base slipped considerably dur
ing 1986-89 (the end of the period examined); earlier,
however, there had been a close link between Swiss infla
tion and long-run growth in the monetary base. Because
the estimation period covers 1975-89, the recent
"breakdown” in the monetary base growth-inflation rela
tionship does not dominate the results.
15For both M2 and M1 aggregates, we take the same
number of lags, 18, in the PDL models.
31
Table 3
PDL Regressions of Growth Rates for Monetary Aggregates on
Growth Rate of Monetary Base____________________________
Growth Rate of
Growth Rate of
Simple-sum
Divisia
M1
M2
M1
M2
PDL order
1
1
1
1
Lag
Coefficients of Distributed Lag for the Growth Rate of the Monetary Base
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
0.39216
0.36648
0.34079
0.31511
0.28942
0.26374
0.23805
0.21237
0.18668
0.16100
0.13531
0.10963
0.08394
0.05826
0.03257
0.00689
-0.01880
-.07127
-.05610
- .04092
- .02574
-0.01057
0.00461
0.01979
0.03496
0.05014
0.06532
0.08049
0.09567
0.11085
0.12602
0.14120
0.15637
0.17155
0.38461
0.35937
0.33414
0.30890
0.28367
0.25844
0.23320
0.20797
0.18273
0.15750
0.13227
0.10703
0.08180
0.05656
0.03133
0.00610
-0.01914
0.20584
0.19467
0.18351
0.17235
0.16118
0.15002
0.13886
0.12769
0.11653
0.10537
0.09420
0.08304
0.07188
0.06071
0.04955
0.03839
0.02722
Sum
t-Stat
3.17360
(2.3518)
0.85237
(0.5770)
3.10648
(2.3538)
1.98099
(3.3068)
AR(1)
S.E.
Adj-R2
D-W
0.57811
0.01852
0.42888
1.97733
0.55397
0.02317
0.48930
1.98227
0.57459
0.01821
0.42632
1.98631
0.40863
0.01137
0.40400
2.12645
S.E. is the standard error of the regression. The critical t-statistic for the 5 percent significance level
is 1.645.
growth. Indeed, significant positive correlation
shows up only th ree years after the changes in
the m onetary base.
broader m onetary aggregates as m onetary
policy targets.
CONCLUSION
This paper examined the potential usefulness
of Swiss M l and M2 m onetary aggregates com
pared with Swiss Divisia M l and M2 aggregates
derived from econom ic aggregation theory. We
showed that M l and Divisia M l generally dis
played similar movements over time and w ere
related similarly both to Swiss inflation (which
justifies their potential usefulness as a target)
and to the m onetary base (which means that
their grow th was potentially controllable by the
Swiss National Bank).
The relationship betw een the Swiss m onetary
base and inflation in Sw itzerland has becom e
more uncertain in recen t years. This phenom e
non has generated considerable interest in using
M2 and Divisia M2, however, displayed sub
stantially different behavior over time and, at
certain key times, yielded substantially different
signals about the stance of m onetary policy.
These results suggest that the Swiss National
Bank can significantly influence the grow th of
M l, Divisia M l and Divisia M2 through changes
in the grow th of the m onetary base. Long-term
simple-sum M2 grow th, how ever, does not ap
pear to be influenced by grow th of the m one
tary base.
SEPTEMBER/OCTOBER 1991
32
More im portantly, M2 grow th was statistically
unrelated to the grow th of the m onetary base.
The results suggest that M l, Divisia M l and
Divisia M2 would be suitable fo r fu rth er study
if the Swiss National Bank is interested in the
possibility of using broader m onetary aggre
gates to replace m onetary base targeting. How
ever, these results indicate that M2 is unlikely
to provide an adequate substitute fo r m one
tary policy purposes.
REFEREN CES
Barnett, William A. “ The User Cost of Money,” Economics
Letters, no. 2, 1978, pp. 145-49.
_______ . “ Economic Monetary Aggregates: An Application
of Index Number and Aggregation Theory,” Journal of
Econometrics: Annals of Applied Econometrics 1980-3,
supplement to the Journal of Econometrics (September
1980), pp. 11-48.
Barnett, William A., and Paul A. Spindt. “ Divisia Monetary
Aggregates: Compilation, Data, and Historical Behavior,”
Staff Study no. 116 (Board of Governors of the Federal
Reserve System, May 1982).
Batten, Dallas S., and Daniel L. Thornton. “ Polynomial
Distributed Lags and the Estimation of the St. Louis
Equation,” this Review (April 1983), pp. 13-25.
Belongia, Michael T., and James A. Chalfant. “ The
Changing Empirical Definition of Money: Some Estimates
from a Model of the Demand for Money Substitutes,” Journal
of Political Economy (April 1989), pp. 387-97.
Diewert, W. E. “ Exact and Superlative Index Numbers,”
Journal of Econometrics (May 1976), pp. 115-45.
Donovan, Donal J. “Modeling the Demand for Liquid Assets:
An Application to Canada,” International Monetary Fund Staff
Papers, Vol. 25, no. 4, (December 1978), pp. 676-704.
Farr, Helen T., and Deborah Johnson. “ Revisions in the
Monetary Services (Divisia) Indexes of the Monetary Ag
gregates,” Staff Study no. 147 (Board of Governors of the
Federal Reserve System, December 1985).
Friedman, Milton, and Anna J. Schwartz. “Monetary
Statistics of the United States: Estimates, Sources, Methods,”
New York, Columbia University Press, 1970.
Hicks, John Richard. Value and Capital, 2nd ed. (Oxford,
Eng.: Clarendon Press, 1946).
Rich, Georg. “Swiss and United States Monetary Policy: Has
Monetarism Failed?” Federal Reserve Bank of Richmond
Economic Review (May/June 1987), pp. 3-16.
________“Geldmengenziele und schweizerische Geldpolitik:
eine Standortbestimmung. In: Schweizerische Nationalbank:
Geld, Wahrung und Konjunktur,” Quartalsheft No. 4 (Dezember 1989), pp. 345-60.
Shiller, Robert J. “The Volatility of Long-Term Interest Rates
and Expectations Models of the Term Structure,” Journal of
Political Economy, (December 1979), pp. 1190-219.
Schweizerische Nationalbank. Revision der Geldmengenstatistik. In: Geld, Wahrung und Konjunktur. Quartalsheft No.
1, Marz 1985.
Thornton, Daniel L. and Dallas S. Batten. “ Lag-Length Selection
and Tests of Granger Causality between Money and Income,”
Journal of Money, Credit and Banking, (May 1985), pp. 164-78.
Appendix 1
Divisia Indexes
T h ere are two types of Divisia index numbers:
the continuous-time version and the discrete
time version. Continuous-time Divisia index
numbers are derived from microeconomic theory;
discrete-time Divisia index numbers are approxi
mations of the continuous-time version.
To understand how continuous-time Divisia in
dex numbers are derived, consider the case where
economic agents want a measure that aggregates
a group of n commodities in the economy. The
quantities of the goods are expressed by the vec
tor q=(q„ q,, ... , qn); their corresponding prices
are denoted by the vector p = (p„ p2,..., pn).
Economic aggregation theory states that the
aggregator function is a utility function g(q) to be
maximized subject to the budget constraint,
first-order necessary condition for utility maximi
zation is
(2) dg(q)/dq = A pi(
where A is the Lagrange multiplier. Because the
aggregator function is linear homogeneous, Eul
er’s equation is satisfied such that
(3) Z (dg/dq)q = g(q).
i~l
Substituting equation 2 into equation 3 yields
Ai=11 q.Pi = g<q) and
n
A E = g(q).
Hence, A = g(q)/E and
(i) 1£=1 qiPi = g(q)flp) = E,
(4) dg(q)/dq = Pig(q)/E.
where f(p) is the price aggregator function and E
is the total expenditure on the specific goods. The
Taking the total differential of the aggregator
function g(q) yields
FEDERAL RESERVE BANK OF ST. LOUIS
33
(5) dg(q) = I (dg/dq,)dq,
i=l
Thus, substituting equation 4 into 5 yields
time, the continuous-time Divisia index must be
transform ed into a discrete-time version to make
it useful. The discrete time approximation of equa
tion 7 is
n
(6) dg(q) = 1 pdqg(q)/E and
i=l
n
dg(q)/g(q) = I
(9) ln(g(t))-In(g(t-l)) = I S'flnlqlt)) - ln(q(t-l))]
i=l
pqdq/Eq,
1=1
If we set Sj = pq/E and call it the i-th good’s
value share in the total expenditure, equation 6
can be transform ed into
(10) g(t) = g(t-l)EXp( I S^n(q(t)) - ln(q(t-l))]l
' i=l
'
(7) d(ln(g(q))) = S Sjddnfq,)).
i=l
where S*t = (Sit + Sit.,)/2,
Solving equation 7 for g(t) yields
n
Sit = pq/E = p.q/ ( I
1= 1
(8) g(t) = EXP/ ( £ S,(t)clfln(qi(t))l) dt.
xi=l
'
Equation 8 is the continuous-time Divisia index.
Because economic variables are observed and
measured in discrete time rather than continuous
p ,q ),
and S*t is the average expenditure share in the
two adjacent time periods. Equations 9 and 10 are
the discrete-time Divisia index equations used in
calculating the Swiss Divisia M l and M2
monetary aggregates.
Appendix 2
Data
To calculate the Swiss Divisia M l and M2 mone
tary services indexes, we used the seasonally
adjusted monthly Swiss M l and M2 series and
their components consistent with the definitions
established in 1975 and incorporating the revision
that occurred in 1985 (for m ore details, see
Schw eizerische Nationalbank, 1985).
The monetary aggregates consist of the fol
lowing assets held by individuals and non-bank
institutions:
M l:
Currency in circulation (C)
Demand deposits with banks (DB)
Demand deposits with the postal giro
system (DP)
M2:
M l plus time-deposits (TD)
MB:
Seasonally adjusted monetary base,
defined as the sum of banks reserves and
banks notes in circulation.
In terest R ates
To compute the user costs of monetary assets,
we need the assets’ own rates of return, the
benchmark rate of return and the cost-of-living
index.
Own R ates
c
Zero
DB
DP
0.25 percent
1975:06 - 1989:04: Zero
1989:05 - 1989:12: 2 percent
TD
three-month rate on time-deposits with
large banks (monthly average)
Cost-of-living In d ex
Monthly Swiss Consumer Price Index (CPI)
calculated by the Federal Statistical Office
(Bundesamt fuer Statistik).
B e n c h m a rk R ate
The highest rate in each period from the fol
lowing interest rates: the secondary market yield
on cantonal bonds, interest rates on cash cer
tificates with the cantonal or large banks and
short-term Euromarket-Swiss franc interest rates.
Short-term rates became the benchm ark rates
during 1979:12 - 1982:04 and 1988:12 - 1989:12,
when the Swiss yield curve was inverted.
SEPTEMBER/OCTOBER 1991
34
Steven Russell
Steven Russell is an economist at the Federal Reserve Bank of
St. Louis. Lynn Dietrich provided research assistance.
The U. S. Currency System:
A Historical Perspective
i
HE USE OF CURRENCY in transactions is a
regular part of our daily lives and a basic feature
o f o u r eco n o m ic system . T h e im p o rta n ce of
cu rren cy derives both from its obvious role in
daily transactions and from the somewhat m ore
subtle role of the cu rren cy sy stem as the basis for
our m onetary and financial systems. The curren cy
system is so fundam ental to econom ic activity that
we tend to give it little thought. Few of us would
have an easy tim e imagining w hat alternative
system s m igh t be like o r w hy th e y m ight be
desirable. Indeed, it seems likely that most of us, if
pressed, would offer the opinion that the present
cu rren cy system is the only one that is feasible—or
at least, the only one that is desirable.
This article has th ree purposes. T he first is to
define the term “cu rren cy ” and explain the special
im portance of cu rren cy and the cu rren cy system
to our economy. The second is to describe the U.S.
cu rren cy system —the system that governs the
form s, uses and roles of curren cy in the m odern
United States. This description will be preceded by
a catalog of the form s curren cy has taken at vari
ous points in the past, so that the m odern U.S.
system em erges as a set of selections from a menu
of choices provided by history. This procedure is
intended to suggest that alternative m enu selec
tions w ere possible—that the curren cy system
which actually evolved in the United States is not
the only one that could have evolved. The article's
third and most ambitious purpose is to present a
b rief but com prehensive account of the historical
1U.S. monetary history from the end of the Civil War through
modern times has been chronicled quite extensively, notably
by Friedman and Schwartz (1963). In addition, most of the key
FEDERAL RESERVE
BANK OF ST. LOUIS
development of the U.S. cu rren cy system. This
account focuses on the period b efore and during
the Civil W ar.1 Its prim ary goal is to provide the
reader w ith historical context that may improve
his understanding of the m odern cu rren cy system.
The historical account has a second purpose,
however. The development of the U.S. curren cy
system is often characterized as a process of slow
but steady advancement: older institutions and
practices, having failed to m eet the demands of
th e ir tim es, w e re re p la ced by m o re e ffic ie n t
successors. This "gradual progress” ch aracteriza
tion implies that the m odern cu rren cy system
m eets the needs of our econom y m ore efficiently
than could any of the alternatives suggested by
history. The historical account is intended to help
determ ine w h eth er this ch aracterization is valid,
and w h e th e r rela tiv e e ffic ie n c y co n clu sio n s
should be based on it.
WHAT IS CURRENCY?
One approach to defining cu rren cy is to contrast
it w ith so m eth ing w h ose d efin itio n is clo sely
related, but m ore fam iliar: money. Most people
have been exposed at some point to an econom ist’s
definition of money; it usually reads something
like “things that serve as media of exchange” or
"things that function as means of paym ent.” While
all cu rren cy is money, all money is not currency.
Currency can be defined as m oney which circu-
decisions that determined the basic form of the U.S. currency
system were arguably made before 1865.
35
lates,
or passes from hand to hand. ("Circulation”
w as on ce com m only used as a synonym fo r
currency.)
Formally, a type of money can be said to circu
late if it usually p asses in e x ch a n g e fro m one
person to another w ithout third-party verification.
One easy way to illustrate the d ifference betw een
circu la tin g and n o n -circu la tin g m on ey is to
contrast dollar bills, w hich circulate, w ith checks,
which do not.2 A dollar bill may pass from one
person to another many tim es in d ifferent tran sac
tions. The only people involved in each transaction
are th e b u y e r and se ller. T ra n sa c tio n s using
checks require m ore complex arrangem ents. It is
unusual for a check, w ritten by one person in
paym ent to another, to be offered in paym ent to a
third person. Instead, the second person usually
deposits the check in a bank account. His bank and
the first person’s bank then conduct a “clearing”
transaction w hich, if successfully completed, vali
dates the paym ent.3
example, commodities such as wampum (colored
beads), tobacco, w heat and rice w ere used as
curren cy at different places and tim es.4 Gold and
silver, th e “p re cio u s m e ta ls ,” had a ttra ctiv e
p ro p e rtie s — p o rta b ility , m alleab ility and d u r
ability—w hich ultimately made them the cu rren
cies of choice in most early economies.
Coin C u r re n c y
As the volume of transactions involving gold and
silver increased, people began to divide these
metals into pieces of readily recognizable size and
shape, called coins. T he earliest coin-producing
facilities (mints) seem to have been privately oper
ated.5 In m ost countries, how ever, the govern
m ent eventually took over coin production.
Commodity C urrency
The rationale behind the governm ent takeover
may well have included the belief that govern
ment-issued coins would be m ore uniform , and
m ore reliable, than their privately issued cou n ter
parts. Early governm ents, however, could have
reso lv ed p ro b lem s o f d iv ersity and fra u d by
regulating private mints and inspecting private
coins, in essentially the same way that govern
ments have long regulated and inspected other
industries.6 A m ore compelling reason fo r govern
m ent coin monopolies, how ever, was the desire to
earn revenue from seign iorage—from periodically
shortweighting or debasing the cu rren cy .7 Unless
a government had a coinage monopoly, its attempts
to earn substantial revenues from seigniorage
would have been frustrated as the public aban
doned its coins in favor of those minted by its
private competitors.
The earliest form s of cu rren cy w ere c o m m o d i
ties (widely traded goods). In colonial America, for
T h e p re v a len ce o f g ov ern m en t c u rre n c y
monopolies gave rise to the twin concepts of a
In the m odern United States, only dollar bills
and coin s, issu ed by a g en cies o f th e fe d e ra l
governm ent, fit the definition of currency. Earlier
in our history (and that of many other nations) the
n u m b er of a lte rn a tiv e typ es o f c u r re n c y w as
la rg e r, and in clu d ed item s issu ed by p riv ate
organizations. The next section presents a brief
catalog of some of the varieties of cu rren cy that
have existed in the past.
WHAT FORMS CAN CURRENCY TAKE?
2Strictly speaking, economists think of the accounts against
which checks are drawn (the demand deposits) as money,
rather than the checks themselves.
3Typically, a person who is offered a newly written check in
payment (the second party) will ask the check-writer (the first
party) to present identification and will record information
from the identification presented. The second party will
deposit the check in his bank account. His bank will “ clear”
the check by sending it to the bank against which it is drawn,
and demanding payment in cash. The two banks are the
“ third parties” which are actually involved in most transac
tions using checks. The clearing transaction is necessary to
verify that the check is drawn on an account that contains
sufficient funds. If the check “ bounces,” it has failed the
verification test. The amount of the check will not be credited
to the second party’s account, and he will use the information
from the check-writer’s ID to pursue him for some alternative
form of payment. [The reason the second party will rarely try
to pass the check along to a third party is that the third party is
unlikely to accept it. (If you doubt this, try passing such a
“ third-party check” at your local grocery store.) A third party
typically has no easy way of obtaining reliable identification
from the (absent) first party.]
4For an extended discussion of the role of commodity money in
the colonies, see Nettels(1934), chapter VIII.
5Feavearyear (1963) describes early English currency as
follows: “ At the beginning of the eighth century the currency
consisted of small silver coins varying in design according to
the fancy of the individual moneyer.” (p. 7)
6Adam Smith (1776/1937) points out that before coins evolved,
governments often stamped ingots of precious metal to certify
their purity (pp. 24-25).
7A coin is said to have been “ shortweighted” if it is minted
with less than its official metallic weight, but represented as
having exactly that weight. A coin is said to have been
“ debased” if it is minted as a mixture of genuine monetary
metal and common scrap metal, but represented as pure
monetary metal. These fraudulent practices were sometimes
practiced by private mints as well. For a discussion of govern
ment seigniorage motives, see Timberlake (1991), pp. 3-5,
50-51.
SEPTEMBER/OCTOBER 1991
36
national cu rren cy and a national m onetary unit.
Typically, a governm ent would define a basic
m onetary unit as a fixed quantity of gold or silver.
It would then mint coins in d en om in ation s that
w ere multiples or fractions of this unit and w ere
scaled appropriately in size and weight.
Most nations had an extended period during
w hich governm ent-issued coins w ere the only
form of currency. One problem with these purecoin cu rren cy systems was that they had difficulty
handling transactions of widely differing scales. If,
for example, coins w ere denominated so that a
single coin of m oderate w eight could be used to
purchase an inexpensive item (say, an apple), then
the coins necessary to purchase an expensive item
(say, a carriage) w ere necessarily quite heavy. One
com m on way in w hich governm ents tried to solve
this problem was by establishing bim etallic coinage
system s. In th ese sy stem s coin s o f low value
contained a relatively inexpensive m etal (typically
silver), while larger-value coins w ere com posed of
a m ore expensive m etal (typically gold). The two
types of coins w ere referred to collectively as
sp ecie.
The U.S. experience w ith specie cu rren cy illus
trates most of the concepts just described. The
U.S. Constitution gave Congress the pow er to
“coin money, and regulate the value th ereo f”—a
provision w hich has been universally interpreted
as prohibiting the states eith er from minting coins
directly or from authorizing private parties to do
so.8 Shortly after the Constitution was ratified,
Congress enacted legislation that defined the basic
m onetary unit, the dollar, as either a fixed weight
of gold or a (different) fixed weight of silver. The
fed era l g o v ern m en t th e n opened a m int th at
produced dollar coins in accordance w ith these
definitions. The mint also produced silver “quar
te rs” containing one-fourth the amount of silver in
a silver dollar, five-dollar gold pieces containing five
times the amount of gold in a gold dollar, and so on.9
The U.S. Mint continued to produce full-bodied gold
coins until the early 1930s, and full-bodied silver
coins until the mid-1960s. (A full-bodied coin con
tains a quantity of m etal whose m arket value is
equal to the face value o f the coin.)
8U.S. Constitution, Article I, Section 8.
9See Huntington and Mawhinney (1910), pp. 474-79. Gold
dollar coins were not actually minted until 1849. See
Carothers (1930), pp. 105,109, and Huntington and
Mawhinney (1910), pp. 508-09.
FEDERAL RESERVE
BANK OF ST. LOUIS
As the magnitude o f econom ic activity increased,
the weight of the gold coins necessary for a m ajor
purchase, or even the quantity that a relatively
wealthy person might desire to have on hand,
becam e unm anageably large. Coins also tended to
w ear away or have their edges clipped. A fter a
few years, coins of the same denomination could
be significantly different in size.10 These problem s
made coins increasingly unsatisfactory, even for
relatively small-scale transactions.
Bills o f E x c h a n g e
An obvious solution to the “weight problem s” of
the coin cu rren cy system was to find or create
lightweight objects that, while not made of coins
themselves, had know n values in term s of coins.
Objects like this already existed: they w ere p ro m is
so ry n otes—contracts betw een b orrow ers and
lenders calling for the repaym ent of fixed sums (in
coin) at fixed future dates.
One special type of prom issory note, the bill o f
exchange, was readily adapted fo r use as currency.
Bills of exchange grew out of com m ercial tran sac
tio n s in w h ich m e rc h a n ts w ou ld a rra n g e to
purchase goods from other m erchants for delivery
at fixed future dates (for example, in 90 days).
Often the seller could not afford to produce and/or
deliver the goods unless he received immediate
payment, while the buyer was reluctant to pay for
the goods b efore receiving delivery. One solution
to this problem was an exchange of contracts. The
seller would con tract to deliver the goods at the
date in question, while the buyer would con tract
to pay the purchase price at the delivery date. The
latter con tract took the form of a conventional
prom issory note.
This exchange of contracts may not seem to
have addressed the seller’s immediate problem : to
obtain the cu rren cy needed to finance the produc
tio n and/or tra n s p o rt o f his goods. Suppose,
however, that the seller, armed w ith his prom is
sory note, sought to purchase m aterials from a
supplier. He could then w rite out another credit
instrum ent—a bill of exchange—calling on the
m erchant who had issued the prom issory note to
pay th e su p p lier th e p u rch a se p rice o f th e m a
te ria ls, plus an allo w an ce fo r in te re s t, in 90
10For a description of the clipping problem in pre-eighteenth
century England, see Feavearyear (1963), pp. 5-6, and
Macaulay (1877), volume V, pp. 85-93.
37
days. This process was called draw ing a bill; the
original goods seller was called the d ra w er and the
issuer of the prom issory note the d raw ee. The
drawee would accep t (agree to cover) the bill as
long as its value was less than that of the prom is
sory note. He would indicate his acceptance by
endorsing the bill.11
By accepting the bill, the supplier was, in effect,
lending the seller the value of the m aterials the
latter had “purchased.” The supplier, how ever,
usually did not expect to hold the bill until it came
due. Instead, he planned to pass it along to some
one from whom he wished to purchase goods; this
person might pass it along to someone else, and so
on, until the bill m atured. The last person in the
chain would demand payment from the drawee.
In betw een, the bill served as paper cu rren cy .12
Notice that drawing a bill was analogous to
w riting a check, with the draw ee of the bill
playing the same role as that of a bank on which a
check is drawn. It seems to follow that bills of
exchange should not have circulated, for precisely
the same reasons that m odern checks do not
circulate. T h ere was a basic d ifference betw een a
bill o f exchange and a m odern personal check,
however. Because no one would accept a bill
unless it was endorsed by the relevant drawee, the
question of "bad ch ecks”—checks w ritten by
individuals with insufficient funds—did not arise.
Stated differently, an accepted bill was purely a
liability of the drawee; a person offered a bill in
paym ent did not need to be concerned about the
creditw orthiness of the d raw er.13
It is true that bills w ere occasionally d ish o n o red
by their drawees, just as m odern banks occasion
ally fail.14 As long as the bills w ere drawn against
well-known m erchants with established reputa
tions in com m erce, how ever, failures w ere un
common. Consequently a person who accepted a
bill in paym ent could be reasonably confident
11See Clough and Cole (1941), pp. 77-78.
12ln England, bills of exchange played a prominent role as
means of payment during the seventeenth and eighteenth
centuries; see Feavearyear(1963), pp. 160-62. In several
English districts, they retained this role well into the
nineteenth century; see Clapham (1944), II, pp. 90-91, 97-98,
Viner(1937), p. 123, and Feavearyear(1963), p. 165.
13lt is worth noting that while it is usually difficult to negotiate a
third-party check drawn on an individual's bank account, a
check drawn on the account of a government agency or
prominent local corporation may be easy to negotiate.
that, if he did not have occasion to pass it along, he
could redeem it w hen it came due.15
Although bills of exchange becam e an im portant
adjunct to coin currency, a num ber of problems
limited their usefulness. Since they w ere typically
drawn in fairly large denominations (of the sort
appropriate fo r trade betw een m erchants), they
w ere not well suited fo r small-scale transactions.
And, as the volume of trade in a given region
increased, it becam e less and less likely that a
person proffered a bill would be familiar, either
personally or by reputation, with the m erchant
against whom it was drawn. Consequently, disho
nored bills becam e a m ore serious problem , and
people becam e hesitant to accept them in payment.
A less fundamental, but still annoying, problem
was that w henever a bill changed hands, interest
had to be calculated and deducted from its face
value. This fairly involved calculation required
consideration of both the rem aining term on the
bill and the m arket rate of interest.
B a n k N otes
The transactions problem s w ith bills of exchange
created opportunities for private entrepreneurs to
profit by providing paper cu rren cy in m ore con
venient form s. Suppose an enterprising m erchant
with a good reputation sold small bills of exchange
in retu rn fo r specie and used the proceeds to buy
large bills with the same m aturity dates. The
proceeds of the large bills would then provide a
fund out of w hich the small bills could be
redeemed. Because small bills w ere m uch m ore
convenient for exchange purposes than large bills,
they w ere slightly m ore valuable, per dollar of
face value, to their holders.16 As a result, small
bills could be sold at smaller percentage discounts
(lower interest rates) than large bills. It followed
that the total purchase price of the large bills
necessary to cover a given face value of small bills
was smaller than the total sale price of the small
bills. This d ifference in total prices represented
the m erch an t’s profits.
14The existence of federal deposit insurance prevents smallscale personal depositors from being endangered by bank
failures. Before 1935, however, this was not true.
1Contemporary criminals sometimes forged merchants’
acceptances, just as modern criminals sometimes forge
checks. The severity of penalties for forgery limited the scale
of this problem, however.
16For a careful description of the logic behind this statement,
see Wallace (1983).
SEPTEMBER/OCTOBER 1991
38
The m erchant had now becom e a ban ker, and
the institution he operated a b a n k o f issu e—a
financial interm ediary whose liabilities consisted
prim arily of paper cu rren cy .17 The small bills
cam e to be know n as b a n k notes.
The m erchant could increase his profits from
note issue by reducing the risk that he would
default on his notes. This would reduce the “risk
prem ium ” that small billholders demanded, and
enable him to sell the bills at smaller discounts.
One strategy fo r accom plishing this was to diver
sify h is la rg e b ill p o rtfo lio as ex ten siv ely as
possible. A nother was to provide, or to obtain
from investors, some capital to act as a cushion
against defaults on the large bills.
A basic problem w ith the schem e just described
was that the tim e and effort necessary to compute
th e a p p ro p ria te d iscou n t on a b an k n o te w as
usually large relative to the face value of the note.
This reduced the usefulness of notes in tran sac
tions and discouraged people from purchasing
them .18 One way to solve this problem was to issue
notes with characteristics so appealing that their
h o ld ers w ould b e w illing to fo rg o in te re s t on
them . How could this be accomplished?
Since we have assumed (perhaps too quickly)
that proper diversification and capitalization made
the risk on bank notes negligible, the need for
interest on them arose purely out of their holders'
tim e p r e fe r e n c e —th eir desire to be com pensated
for giving up their money (in this case, their specie
currency) for fixed periods. Suppose, however,
that a m erchant prom ised to redeem his notes on
d em a n d (at any time) instead of at a fixed future
date. Since the pu rchaser of such notes could
reclaim his specie w henever he chose, he would
not be giving it up for any fixed period, and would
have no reason to demand interest. The bills could
then be sold at p a r (undiscounted).
How could a m erchant make such a convertibility
com m itm ent credible? Clearly, he would need to
hold back some of the (specie) proceeds of his note
sales fo r use as reserv es. These reserves would not
have to be large, how ever, because as long as note
holders w ere confident that they cou ld redeem
17For a description of banks that dealt in bills of exchange, see
Feavearyear(1963), pp. 162-65.
,8White (1987) provides an analysis of the transactions
problems associated with interest-bearing currency.
19Calomiris and Kahn (1991) construct a formal model in which
note- or deposit-holders can use redemption demands as a
device for preventing bank frauding by forcing a preemptive
FEDERAL RESERVE
BANK OF ST. LOUIS
their notes, there was no particular reason why
they w ould do so. A fter all, the holders had bought
the notes because they w ere m ore convenient for
exchange purposes than specie.
Notice that th ere is some circularity in the argu
ment just presented. Convertibility, it asserts, was
necessary to prevent noteholders from demanding
com pensation for giving up their specie, w hich
they had been holding fo r use as money. But these
people had exch an g ed th e ir sp ecie fo r n otes
precisely because the notes w ere a m ore conven
ient form of money! This paradox m akes it seem
possible that convertibility is not really necessary;
indeed, th e re are both historical and theoretical
reasons for suspecting that it may not be. In p rac
tice, how ever, the vast m ajority of private banks
of issue have attached convertibility com mitments
to their n otes.19 T he Bank of England, fo r example,
began circulating convertible notes shortly after it
received a royal ch arter in 1695. These notes b e
cam e the principal paper cu rren cy of the relatively
developed reg io n su rro u n d in g L ondon (the
“M etropolis”).” 20
G o v ern m en t P a p e r C u r re n c y
Governments eventually acquired a role in the
paper cu rren cy system by regulating the issuance
of private paper currency, and/or by issuing paper
cu rren cy directly. The motives for this decision
w ere essentially the same as those w hich drove
governm ents to acquire a monopoly over coinage;
some com bination of a desire to improve effi
ciency by facilitating the development of uniform
and reliable paper money, and a desire to earn
revenue by regulating or replacing the private
banking system. This revenue has been earned in
a variety of ways. In some cases, governm ents
have earned substantial sums by granting private
institutions the right to issue paper cu rren cy in
retu rn for some kind of financial consideration.
(See, for example, the discussion of the establish
m en t o f th e Bank o f England w h ich ap p ears
below.) In oth er cases, revenue has been earned
through direct cu rren cy seigniorage, in w hich the
governm ent issues paper cu rren cy to purchase
goods and se rv ice s, o r th ro u g h in d ire c t seig-
liquidation. Incomplete information problems make it impos
sible for these agents to detect fraud without a liquidation.
While this model is not reasonable in every historical context,
it represents a first step toward explaining the prevalence of
convertibility.
20For the early history of the Bank of England, see Clapham
(1944), volume I.
39
niorage, in w hich the governm ent issues paper
cu rren cy to purchase and retire its own bonds.
Government paper curren cy can take a variety
of form s. The earliest form o f governm ent paper
c u r r e n c y —and u ntil quite re c e n tly , th e m ost
common form —was rep resen tativ e currency. A
governm ent curren cy is said to be "rep resen ta
tive” if it is issued under a convertibility com m it
ment; that is, a governm ent prom ise to redeem the
curren cy in specie, at par and on demand. Rep
resentative curren cies are the government-issued
an alog u es to p riv ate, c o n v e rtib le b an k n otes.
W h ile they have u sually b een issued by g o v e rn
m en t-o rg an ized " c e n tr a l b a n k s ,” th e y have
sometimes been issued directly by the govern
ment. The United States, for instance, had directly
issued representative cu rren cy during 1879-1913
(the U.S. n otes, o r " g r e e n b a c k s ,” w h ich w ere
issued by th e T re a su ry ) and re p re se n ta tiv e
currency issued by a central bank during 1914-1933
(the Federal Reserve notes, w hich w ere issued by
the Federal Reserve Banks).21
Governm ents have also issued curren cy that is
not convertible into specie, or anything else. This
type of cu rren cy is often referred to as fia t cu r
rency.22 During the Civil W ar, both the Union and
the Confederacy issued fiat cu rren cy to finance
p a rt o f th e ir m ilitary p u rch a ses. T h e U nion
cu rren cy was the greenback m entioned above.
M odern U.S. c u r re n c y is also fia t in n a tu re.
Federal Reserve notes (our dollar bills) have not
been convertible for dom estic holders since 1933;
since 1971, they have not been convertible fo r any
holders w hatsoever. The Federal Reserve Banks
21The greenbacks were first issued in 1863, but were not
convertible until 1879. Federal Reserve notes were convert
ible for domestic holders from the establishment of the
Federal Reserve System in 1914 until March 1933. They
remained convertible for certain foreign holders until 1971.
22A distinction is sometimes made between inconvertible
government currencies that are issued in purchase of assets
(and so form the liabilities side of a “ balanced” balance
sheet), and currencies which are issued in purchase of goods
and/or services. Currencies of the former type are referred to
as fiduciary. Many economists believe that currencies derive
much of their value from the assets which back them.
Descriptions of this view appear in Smith (1985b) and Russell
(1989a). It suggests that fiduciary currencies may be less
likely to decline in value (that is, to depreciate) than fiat
currencies.
23Strictly speaking, the Federal Reserve System pays for the
Treasury securities it purchases by issuing claims on the
Federal Reserve Banks. These claims can be redeemed in
currency—Federal Reserve notes—which can be held by the
general public as cash balances, or by commercial banks as
reserves. Alternatively, the claims can be converted into
demand deposits at the Federal Reserve Banks, which can
issue most of these notes in purchase of U.S.
Treasury securities.23
WHAT IS A CURRENCY SYSTEM?
A nation’s cu rren cy sy stem can be defined as the
set of laws, conventions and practices that deter
mine the form and role of cu rren cy in the nation’s
economy. A com plete description of a nation’s
cu rren cy system would provide answ ers to ques
tions like: "W hat things does the econom y of this
nation use as cu rren cy?”, "W hat sorts of institu
tions (private and/or government) are perm itted to
issue cu rren cy under the nation’s laws?”, "W hat
role (if any) does the nation’s governm ent play in
defining the econom y’s curren cy unit, or in p re
serving its value?", and ’’W hat is the nature of the
relationship betw een the nation’s cu rren cy system
and its m onetary and financial system s?”
HOW DOES TH E U.S. CURRENCY
SYSTEM W O RK ?
This section will provide a b rief summary of the
history and legal fram ew ork of the U.S. cu rren cy
system. It will focus on a pair of legal restrictions
that play a critical role in shaping the system.
These restrictions would be prim e candidates for
revision or repeal if the system w ere to be refo rm
ed or deregulated.
As previously noted, the U.S. Constitution gave
Congress exclusive pow er to define a national
m onetary unit and produce coined currency. In
addition, the states w ere explicitly prohibited
from issuing paper cu rren cy directly.24 The
also be used by commercial banks as reserves. The decision
concerning how the claims are divided between these
competing uses is made by the private sector.
At present, currency held by the public, or as reserves,
accounts for about 85 percent of total claims on the Federal
Reserve Banks, while U.S. Treasury securities account for
about 75 percent of their total assets. In addition, the
economic implications of the scheme for paying for these
securities just described are identical to those of an alterna
tive scheme under which the System paid for Treasury securi
ties with newly issued Federal Reserve notes, and the private
sector decided how much of this currency to retain and how
much to deposit with the Reserve Banks.
The Federal Reserve Act prohibits the System from
purchasing newly issued Treasury securities—an action that
would amount to issuing currency (and/or Reserve Bank
deposits) to finance government purchases. [See the defini
tion of “ indirect currency seigniorage” presented earlier in
this section.]
24Article I, Section 10 of the U .S. Constitution denies the states
the power to “ emit Bills of Credit;” this was almost universally
understood to prohibit them from issuing their own currency.
SEPTEMBER/OCTOBER 1991
40
Constitution was silent, how ever, on two ques
tions that ultimately becam e controversial: Does
the federal governm ent have the right to issue
paper currency? Do eith er the federal govern
ment or the states have the right to authorize
private institutions to issue paper cu rren cy—do
they have the right, th at is, to grant ch arters to
private banks?
Shortly after the Constitution was ratified, the
states began to ch a rte r private banks o f issue.25 In
1791, and again in 1816, the federal governm ent
chartered a single private bank—the Bank of the
United States. For the next three-quarters of a
century, the bulk of the paper curren cy that circu
lated in the U.S. was issued by state banks; virtu
ally all of the rem ainder was issued by the United
States Bank. The rights of the federal governm ent
and the states to ch a rter private banks w ere even
tually affirm ed (in separate decisions) by the U.S.
Suprem e Court.26
In 1865, Congress imposed a tax on note issue by
state banks that was high enough to make the
activity unprofitable. This action, w hich cam e one
year after Congress had established a system of
fed era lly c h a rte r e d b a n k s o f issu e called the
National Banking System, was evidently intended
to put an end to state banking.27 A nother wartim e
innovation was the issuance, beginning in 1862, of
"green backs.” For the next 50 years, the U.S. stock
of paper curren cy consisted almost entirely of
national bank notes and greenbacks.28
The Civil W ar produced a dram atic expansion of
the federal governm ent’s role in, and pow ers over,
the U.S. m onetary system. In the years immedi
ately following the w ar, the right of the federal
governm ent to play this role, and to exercise these
exp an d ed p o w ers, w as a ffirm ed in a se rie s of
25For an exhaustive list of banks chartered by the states prior to
1837, see Fenstermaker(1965).
26Key decisions upholding the right of the states to charter
banks are Craig v. Missouri (1830) and Briscoe v. the Bank of
the Commonwealth of Kentucky (1837). Key decisions
regarding the right of the federal government to charter banks
are McCulloch v. Maryland (1819) and Farmers and
Mechanics Bank v. Dearing (1875).
27State banking survived because it proved possible for many
state banks to convert to pure deposit banking, which was not
taxed prohibitively.
28They were eventually supplemented by substantial quantities
of silver certificates, and by minor quantities of federal govern
Digitized forFEDERAL
FRASER RESERVE BANK
OF ST. LOUIS
Supreme Court decisions. The w ar converted a
political system in w hich the m onetary pow ers of
federal governm ent w ere sharply circum scribed
into one in w hich they w ere virtually unlimited.
Almost half a century later, the federal govern
m ent’s m onetary pow ers w ere wielded in dram atic
fashion w hen Congress passed the Federal Beserve Act o f 1913. This legislation established 12
"Federal Beserve Banks” that collectively constitu
ted a "cen tral bank” fo r the United States.29 The
Beserve Banks issued a new form of representative
paper cu rren cy called "Federal Beserve notes.”
T h e se n o tes b ecam e th e b asis fo r th e U.S.
c u r re n c y system .
During the first two decades following the
passage of the Federal Beserve Act, the national
banks retained the right to issue limited quantities
of notes. In 1935, however, the national banks’
issue rights expired; Congress declined to renew
them, and made provisions fo r the gradual retire
ment of all national bank notes still outstanding.30
Since 1935, the Federal Beserve Banks have been
the only U.S. organizations authorized to issue
paper cu rren cy on a regular basis.31
T h e re s u lt of th is h isto ric a l p ro ce ss can be
sum marized as the first of two basic legal restric
tions w hich govern the U.S. cu rren cy system: the
federal governm ent has a legal m o n o p o ly over the
issuance of currency, w hether in coin or paper
form.
The second basic legal restriction involves the
relationship betw een currency, which is now
exclusively federally issued, and "m oney” of other
so rts, w h ich co n tin u e s to b e provided by th e
p riv ate se cto r. P riv ately issu ed m on ey is r e
quired to be convertible (redeem able at par and on
ment currency of other sorts. See Friedman and Schwartz
(1963), pp. 124-34, and Timberlake (1978), chapter 10.
29The text of the act appears in the Annual Report of the
Federal Reserve System for 1914.
30See Friedman and Schwartz (1963), p. 442.
31ln principle, Congress retains the right to authorize the
Treasury to issue paper currency directly. It has declined to
do so, however, since the establishment of the Federal
Reserve System.
41
demand) in governm ent cu rren cy .32 In practice,
privately issued money consists of deposits at
com m ercial banks and thrift institutions that are
potentially convenient as media of exchange—i.e.,
that are readily transferable (checkable) and avail
able in small denominations. The governm ent
requires that these deposits be convertible, and
they are referred to as "demand deposits.”
If demand deposits are to be convertible into
governm ent curren cy, they m ust be denominated
in the same units, and have the same m arket value
per unit, as governm ent currency. Consequently,
the convertibility restriction, com bined with the
governm ent’s cu rren cy monopoly, imposes a
com m on denominational and value standard on all
U.S. money. The denom inational standard is of
course the "dollar,” the basic unit of governm ent
curren cy; the value standard is the purchasing
pow er of a dollar (or any fixed num ber of dollars)
of this currency.
It is w orth noting that governm ents usually
attem p t to e n h a n ce th e a cce p ta b ility o f th e ir
cu rren cy by making it legal tender. Legal tender
laws either require or strongly encourage people
to accept governm ent cu rren cy in paym ent of
nom inal debts—debts denom inated in national
cu rren cy units. In the United States, both coins
minted by the Treasury and Federal Reserve notes
are legal tender.
HOW DID THE MODERN U.S.
CURRENCY SYSTEM D EVELO P?
Scien ce h a s b e e n slo w to adm it th e d iffe r e n t explan
atory w orld o f h istory into its d om ain —an d ou r
in te r p r e ta tio n s h a v e b e e n im p o v e r is h e d b y this
o m is s io n . S c ie n c e h a s a ls o t e n d e d to d e n ig r a te
history, w hen f o r c e d to a con fron tation , by reg ard
ing an y invocation o f contingen cy a s less elegant o r
less m ean in gful than explanation s b a s e d directly on
tim eless “law s o f n atu re.’’
—Stephen Jay Gould (1989).
32During the earliest decades of U.S. banking history, the
convertibility requirement was largely implicit. Early in the
nineteenth century, however, states began to pass legislation
which explicitly imposed convertibility on the banks—or,
alternatively, to include convertibility requirements in bank
charters. The federally chartered United States Banks had
convertibility requirements in their charters. Convertibility
requirements were standard features of state charters issued
under the “ Free Banking” laws of 1836-63. The federal
E n glish O rigins
S y n o p s is : The origins of the m odern U.S. cu r
rency system can be traced in large part to Eng
land. Many im portant features of the U.S. cu r
rency system w ere based on English models. The
early history of paper curren cy in England was
dominated by the governm ent’s need for specie
revenues to finance its foreign wars. This need
caused the governm ent to establish two princi
ples—Bank of England monopoly, and strict specie
convertibility—as the basis for England’s system of
paper currency. These principles had a profound
effect on the evolution of paper curren cy and
banking in the United Kingdom, and later in the
United States.
In England, the notion o f organized note issue
seems to have arisen during the latter part of the
seventeenth century. At the time, England had
had a gov ern m en t-m on op oly co in c u r re n c y
system for several centuries, and had begun to
develop a paper cu rren cy system based on bills of
exchange.33 During the last decade of the seven
teenth century several groups of entrepreneurs
recognized an opportunity to profit by providing a
m ore convenient paper currency. Each of these
groups sought royal ch arters for banks of issue.
Horsefield (1960) singles out fou r groups for
special study. One of them, led by William Paterson,
proposed a bank w hich would lend convertible
n otes on co m m ercia l se cu rity . T h e new b an k
w as called th e Bank o f England. It re ce iv e d a
c h a r te r in 1 6 9 5 , and has o p e rate d co n tin u o u sly
sin ce; it is now th e c e n tra l b a n k o f th e U nited
K ingdom .34
T h ree oth er groups, led by Hugh Chamberlen,
Joh n Briscoe, and Joh n Asgill and Nicholas Barbon,
respectively, proposed “land banks” which would
lend inconvertible notes on the security of land
and other real property. The land banks of Briscoe
and Asgill-Barbon actually operated fo r a short
time during 1695-96. In the latter year, they w ere
consolidated pursuant to a schem e to secure a
royal ch arter by raising £2,000,000 in specie to be
len t to th e B ritish g o v ern m en t, w h ich w as des-
charters issued during the late nineteenth century, under the
National Banking Act, also included convertibility
requirements.
33For a description of early English monetary history, see Feavearyear (1963), chapters l-IV.
34The definitive history of the Bank of England has been written
by Sir John Clapham (1944, two volumes).
SEPTEMBER/OCTOBER 1991
42
perate fo r funds. T he ch arter of this “National
Land Bank’’ required that it raise half of the
specie loan prior to beginning operation. W hen
it proved unable to do this, the ch arter lapsed
and the schem e fell apart.35
The fear that a public accustom ed to coin cu r
ren cy would not accept inconvertible notes c e r
tainly played a role in the collapse of the Na
tional Land Bank and other land bank schemes.
Nevertheless, even Horsefield (1960), who is
generally unsym pathetic to the concept of land
banking, points out that “the m ajor cause of
these events was an accident o f tim e.”36 The
Bank of England, w hich had obtained its ch arter
by m eans of a similar com m itm ent to provide
specie, had drained the capital m arket of funds.
This problem was exacerbated by the onset
of a com m ercial crisis, w hich forced the Bank
of England to suspend specie convertibility of
its notes.37
In 1697 the Bank of England obtained, in
retu rn for a fu rth er extension of credit to the
governm ent, a form al com m itm ent that Parlia
m ent would authorize no other banks so long
as the Bank existed. Its urgent desire for this
co m m itm en t su ggests th a t it co n tin u ed to
regard land banking as a viable competitive
threat. Clapham (1944) w rites that “the General
Court [the directors of the Bank of England]
wanted no m ore Land Banks.”38 In 1708, in
retu rn fo r fu rth er loan com m itm ents, the m o
nopoly grant was “reenacted and made more
precise.” Parliam ent explicitly prohibited any
firm consisting of m ore than six partners from
issuing notes in England.39 Thereafter English
note issue was dom inated by the Bank of En
gland. T h e small “cou n try b an k s” operated in
35See Horsefield (1960), chapters 14-16, and Clapham (1944),
volume I, pp. 33-34.
“ Horsefield (1960), p. 246.
37lbid., pp. 246-47.
38Clapham (1944), vol. 1, p. 47.
39See Clapham (1944), p. 65, and Feavearyear (1963), pp. 167-68.
40During 1797-1821 (the era of the Napoleonic Wars), the Bank
of England suspended specie payments. Although its notes
were not officially legal tender, they became so operationally.
Specie virtually disappeared from circulation, most payments
were made in Bank of England notes, and other English
banks redeemed their notes in Bank of England notes. See
Feavearyear (1963), pp. 182-85, and Viner(1937), p. 154.
41Macaulay (1877) provides a colorful and illuminating para
graph describing this relationship (IV, pp. 551-2).
42Clapham (1944), describing the Bank’s first summer,
observes that “ what the government—like the Bank—most
Digitized for FEDERAL
FRASER RESERVE BANK
OF ST. LOUIS
its shadow, and evolved along strictly convertible
lines.40
The development o f the British cu rren cy sys
tem can be properly understood only in the
context of the symbiotic relationship betw een
the British governm ent and the Bank o f En
gland.41 The Bank regarded its paper curren cy
monopoly as critical to its profitability, and was
willing to make large financial concessions to
the British governm ent in order to protect and
extend it. The British governm ent, on the other
hand, was willing to grant the Bank a monopoly
because it needed the Bank’s financial assist
ance—in particular, to help it obtain specie to
finance foreign w ars.42 Under the circu m
stances, it was profoundly in the in terest of
both parties fo r governm ent liabilities to be
identified as closely as possible w ith Bank liabili
ties, and fo r Bank liabilities to be identified as
closely as possible with specie.43 The simplest
and most certain way to achieve this was for
the Bank to lend extensively to the governm ent
and make its notes strictly convertible.44
In Scotland the situation was quite different.
The convertible, com m ercial Bank of Scotland
was chartered by the Scottish Parliam ent in
1695. This bank, unlike the Bank of England,
w as sta tu to rily u ninvolved in g o v ern m en t
finance. It was granted a 21-year note issue mo
nopoly (which was not renewed). In 1705, both
Hugh C h am b erlen , w ho had now m oved to
Scotland , and Jo h n Law , w ho w as la te r to
achieve m onetary infam y in France, proposed
land banks fo r Scotland. Both proposals w ere
rejected because they involved notes that w ere
legal ten d er—a status Parliament was unwilling
wanted in the summer of 1696 was not a circulation of notes
but cash, hard cash for the Army in Flanders” (vol. 1, p. 39).
43Clapham (1944) points out that under the Act of 1697, which
formalized the Bank’s monopoly, “ forgery of the Bank’s notes
was to be punished with death, the penalty for clipping or
coining the King's money. Bank notes were not yet the King’s
money, but they were getting near to it.” (vol. 1, p. 50). In
October of 1698, he writes, the English Treasury agreed to
“ 'receive such bills of the Bank of England commonly called
Bank Bills ... provided the said bills are not at any discount. ’
The time was getting nearer when the Bank would circulate
the Exchequer Bills for the Treasury, cash them on demand,
accept them as deposits, make generous advances on their
security, and even pay a dividend on them.” (vol. 1, p. 56; my
emphasis). And in 1710, just two years after it had acted to
further strengthen the Bank’s monopoly status, the British
Parliament passed an Act “ ‘for engaging and obliging the
Bank of England ... to exchange all Exchequer bills for ready
money on demand.’ ” (vol. 1, p. 67).
44Santoni (1984) asserts that “ the Bank’s contract with its
customers to redeem its notes at a fixed price in terms of gold
43
to grant private liabilities.45 Apparently, the
failure of the English land bank schem es had
created a belief that inconvertible notes would
be accepted only if they w ere legal tender. As
we shall see, this belief was also widespread in
the American colonies—w here experience ulti
mately refuted it.
During the eighteenth and early nineteenth
centuries, Scottish banking was considerably
m ore competitive than English banking. The
m ajor banks fought bitter "note duels,” present
ing their com petitors’ notes for payment in an
effort to drain their specie reserves and force
them to retren ch . One defensive response to
these duels was the issuance b y Scottish banks
of notes w hich contained an “option clause”—a
clause that granted them the right to defer
specie payments for a fixed period in retu rn for
legal in terest.46
Experim ents like the option clause might well
have led to fu rth er departures from convertibili
ty in Scotland, w here both the public and the
governm ent w ere m ore com fortable with banks
and paper curren cy than their counterparts in
England. During the financially troubled years
of the early 1760s, how ever, the option clause
and other "irregu lar” Scottish banking practices
a ttra c te d u n fa v o ra b le a tte n tio n in England
(whose Parliament had absorbed that of Scot-
was a voluntary arrangement” (p. 15). He justifies this asser
tion by noting that the gold standard was not imposed by
British law until 1821, more than a century after the Bank was
founded. While this is certainly true, it is also true that until
the early nineteenth century much of British bank regulation
was implicit rather than explicit. One illustrative example
involves the monopoly status of the Bank of England. Most
historians describe the Bank as having possessed a
monopoly over joint-stock (corporate) banking in England.
However, the Act of 1708 gave it a monopoly only over jointstock note issue. Nevertheless, none of the deposit banks in
England attempted to organize in corporate form. Feavearyear (1963) explains this by noting that, despite the lack of
an explicit legal prohibition against joint-stock deposit
banking, “ there can be no question whatever that the inten
tion was to give the Bank of England a monopoly of joint-stock
banking, and that had any other institution of more than six
partners attempted to carry on banking in England in any
manner whatever at any time during the first half of the
century it would have been suppressed” (pp. 167-68). Simi
larly, despite the lack of an explicit legal requirement that
bank notes be convertible, there can be little question but that
the intention was that they should be convertible, and that
any attempts by the Bank of England, or any of the English
country banks, to issue inconvertible or semiconvertible
notes would have been suppressed—just as attempts of this
sort by Scottish banks were suppressed (see below).
Santoni also asserts that the establishment of an official
gold standard occurred as a result of “ the Bank’s continuous
prods to an unwilling government.” He justifies this assertion
by noting (1) that the deadlines set by Parliament for resump
land in 1707). In 1765 the British Parliament
stepped in with an act prohibiting notes con
taining an option clause, or any other depar
tures from strict convertibility.47
C olonial O rigins
S y n o p s is : The Am erican colonies experim ented
with a variety of cu rren cy systems based on
inconvertible notes issued by colonial govern
ments. During the early eighteenth century the
British governm ent began to regulate these sys
tems. British regulation forced some of the col
onies to back their notes m ore carefully, and
eventually prevented all the colonies from m ak
ing their notes legal tender. By the end of the
colonial period many of the colonies had deve
loped successful and popular cu rren cy systems.
These systems w ere based on inconvertible
notes w hich w ere carefully backed, and w ere
not legal tender.
Conditions in early colonial America dif
fe re d fro m th o se in England ev en m ore
profoundly than did conditions in Scotland. In
the colonies, the most pressing m onetary
problem was a specie shortage: the quantity
o f sp ecie th e co lo n ists w e re able to re ta in
seems to have been insufficient to m eet their
tion were repeatedly postponed, and (2) that the Bank
resumed paying specie for small notes in 1817, and Parlia
ment intervened in 1819 to prohibit it from doing so. However,
Viner (1937), Clapham (1944), and Feavearyear all report that
the government’s primary goal in postponing resumption was to
avoid embarrassing the Bank. [Viner writes that “the govern
ment continued to refuse to obligate the Bank of England to
resume cash payments, and both government and Bank were
obviously waiting for the course of events to disclose the auspi
cious occasion for resumption” (p. 172, my emphasis); Feav
earyear notes that “the Government refused to allow the Bank's
hand to be forced, and repeatedly extended the term of the
Restriction Act” (p. 214).] These authors also describe the
resumption attempt cited by Santoni as a failure, and the
government prohibition as an attempt to protect the Bank.
Finally, both Clapham (1944, vol.2, p. 70) and Feavearyear (pp.
221-22) report that the Bank opposed the legislation which actu
ally compelled it to resume specie payments.
45For a history of the early years of Scottish banking, see
Checkland (1975) or White (1984). For a description of the
Scottish land bank proposals and their fate, see Horsefield
(1960), pp. 175-78, 215-16.
46The option clause has attracted a good deal of academic
attention in recent years. See White (1984), pp. 25-30,
141-42, Rockoff (1986) and Dowd (1988), for example.
47See Checkland (1975), pp. 118-21; White (1984), pp. 29-30.
The legislation originally proposed ruled out the option clause
only; it was later amended to rule out any notes not redee
mable on demand.
SEPTEMBER/OCTOBER 1991
44
needs for a medium of exchange.48 During the
1650s, the M assachusetts Bay Colony attem pted
to allay the shortage by operating its own mint
(which produced the renow ned "Pine T ree Shill
ings”). The British governm ent viewed this ac
tion as usurping a royal prerogative, how ever,
and forced M assachusetts to close the m int.49
The colonies also experim ented with commodity
currencies of d ifferent types; these included
wampum (Indian beads), rice and tobacco.50
Despite the public's need for m ore convenient
means of paym ent, the introduction of a new
form of governm ent paper cu rren cy was m oti
vated in the first instance by the fiscal exigen
cies of a colonial governm ent. In 1696, the
M assachusetts legislature experienced great
difficulty financing an expedition against the
French in Canada. It decided to issue "bills of
credit" in the form of paper curren cy to use to
purchase supplies. These bills w ere neith er con
vertible nor ultimately redeem able in specie;
they could, however, be used to defray future
tax liabilities. This financing expedient proved
quite successful, and the colony used it rep eat
edly during the ensuing 50 years. W ithin a very
few years, other colonies began to adopt the
practice—first in New England, and later else
where. By 1730 or so, bills of credit had become
the principal currency of the American colonies.51
The earliest colonial bills w ere issued, like
these M assachusetts bills, in anticipation of fu
tu re taxes. A fter a few years, how ever, certain
colonies began experim enting with bills that
w ere issued on loan. Typically the issuing colo
ny would pass laws providing that relatively
small sums in new bills could be lent to individ
uals w ho w ere able to provide land or other
sorts of property as collateral. (Often these
loans w ere m ortgage loans and w ere intended
in part to encourage the colonists to settle and
improve land.) These "loan office” or "land bank”
48For discussions of the specie shortage, see Nettels (1934),
pp. 202-207, and Brock (1975), pp. 1-9.
49See Felt (1839), Bullock (1900), Chapter III, Breckinridge
(1903), pp. 55-56 and Nettels (1934), p. 276.
50None of these experiments proved particularly satisfactory.
For a description of colonial experiments with commodity
currency see Nettels (1934), chapter VIII, and Brock (1975),
pp. 9-16.
51For an encyclopedic account of colonial currency history prior
to 1764, see Brock (1975).
52For discussions of colonial land banking, see Davis (1900),
Kemmerer (1939), Thayer (1953), Billias (1959), Brock (1975)
and Smith (1984), among others.
Digitized forFEDERAL
FRASER RESERVE BANK
OF ST. LOUIS
issues becam e increasingly popular during the
first half of the eighteenth cen tu ry.52
As previously indicated, the legislation auth
orizing the emission of tax anticipation or loan
office bills was typically accom panied by legisla
tion providing fo r th eir eventual retirem en t—
either by imposing future taxes which the bills
could be used to pay, or setting out the term s
according to w hich the loans would be secured
and repaid. The legislators clearly believed that
it was these retirem ent com m itm ents that con
veyed value to the bills.53 Unfortunately, th ere
w ere often great political and financial incen
tives for the colonies to violate these com m it
ments by declining to levy or collect the future
taxes, by declining to collect the loan payments,
or by stretching out the period over w hich loans
could be repaid. W hen these things happened,
the bills would often depreciate in value relative
to specie and goods.54 The extent of the depreci
ation was typically m easured by the discount on
paper cu rren cy relative to specie currency; that
is, by the difference, in percent, betw een unity
(one) and a fraction equal to a given quantity of
specie cu rren cy divided by the quantity of
paper cu rren cy it could be sold for in the open
m arket. If it took 50 shillings in paper curren cy
to purchase 40 shillings in specie currency, for
instance, then the discount on paper curren cy
was 20 percent. During the early decades of the
eighteenth century, many colonial curren cies ex
perienced significant depreciation. In some cases,
the depreciation was quite severe.55
Currency depreciation became particularly con
troversial because most colonies gave their bills
of credit the status of legal tender. Legal tender
laws compelled creditors to accept bills at face
value in payment of debts. If, for example, a loan
agreement called for a repayment of 500 shillings
at the end of five years, the lender could be
forced to accept 500 shillings in bills of credit,
53Nettels (1934) writes that “ In the opinion of the colonists, the
principal factor affecting the specie value of their paper was
the provision made for redeeming it from tax revenues”
(pp. 257-58).
54For analyses of the link between backing and depreciation,
see Smith (1984,1985a, 1985b) and Russell (1988).
55Data on the specie prices of the currencies of various
different colonies are presented by Brock (1975) and Smith
(1984,1985a, 1985b), among others.
45
even though he might prefer specie. The pen
alties for refusing to accept bills were relatively
harsh: the creditor might forfeit the entire
amount of the debt or, in some cases, a multiple
thereof.56
The original idea behind legal tend er laws was
to protect borrow ers, and to reduce the fre
quency of lawsuits, by providing a method of
repaym ent w hich was beyond legal challenge.57
W hen unexpected depreciation occured, how
ever, legal tend er laws tended to benefit debtors
(by reducing the real value of their obligations)
at cred itors’ expense. This made them popular
with farm ers and oth er debtors, and unpopular
with creditor interests.
The creditor interests included a number of
British merchants who did business with the
colonies. Many of these m erchants were wellconn ected in Great Britain; th e ir com plaints,
which were seconded by those of indigenous
merchants and creditors, received sympathetic
attention from the British colonial administration,
and eventually from the British Parliament. After
1730, the colonial administration began to issue
regulations eliminating or restricting the right of
particular colonies to issue new bills or (more fre
quently) to make them legal tender. As the prob
lem of depreciation worsened, however, Parlia
ment considered comprehensive legislation. The
Currency Act of 1751 deprived the New England
colonies of the right to issue legal tender bills and
greatly restricted their powers to issue paper cur
rency of any description. In 1764, a second Cur
rency Act extended the legal tender prohibition
to all the colonies.58
56See Bullock (1900), p. 131, Nettels, (1934), p.265, and
Russell (1988), pp. 47-48.
57See Breckinridge (1903), p. 52, and Hurst (1973), p. 40. West
(1 978) stresses the role of colonial paper currency in
providing a “ means of settlem ent,” but does not mention
legal tender laws explicitly.
58Even before the blanket legal tender prohibition, the British
government had intervened to prevent particular colonies
from making their currencies legal tender. It also intervened
to force some of the colonies to back their legal tender curren
cies more carefully with future tax receipts, and to prevent
others from issuing currency on loan. See Davis (1900, vol. I
and II), Ferguson (1953), Ernst (1973), Brock (1975), Smith
(1984,1985b) and Russell (1988).
For the history of the Currency Acts, see Davis (1900),
Greene and Jellison (1961), Ernst (1973), Brock (1975), Smith
(1985) and Russell (1988).
59ln 1767, Ben Franklin wrote that “ On the whole, no method
has hitherto been formed to establish a medium of trade, in
lieu of money, equal, in all its advantages, to bills of credit,
funded on sufficient taxes for discharging it, or on land secu
rity ... and in the mean time made a GENERAL LEGAL
TENDER.” [Franklin (1971), p. 354; his emphasis.] Ferguson
Many colonies responded to the legal tender
prohibitions by issuing non-legal tender bills of
credit. Although many contem porary analysts be
lieved that giving the bills legal tender status was
essential to preserve their value, this does not
seem to have been the case in practice.59 The
non-tender bills remained quite stable in v a lu e far more stable, in many cases, than their legal
tender predecessors. This was particularly strik
ing because many non-tender issues took the
form of land banks—a mode of issue the British
regarded as particularly prone to depreciation.60
During the decade prior to the Bevolution, the
colonies appeared to be moving toward a system
of non-legal tender land bank currency.61
R ev o lu tio n a n d R eo rga n iza tio n
S y n o p sis : The Revolution completely disrupted
the evolution of the American currency system.
The Continental Congress was forced to finance
wartime expenditures by money creation—a policy
which led to a virtual hyperinflation. The war,
and the depression that followed it, produced
financial problems for both state governments
and the general public. One symptom of these
problems was large public and private debts—
many of which were held by the domestic prop
ertied classes. This situation, combined with mem
ories of the recent inflation, created fears among
the m em bers of th ese classes th at popularly
elected state governments would adopt monetary
policies designed to partially repudiate these debts.
Representatives of the propertied classes domi
nated the Constitutional convention. They moved
(1953) writes that “ The restraining act of 1764 ... prohibited
legal tender laws and required that existing legal tender
currencies be sunk at their expiration dates. Many colonies
protested, in the belief that the legal tender feature was an
essential prop to their currency. Experience was to show,
however, that the restriction did not materially impair the
workings of the currency system” (p. 177).
60For discussions of the strong performance of nontender
paper currency, see Ernst (1973), Smith (1984b) and Russell
(1988).
61See Ferguson (1953), pp. 177-180. Inconvertible government
currency which was not legal tender is of special historical
interest because it was issued under circumstances that
approximated relatively closely the circumstances under
which inconvertible private currency might have been issued.
This was especially true when, as was frequently the case,
the currency was issued on loan rather than in anticipation of
taxes. Russell (1988) argues that the success of government,
non-legal-tender, inconvertible land banking before the Revo
lution provides indirect evidence that private inconvertible
banking might have been feasible after the Revolution, had it
been legally permitted.
SEPTEMBER/OCTOBER 1991
46
to prevent repudiation by prohibiting the states
from issuing their own currency, or from making
privately issued cu rren cy legal tender.
During the Revolutionary W ar, the Continental
Congress was the Am erican central governm ent,
and bore prim ary responsibility for conducting
and coordinating the w ar effort. It also faced a
critical financing problem : under the Articles of
Confederation, it lacked the pow er to levy taxes.
(Colonial opposition to British taxation had been
one of the most im portant causes of the rebellion.)
During the early stages of the Revolutionary W ar,
the Congress attem pted to subsist on voluntary
contributions from the colonies. W hen this source
of revenue proved insufficient, it began to issue
bills of cred it—the renow ned "continentals”—
which w ere backed by little m ore than the pious
hope that the states would eventually provide
funds, or authorize tax levies, to retire them. The
likelihood of such retirem ents becam e ever m ore
distant as the quantity of continentals increased
and the states supplemented them w ith their own
cu rren cy issues. Both form s of paper curren cy
began to depreciate—at first gradually, and later
very rapidly. By the end of the w ar, they w ere
virtually w orthless.62
During the b rief “critical period” betw een the
end of the w ar (in 1783) and the ratification of the
U.S. Constitution (in 1789), the newly independent
states began to reorganize th eir finances and con
sider the problem of providing a paper currency.
A num ber of states issued or seriously considered
issuing bills of credit in anticipation of taxes or on
loan—m uch in the m anner of the prerevolutionary
colonies.63
Unfortunately, the continental hyperinflation
had fractu red the prerevolutionary consensus
reg ard in g th e u sefu ln ess o f p ap er c u rre n c y .
People who had accepted continentals or contin
ental-denom inated securities from governm ent or
private parties w ere outraged that the states ap
62For a good account of the history of the Continentals, and
indeed of Revolutionary War finance, see Ferguson (1961).
See also Calomiris (1988).
63For general discussions of currency issues by (incipient)
states during the critical period, see Nevins(1924), Ferguson
(1961), Nettels (1962), Russell (1988) and Schweitzer (1989).
64For descriptions of public attitudes toward paper currency
during the critical period, see Libby (1894), Hammond (1957)
and Ferguson (1961).
65Most of the other colonies that issued legal tender bills also
experienced serious depreciation—though not on the scale of
Digitized forFEDERAL
FRASER RESERVE BANK
OF ST. LOUIS
peared to have no intention of redeem ing them at
anything close to the values at w hich they had
traded during the early years of the w ar. Indeed,
the propertied classes cam e to view paper cu rren cy
as a device by which popularly elected govern
ments sought to perm it the common people to
escape the burden of their public and private
debts. (The lengthy trade depression that followed
the end of the w ar had increased private debt
burdens.) W ealthy Am ericans becam e terrified
that the state legislatures, which w ere now free
from British restraint, would rep rise the Revolu
tionary experience by issuing large volumes of
inadequately backed legal tender bills—bills which
would rapidly depreciate, and w hich could be
used to retire debts at a fraction of their real value.
As a result, proposals to issue paper cu rren cy that
would have received consensus support before
the Revolution now becam e the subjects of intense
political controversy.64
In Rhode Island, radical populists gained control
of the legislature. They confirm ed the w orst fears
of the anti-paper m oney conservatives by issuing a
legal tender currency, and then engineering a
rapid inflation that seemed clearly designed to
en ab le b o rro w e rs to escap e th e ir d e b ts .65 In
M assachusetts, an agrarian insu rrection (Shay's
Rebellion) erupted as a result of the refusal of the
legislature to issue legal tender paper cu rren cy .66
As it happened, the Constitution was w ritten
and ra tifie d d urin g a p eriod o f co n serv a tiv e
ascendancy— a reaction against excesses of the
sort epitomized by events in M assachusetts and
Rhode Island.67 The conservatives desired a “h ard ”
c u rre n c y im m une from depreciation. As a result,
the fram ers of the Constitution w ere not content
merely to deprive the states of the right to issue
legal tender bills; instead, they w ere prohibited
from issuing cu rren cy of any kind. Specie was
established as the new nation’s sole legal tender
cu rren cy—and, in the minds of many, as the
nation's only legitimate currency.
Rhode Island. On the other hand, colonies that issued
nontender bills experienced little or no depreciation. For a
discussion of this question, see Russell (1988). For descrip
tions of Rhode Island’s post-revolutionary currency policy, see
Phillips (1865), Bates (1898), Nevins (1924) and Ferguson
(1961).
66The classic study of Shay’s Rebellion is Taylor (1954).
67This point is made by Ferguson (1961), pp. 249-250, Nevins
(1924), p. 537, and Schweitzer (1989), pp. 319-320.
47
Thus the peculiar historical circum stances of
the post-revolutionary critical period had a pro
found and lasting im pact on the nature of the U.S.
m onetary system. The traum a of the Revolution
made the curren cy system controversial, and ulti
mately produced a system very d ifferent from the
relatively uncontroversial system of late colonial
times. Indeed, it seems likely that had the colonies
been able to escape British domination without
fighting an expensive w ar, or had the principal
casu s bellu m not been one w hich required that the
w ar be financed by m eans w hich sowed the seeds
of a divisive struggle betw een classes, the United
States might have begun its existence with a decen
tralized cu rren cy system based on (non-legal tend
er?) bills of credit issued by state governm ents.68
The Constitution was silent on the question of
privately issued currency. Indeed, during the
years immediately following its ratification, issu
ance of small-denomination liabilities w hich might
circulate as cu rren cy (which might "pass cu rren t,”
to use the contem porary phrase) was regarded as
a right of all free persons. By the second decade of
the nineteenth century, how ever, the legislatures
of m ost of the states had acted to eliminate or
greatly restrict th atrig h t.69 T h ere w ere at least two
reasons for this. One was the problem s caused by
irresponsible, or downright fraudulent, private
issues. Another, w hich was perhaps m ore com pel
ling, was the desire of the state legislatures to re
serve the right of note issue to state-chartered
banks.70
The Constitution had also been silent regarding
the right of the states (or the federal government)
to ch arter banks of issue—perhaps because private
b an k in g had little h isto ry in th e co lo n ies. In
M assachusetts, the most econom ically sophisti
cated colony, efforts to organize a private land
68Many historians believe that resentment over British efforts to
regulate colonial currency practices played a major role in
stimulating the Revolution. [See Bullock (1900), pp. 56-59,
Davis (1900), vol. 1, chapter XXI, vol. 2, pp. 256-61, Brock
(1975), pp. 561-63, Billias (1959), p.42, and Ernst (1973), pp.
359-60, for example.] This makes it seem very ironic that the
currency restrictions the U.S. Constitution imposed on the
states were generally more restrictive than any the British had
ever imposed. [One exception is that the states could charter
private banks of issue, something the British had prohibited
the colonies from doing. It is not completely clear that the
framers actually intended to authorize state-chartered
banking, however (see below).]
69See Hammond (1957), pp. 27-29,184-85 and Fenstermaker
(1965a), pp. 21-22.
70Both these motives are mentioned by Hammond (1957), pp.
27-29,159-60,184-85. See also Fenstermaker (1965a), pp.
bank along the lines of public land banks began
late in the seventeenth century and persisted
episodically for the next five decades. The m ercan
tile community was somewhat skeptical of land
banking, how ever, since it threatened to compete
with their own lending activities. [M erchants in
the coastal cities provided a good deal of trade
credit to the m erchants and farm ers of the interior.]
In addition, the colonial authorities (both adminis
trative and legislative) w ere reluctant to give up
their monopoly over paper cu rren cy—partly out
o f fear that a form of m oney issued outside of offi
cial control might be subject to manipulation, and
partly out of con cern that it might reduce poten
tial revenues from seigniorage. Although various
private land bank projects received considerable
popular support, they w ere unable to surm ount
this political opposition.71
By 1740, how ever, the restrictions on colonial
issues w hich had been imposed by the British
governm ent had becom e so onerous (land bank
issues, in particular, had been entirely prohibited)
that the M assachusetts legislature was willing to
ch arter a private land bank. W hile the land bank
project received broad support from the public, it
was vehem ently opposed by the colony's governor
(a creatu re of the British), who viewed it as w eak
ening the mother country’s control over the colony’s
economy. The land bank was also opposed by
British m erchants, who saw it as a th reat to their
n ear monopoly over trade credit. Both groups ap
pealed for relief to Parliament, which responded by
enacting legislation prohibiting the establishment
of banking corporations anywhere in the colonies.72
The legislatures of the newly independent states
saw chartered banking as a means by which they
could provide their citizens with paper currency
while at the same time (in many cases) providing
15-16. The importance of the latter motive is indicated by
Hammond’s comment that restrictions on unincorporated
note issue were enacted “ on the complaint of chartered
banks” (p. 184).
71In 1714 there was a well-organized and determined attempt to
organize a private land bank—an attempt which was
supported by some influential British merchants, and
received the endorsement of the Board of Trade. The govern
ment of Massachusetts responded by establishing a public
land bank. See Billias (1959), pp. 3-5, Nettels (1934), pp.
271-275, Davis (1900), volume I, pp. 56-61, volume II, pp.
82-91, Ernst (1973), pp. 27-28, and Metz (1945), chapters 3,4.
72See Billias(1959), Davis(1900), volume II, pp. 130-261,
Hammond (1957), pp. 24-25, Brock (1975), pp. 123-27, Ernst
(1973), pp. 34-35, and Metz (1945), chapter 10.
SEPTEMBER/OCTOBER 1991
48
themselves with revenue. Of course bank charters
would be m ore valuable to their holders, and thus
issuing charters would be more lucrative for the
states, if the charters conveyed an exclusive right to
issue paper currency. This accounts for the prohi
bitions against private issues from other sources.73
But what sort of banks should the states charter?
While it might seem that the colonies' extensive
experience with public land banks should have led
them, as states, to charter private land banks, they
did not in fact do so. Two factors may help explain
this situation. The first is that the right of the states
to charter banks of any sort was not altogether
clear; indeed, the view that they did not have this
right was widely held.74 Doubts about state charter
rights seem to have existed on two levels. Many,
and perhaps most, informed Americans believed
that the Constitution established specie as the only
legitimate form of "money." The question then
becam e "W hat is ‘m oney’? ” and, in particular, "Are
bank notes ‘m on ey?” A conservative view was that
the category "m oney” did include bank notes, and
indeed paper bills of all descriptions, so that paper
curren cy in any form was proscribed. A m ore
m oderate view was that bank notes con v ertible in
sp ecie w ere not money, but m erely its “rep resen ta
tive,” and thus w ere not constitutionally prohibi
ted.75 This view implied that private land banking
conducted along colonial lines, w hich is to say
through the issuance of inconvertible notes, in
volved direct creation of money, and was th e re
fore unconstitutional—even if the bills w ere not
legal ten d er.76
A second possible reason why the states did not
opt for some form of land banking was that, given
73The motives (and actions) of the states in this regard were
similar to those of any license-granting monopolist. The
strategy was evidently based on the British example. During
the eighteenth century, the British government had repeat
edly extracted large payments, or loans on favorable terms,
from the Bank of England in return for extending or strength
ening its monopoly on note issue. A distinctively American
variant of this strategy was for a state government to require a
bank’s organizers to cede the state an equity interest in
return for granting the bank a charter. Sometimes the interest
was ceded gratis, and sometimes merely on favorable terms.
This practice was particularly common in the southeastern
states. In several of these states bank dividends accounted
for a substantial portion of state government revenues. [See
Fenstermaker (1965a), pp. 17-20. For a thorough analysis of
the importance of dividends from bank stock in state govern
ment finance during this period, see Sylla, Legler and Wallis
(1987).]
74For a general description of the nature and source of doubts
about the constitutionality of state banks see Hammond
FEDERAL RESERVE
BANK OF ST. LOUIS
their British-imposed lack of experience with
p rivate b an k in g , th e ir only m odels fo r b an k
c h a rte rs w e re th o se o f B ritish b a n k s—w h ich
meant, for all intents and purposes, the ch arter of
the Bank of England. Indeed, the ch arters of many
of the earliest state banks w ere virtual carbon
copies o f th e Bank of England 's c h a r te r . T h e
attractiveness of the British model may have been
enhanced when, shortly after the Constitution
was ratified, the Federal governm ent decided to
seek a ch arter for a single “National Bank” along
the lines of the Bank o f England. This institution
was called the "Bank of the United States,” and
was established in 1791 with a 20-year charter.
Though the U.S. Bank was basically a private
organization, the Federal governm ent held a
minority interest, and the Bank was expected to
provid e a v a riety o f fin a n cia l se rv ices fo r th e
g o v ern m en t in addition to its p riv ate lending
activities.77 The early state banks seem to have
been intended as state versions of the U.S. Bank.
This is reflected in the fact that until the ch arter of
the U.S. Bank expired in 1811, most states ch ar
tered just one or, at most, a handful of banks.78
The failure of the first U.S. Bank to secure a new
ch arter was due to a com bination of doubts about
its co n stitu tio n a lity , su sp icion of its p ow er, and
d isco m fo rt w ith th e fa c t th a t m uch o f its stock
w as fo re ig n -o w n e d .79 T h e dem ise o f th e b an k
coin cid ed w ith a p eriod o f n atio n al eco n o m ic
exp an sion asso ciated p artly w ith th e im p act of
the N apoleonic W ars on com m od ity p rices, and
p artly w ith th e se ttle m e n t of th e w e ste rn
(trans-A ppalachian) reg ion . A cro ss th e U nited
S tates, and p a rticu la rly in th e new states and
(1957), pp. 103-13, 564-71, and Hurst (1973), pp. 11-12,
141-45.
75See Hammond (1957), pp. 61,105-06, Gallatin (1879), pp.
254-55, (1879), p. 379, Bancroft (1831), p. 40.
76The relatively radical view that private or even public land
banking was consistent with the Constitution, so long as the
states did not try to make the notes of such institutions legal
tender (or even, perhaps, ifthey did), did not become popular
until the economically troubled period following the Panic of
1819.
77For the history of the first U.S. Bank, see Holdsworth (1910).
78For information concerning state banks chartered before
1819, see Fenstermaker (1965b).
79See Hammond (1957), pp. 209-26, and Holdsworth (1910).
49
te r r ito r ie s of the W est and South , b an k s w ere
c h a rte r e d in larg e n u m b e rs.80 T h ese n ew b a n k s
d iffered fro m the esta b lish ed b a n k s in one c r it
ical re sp e c t: th e y w e re o rg an ized p rim arily to
provid e c re d it to fa r m e r s .81
E c o n o m i c D e p r e s s i o n a n d Its
C onsequences
S y n o p s is : The lengthy and severe depression
that followed the Panic of 1819 placed great strains
on the U.S. banking and cu rren cy system. The
crisis exp o sed ab asicin con sisten cv betw een two
goals of the developing banking system: specie
convertibility of bank cu rren cy , on one hand, and
liberal extension of farm cred it, on the other.
Many of the southern and w estern states, in w hich
farm erspred om inated , responded by exp eri
m enting w ith system s in w hich banks issued in
convertible notes. W hile some of these systems
w ere clearly not viable, others appear to have
had prom ise. The fed erally ch a rtered Bank of the
United States intervened to put an end to all of
them , how ever.
At this point in U.S. history, a conflict arose be
tween the needs of economic development and the
devotion to “hard money” which grew out of the
Revolution. Economists since the time of Adam Smith
had understood that banks which issued convertible
notes, and thus were vulnerable to runs, could not
safely lend to farmers: farm loans were typically long
term, illiquid and relatively risky.82 The need for
farm credit was sufficiently great, and public under
standing of banking sufficiently slight, however, that
the legislatures of the southern and western states
80At the time, the western region included western Pennsyl
vania, Ohio, Kentucky, Tennessee, Indiana, Illinois and
Missouri. (The following account applies most closely to
the experience of the last five states listed.) The southern
region included Virginia, North Carolina, South Carolina,
Georgia and Alabama. (The following account applies most
closely to the experience of North Carolina, South Carolina
and Georgia.)
For information concerning state banks chartered before
1837 see Fenstermaker (1965a).
81For a discussion of the development of agricultural (and
other types of long-term) banking see Hammond (1934),
Hammond (1957), pp. 676-80, and Redlich (1951), vol. 1,
pp. 11-13, 44-45.
82Many farm loans were seasonal loans to finance planting
or harvesting the crop. Variation in weather conditions and
crop prices could make such loans quite risky, and by
contemporary commercial standards they were relatively
long term. However, a good deal of farm credit involved
much longer-term loans to finance the purchase and/or
clearing of land, the purchase of equipment (and in the
South, slaves), etc. Most loans for these purposes were
“ accommodation loans.” An accommodation loan did not
permitted (and indeed encouraged) their banks to
finance large quantities of farm credit by issuing
convertible notes.83
The collapse in agricultural prices which occurred
during 1818-19 (and led to the Panic of 1819) made
the two mandates of the southern and western
banks—prompt specie redemption of notes, and
liberal extension of farm credit—impossible to recon
cile.84 The price collapse produced widespread loan
defaults and runs on banks. Since agricultural loans
were impossible either to collect or to sell in a short
time, the runs could be stopped only by suspending
specie paym ents—by refusing to redeem bank notes
in specie on demand. Even after the runs subsided,
however, the defaults represented tremendous losses
for the stockholders of banks that were willing, or
could be compelled, to honor their convertibility
commitments. Indeed, a large number of western
banks had become insolvent.85 The managers of the
southern and western banks responded by declining
to resume payments, and their notes continued to
trade at substantial discounts in the open market.
The holders of these notes were forced, in effect, to
bear some of the financial losses associated with the
Panic.
The governments of the western and southern
states responded to this situation in very different
ways. Most western states had banking systems
composed of large numbers of relatively small banks.
Since the losses associated with the price collapse
and Panic were particularly heavy in the West, most
western banks were insolvent, or nearly so; their
notes were trading far below par. Many westerners
viewed the events of 1818-19 as a conspiracy on the
part of the "monied interests” to ruin them and seize
have a fixed term; the borrower was expected to pay an
“ installment” equal to a fixed fraction of the principal (typi
cally, 10 percent) every 90 days. In practice, installments
could be deferred and/or reduced, so that the lifespan of
an accommodation loan could greatly exceed the term im
plied by these conditions. Crop loans might also be ex
tended as accommodation loans, with the presumption
that they would be fully repaid at the end of the season.
Very often adverse circumstances made this impossible,
however. See Fenstermaker (1965a), pp. 47-49, Redlich
(1951), volume I, p. 11, Holder (1937), pp. 119-22, and
Russell (1989a), pp. 69-73.
83See Hammond (1957), pp. 178-83, Redlich (1951), pp.
9-12.
84For descriptions of and data on the price collapse, see
Cole (1938), Berry (1943), Smith (1953) and Russell
(1989c).
85A list of the banks which failed during or shortly after the
Panic can be compiled from information presented by Fen
stermaker (1965a). Berry (1943) provides information con
cerning bank failures in Ohio during this period.
SEPTEMBER/OCTOBER 1991
50
their property. The banks, they believed, had
been agents of this conspiracy. Consequently,
the w estern legislatures moved to revoke their
ch arters and/or to fo rce them to liquidate. Pri
vate banks w ere replaced by monolithic, statemanaged organizations called "Banks of the State”
or “relief banks.” They w ere supposed to lend
inconvertible notes w hich would be given quasilegal tend er status by their states.86
The history of the relief banks was brief, con
troversial, and generally undistinguished. Since
their m andate was to lend liberally to financially
distressed farm ers, the m arket value of their
loan portfolios was low relative to the nominal
value of th eir outstanding notes. This circu m
stance was reflected in the deep discounts on
the notes.87 The lending standards of the Illinois
relief bank w ere so lax, and its efforts at collec
tion so ineffectual, that its notes soon becam e
virtually w orthless.88
Both the legislation that created the relief
banks and the "stay law s” w hich made their
notes quasi-legal ten d er w ere of doubtful con
stitutionality. These laws w ere challenged vig
orously in state and federal courts. In Kentucky,
Tennessee and Missouri, these challenges ulti
mately led to the demise of the relief banks.
Some of the federal cou rt challenges w ere o r
chestrated by the second Bank of the United
States. In 1816, Congress had responded to fi
nancial problem s created by the W ar of 1812
by chartering a second U.S. Bank. The new
bank was (again) a private institution, though
the federal governm ent held a sizable m inority
interest. The term of its ch a rter was 20 years.89
The U.S. Bank's ch a rter required its notes to
be strictly convertible. The Bank’s m anagem ent
believed that it could circulate convertible notes
in the w est and south only if the notes of its lo
cal com petitors w ere also convertible. The m an
agem ent also believed that acquiring a large
local circulation was essential if the Bank w ere
86The best available account of the Panic and its aftermath
is Rothbard (1962).
The legislatures of these states enacted “ stay laws”
which provided that foreclosures and other legal actions
for the collection of delinquent debts (“ executions,” to use
the contemporary description) could be delayed for long
periods—typically a year or more—unless the creditors in
question were willing to accept Bank of the State notes at
par. See Rothbard (1962), chapters II, III.
87Data on the discounts on the notes of the Kentucky relief
bank are available from Berry (1943) and Sumner (1896).
Fenstermaker (1965a) presents somewhat less complete in
formation on the discounts on other relief bank currencies.
Digitized for FEDERAL
FRASER RESERVE BANK
OF ST. LOUIS
to earn a profit on its southern and w estern
operations, or effectively perform its duties as
the payments agent of the federal governm ent.90
The Bank consequently made strenuous efforts
to force the w estern banks to retu rn to the
specie standard. This policy allowed it to extend
the scope of its activities in the W est, while
striking a pose as the defender of sound cu rren
cy. In the meantim e, bitter political controversy
over the redistributive implications of the activi
ties of the Banks of the State greatly reduced
their effectiveness.91
In many ways, the relief banks w ere direct
descendants of the public land banks that had
been quite successful, and had enjoyed consen
sus public support, during the years preceding
the Revolution. Unfortunately, the extraordinari
ly adverse econom ic circum stances w hich led
the w estern states to circum vent the Constitu
tion by creating these institutions also served to
ensure th eir ultimate demise. The relief banks
w ere created by the state legislatures for the
purpose of relieving the financial plight of their
constituents, rath er than to provide a sound
currency, or to earn revenue (which is to say
profits) fo r the states.92 This m andate, com bined
with the depth of the agrarian distress, made it
impossible for the banks’ m anagers to resist
making too many loans to troubled farm ers—
who w ere already burdened with debt and
whose ability to repay was very doubtful. The
same circum stances led the states to supplement
the relief bank legislation w ith stay laws and
related provisions. These provisions made the
banks extrem ely controversial, earning them the
enmity of creditor interests in general and the
pow erful United States Bank in particular.
The failure of the relief bank experim ent had
an effect on public attitudes tow ard m onetary
issues that was rem iniscent of public reaction to
the Revolutionary hyperinflation. People becam e
increasingly suspicious of banks and paper cur-
88See Rothbard (1962), pp. 41-42, 80-83; see also Dowrie
(1913).
89See Catterall (1902), chapter I.
90See Catterall (1902), pp. 96-99, Redlich (1951), pp. 109-10,
124, 128-29, 440-44, Temin (1969), p. 49, and Fraas
(1974).
91For accounts of the demise of the relief systems, see
Rothbard (1962), Chapter III, Hammond (1957), pp.
282-285, and Sumner (1896).
92See Fenstermaker (1965a), pp. 26-27, Rothbard (1962),
pp. 81-83, 85-86, 102-03, and Sumner (1896).
51
rency, and increasingly enam ored with "hard
m oney”—specie, or bank notes rigidly convert
ible into specie. The demise of the relief banks
also gave the Bank of the United States a virtual
monopoly over banking in m uch of the W est.
notes.94 These notes dropped to variable discounts
(against specie) in the open market. Variation in
these discounts seem to have reflected changing
market conditions—much like the variation in
modern national currency exchange rates.95
The southern states tried a d ifferen t—and ini
tially, at least, m ore promising—experim ent
with inconvertible banking. The southern banks,
unlike their w estern counterparts, w ere large
but few in num ber; frequently, they operated
b ranch es across their respective states. The
southern state governm ents held large minority
interests in these banks. Since the dividends on
the shares provided an im portant source of
state revenue, the state legislatures had no
desire to see the banks close down.93 In addi
tion, since the financial distress that accom pa
nied the Panic was less severe in the South than
in the W est, the popular outcry against the
banks was somewhat less strident there. Finally,
because of their large size and branch systems,
the banks of the region w ere m ore effectively
diversified than their w estern counterparts, and
requirem ents imposed by their ch arters had
kept them relatively highly capitalized.
Economic historians have usually viewed the
suspensions as irregular events completely in
consistent with the m aintenance of m onetary
and financial stability.96 T h ere are good practi
cal and theoretical argum ents against this view,
however. On a practical level, the banks stayed
in business, and continued to supply badly needed
currency and credit, despite the depressed condi
tions created by the collapse in prices and the
financial panic. This situation stands in marked
contrast to that of the W est. There, the banking
system collapsed, and a scarcity of cu rren cy
and credit threatened to bring econom ic activity
to a standstill.
W hen the southern banks suspended the con
vertibility of their notes, the governm ents of the
southern states did not force them to liquidate,
or even to close down. Instead, the banks w ere
permitted, and often encouraged, to continue to
do business—lending, collecting on loans, and
conducting other financial transactions, all
through the medium of their now-inconvertible
93See note 73 above.
94For a description of banking in North Carolina after the
suspension, see Russell (1989a).
95Note discount data for North and South Carolina state banks
during 1817-1829 are provided by Russell (1989a). Fenster-
maker (1965a) provides less complete data for all the
southern states.
96For expressions of this view see Gouge (1833), Sumner
(1896) and Klein (1974).
97Before the suspensions, the notes of southern state banks
had traded at or near par with specie; afterwards, they traded
at variable discounts. Thus it seems clear that the suspen
sions exposed noteholders to risks they had not previously
borne. The author believes that after the suspensions, the
market priced bank notes in much the same manner as
modern mutual fund shares. This sort of pricing scheme
would have linked the value of a bank’s notes to the value of
its assets—a link which was absent under convertibility. [See
Russell (1989b,c).] Other theories of inconvertible note
pricing have broadly similar implications, however. The most
popular alternative theory is that the notes of a suspended
bank were priced as risky titles to future specie, payable if
and when the issuing bank resumed payments. Since the
state of a bank’s portfolio was probably the biggest single
factor influencing the prospect that it would be able to re
On a theoretical level, the suspensions shifted
some of the burden of the banks' portfolio risk
from shareholders to noteholders.97 The price
collapse and panic had revealed the true extent
of the risks the banks faced, and had ex acer
bated the already acute scarcity of financial
capital—particularly concentrated financial capi
tal—in the relatively undeveloped southern
states.98 Under these circum stances, it seems
doubtful that current or future bank shareholders
would have been willing to continue to b ear all
of the banks’ portfolio risk—particularly in light
of the heavy losses that a prompt retu rn to
specie payments would have im posed.99
sume, this theory also implies that the holders of inconvertible
bank notes were bearing portfolio risk. And because all of the
portfolio risk had to be borne by the holders of the banks’
liabilities, any risk borne by noteholders reduced the risk born
by shareholders (and vice-versa).
Another alternative theory is that bank notes traded at a
discount during suspension because holders’ inability to
convert them into specie temporarily reduced their useful
ness, relative to specie, as media of exchange. Advocates of
this theory typically view suspensions as a sort of “ safety
valve” protecting fractional reserve banking systems oper
ating under laissez-faire—operating without, that is, govern
ment deposit insurance and/or a government lender of last
resort. For expressions of this view see Friedman and
Schwartz (1963), pp. 163-68, 324-32, and Dwyer and Gilbert
(1989).
98The charters of contemporary banks specified minimum
denominations for shares which were usually well out of
reach of the common people. The rationale behind these
minima is not entirely clear. See Russell (1989a), pp. 35-36.
"T he North Carolina banks, in particular, decided to close
down when specie payments were finally imposed on them in
1828-1829. See Russell (1989a), pp. 25-32, 78-80, Holder
(1937), pp. 250-51, and Flanagan (1934).
SEPTEMBER/OCTOBER 1991
52
A case can be made that the post-Panic suspen
sions began a process w hich, had it been allowed
to proceed unhindered, might have enabled the
South to develop an alternative banking system
which was peculiarly suited to its distinctive needs.
Because the southern econom y was dominated by
agriculture, banks could be useful to southern
econom ic development only if they w ere able to
make farm loans in large volumes. Farm loans,
however, w ere relatively risky—and under con
vertibility, these risks w ere born e almost exclu
sively by bank stockholders. These stockholders
had both the opportunity and the means to lend
outside the South, and could be induced to take
these large risks only in retu rn fo r high average
rates of retu rn. Such rates would have made bank
credit too expensive fo r many farm ers—and usury
laws might have prevented the banks from charging
them in any case. One solution to this problem was
the development of an alternative banking system
w hich could bring the diffuse financial capital of
the com m on people into the risk-bearing process.
Inconvertible priv ate banking seem s to have had
the potential to provide such a system .100
Unfortunately for the southern banks, both the
federal governm ent and its financial agent, the
Bank of the United States, regarded the suspen
sions with almost unalloyed hostility. The U.S.
Treasury D epartm ent was deeply (and somewhat
irrationally, under the circum stances) committed
to fiscal arrangem ents under w hich payments to
the federal governm ent (for taxes, land purchases,
etc.) w ere made exclusively in par cu rren cy .101
Southerners, how ever, w ere used to making
governm ent paym ents in local (bank) currency,
w hich was no longer trading at par. The Bank of
the United States, w hich was charged with the
100This argument is presented in detail in Russell (1989a,b).
101It seems likely that the exchange problems described here
could have been avoided if the Treasury had been willing
to accept and disburse state bank notes at their market
rates of discount. A similar situation had arisen after the
general suspension, of specie payments which occurred in
August of 1814, near the end of the War of 1812. During
the months after the suspension, the Treasury needed
funds to service debt held by residents of New England.
New England was the only region of the United States in
which the banks had not suspended specie payments, so
New England bank notes were trading at par with specie.
Unfortunately, most of the federal government’s revenue
was received in the mid-Atlantic region. The mid-Atlantic
banks had suspended, and their notes were trading well
below par. The Treasury insisted on accepting local cur
rency at par in payments to the federal government, and
on disbursing such currency only where it would be ac
cepted at par. This forced it into temporary default on its
debt service payments, despite its large balances of midAtlantic bank notes. See Catterall (1902),pp. 4-7, and Ban
croft (1831), pp. 47-49.
FEDERAL RESERVE BANK OF ST.
LOUIS
responsibility of receiving and clearing the pay
ments, found itself wedged very uncom fortably in
betw een. If the Bank accepted discounted state
bank notes at par for governm ent payments, the
Treasury would insist that it clear them at par,
and the Bank would take large exchange losses. If
it did not accept state bank notes at all, it would at
the very least offend the people of the South—a
region w h ere it greatly desired to extend its busi
ness—and might well materially increase their
econom ic troubles. The Bank would also offend
the Treasury, its patron, w hich wished to ensure
that federal payments could be made in currency
readily available to the public.102 Finally, if the
Bank accepted state bank notes at th eir m arket
rates of discount, it would be accepting a situation
which, it believed, prevented it from operating
profitably and effectively in the region.103
The Bank’s problem s would have been solved if
the southern banks had resum ed specie payments
promptly after the Panic. W hen they did not, it
launched a campaign to force them to do so. It
continued to accept discounted state bank notes in
payment of federal debts and, w hen it had accu
mulated them in large quantities, presented them
at the counters of the state banks fo r payment.
W hen the southern banks refused to pay, the U.S.
Bank filed suit against them in federal cou rt.104
While the suspensions had no formal legal validi
ty—in principle, each bank note was redeemable at
par and on demand, and a bank which declined to
redeem its notes had defaulted on its debts—they
were implicitly (and sometimes explicitly) tolerated
by state legislatures and courts. The U.S. Bank, how
ever, was in a position to sue in federal courts, which
provided the state banks no such protection.103
102For a general description of the U.S. Bank’s problems as
a federal collection and clearing agent, see Catterall
(1902). For accounts of its disputes with the banks of Ge
orgia and North Carolina, see Govan (1937), Heath (1954),
Holder (1937) and Russell (1989a). These accounts are
based on correspondence between the U.S. Bank, the
Treasury Department, and the state banks that is recorded
in the American State Papers, Finance, Volume 4.
103See note 90 above.
104Govan (1937) and Russell (1989a) describe federal court
suits filed by the U.S. Bank against state banks in Georgia
and North Carolina, respectively.
105See Hammond (1957), pp. 283-84, and Russell (1989a),
pp. 42-43. The efforts of various southern and western
states (notably Georgia) to evade adverse federal court de
cisions concerning the U.S. Bank were ultimately rejected
by the U.S. Supreme Court. See Catterall (1902), pp. 88-91,
Govan (1937), and Hammond (1957), pp. 263-68, 272-73.
53
The efforts of the U.S. Bank slowly forced the
banks of the various southern states to resume
payments: Virginia and South Carolina in 1823,
Georgia in 1825, Alabama in 1827 and North Caro
lina in 1828. In view of the conventional wisdom
regarding suspensions, it should come as no sur
prise that most economic historians have regarded
the Bank’s resumption campaign as virtuous and
constructive; the Bank is lauded, in particular, for
having created a “uniform national currency.’’106
Southern farmers, and other southerners whose
livelihood was based, directly or indirectly, on
farming—groups which collectively comprised the
bulk of the region’s population—had less reason to
sing the Bank's praises. The cost of resumption was
that state banking systems (or those portions of
them which survived) became reluctant to lend to
farmers. Since there were few alternative sources
of credit available, farm loans becam e substan
tially m ore difficult to obtain.107 It should thus be
equally unsurprising that, just a few years later,
the southern states w ere in the forefron t o f the
opposition to the Bank’s efforts to secure a new
ch a rte r.108
A B a n k W ar a n d th e B is e o f F r e e
B a n k in g
S y n o p s is : The political controversy that led to
the demise of the Bank of the United States had a
profound effect on public attitudes tow ard the
banking and cu rren cy system. The American
public becam e suspicious of any hint of monopoly
pow er in banking and of any link betw een the
federal governm ent and private banks. It also
becam e increasingly devoted to the concept of
hard money. One outgrow th of these attitudes
was that many states adopted laws providing for
“Free Banking.” Free banking laws encouraged
entry into banking, and resulted in the establish
ment of large num bers of banks. The free banks
w ere heavily regulated, how ever; their notes w ere
to be carefully secured, and strictly convertible.
O ther results of changed public attitudes w ere the
"Specie Circular” and the "Independent T reasu ry.”
The federal government became reluctant to accept
106See in particular Catterall (1902), Redlich (1951), Ham
mond (1957) and Temin (1969).
107For the case of North Carolina, see Russell (1989a), pp.
71-74. In 1828, Georgia established the Central Bank, a
state-owned institution designed to extend long-term loans
to farmers and planters. Heath (1954) ascribes this deci
sion to a shortage of long-term commercial bank credit
that developed during the 1820s.
b an k c u r re n c y in p aym en t, and attem p ted to
conduct its financial operations w ithout the aid of
banks.
The election of 1828 tran sferred control over
the federal adm inistration from the Whigs, led by
defeated President Joh n Quincy Adams of Mas
sachusetts, to the Dem ocrats, led by President
elect Andrew Jackson of Tennessee. The Demo
crats regarded them selves as the party o f the com
mon (and thus largely agricultural) people; they
had long been advocates of competitive, decentral
ized state banking and opponents of the Banks of
the United States. T he party also contained a hard
money faction which was deeply suspicious of
banking of any kind. Jackson him self seems to
have had somewhat ambiguous feelings tow ard
banking. On one hand, many of his principal
advisors w ere m en who had defended the relief
banks and bitterly resented the damage the U.S.
Bank had done to the banks and people of the
w estern states. Some of these men w ere now con
nected with the state banks, and thus tended to
form ulate policies w hich favored their interests
vis-a-vis those of the U.S. Bank. On the oth er hand,
Jackson is said to have been personally opposed to
banking and paper curren cy of any sort; late in his
administration he took actions that greatly in
creased the problem s of the state banking
system s.109
Jackson ’s State o f the Union message in 1829
cam e out against the rech a rter of the United
States Bank. (The Bank's ch arter did not expire
until 1836, but its friends in Congress had begun
to agitate fo r an early rech arter act.) Jackson
argued that the Bank’s constitutionality was
doubtful, and that its concentrated financial
pow er was inconsistent with the tenets of
representative dem ocracy. Jackson ’s message
m arks the beginning of the "Bank W a r,” a period
of five years or so during w hich the W higs (who
controlled Congress) attem pted to defend the
Bank against the increasingly vituperative attacks
of the Dem ocratic adm inistration. During this
period, the "Bank question” becam e the single
biggest issue in national politics. The Dem ocrats
108Catterall (1902), pp. 164-65, 235, Hoyt (1960) and Wilburn
(1967), pp. 7-11, 17-19, describe southern support for the
Bank War.
109For descriptions of Jackson’s attitude toward banking in
general, and the United States Bank in particular, see Cat
terall (1902), pp. 182-85, Schlesinger (1953), pp. 76-78,
Redlich (1951), pp. 162-71, Hammond (1957), pp. 346-50,
and Temin (1969), chapter 2.
SEPTEMBER/OCTOBER 1991
54
simultaneously exploited and encouraged public
feeling against the banks so as to build up their
political pow er base at the Whigs' expense. After
Jackson was re-elected in 1832 the issue cam e to a
head; Congress passed legislation rech artering the
Bank; Jackson vetoed the legislation, and Congress
failed (narrowly) to override his veto. The second
U.S. Bank effectively ceased to exist as a national
institution w hen its federal ch a rter expired in
1836; its relationship w ith the federal governm ent
had been severed, and its pow er thus greatly
reduced, two years earlier.110
The demise o f the U.S. Bank raised two im por
tant questions: how should the federal govern
m ent adm inister its financial affairs, and how
should the states regulate their banking systems?
T he first question arose because the U.S. Bank had
acted as the financial agent fo r the U.S. T reasury—
in other words, as the federal governm ent’s bank.
The second arose because the U.S. Bank no longer
existed to regulate the state banks and, in particu
lar, to en force specie payments. By the second half
of the 1830s, Jacksonian hard-money notions had
becom e so pervasive that few people w ere p re
p ared to a c c e p t a re tu r n to rela tiv ely la iss e z f a i r e b an k in g , o r to a c u r re n c y w h ich m ight
com e to co n sist larg ely of in co n v ertib le b an k
n o tes. T h e e ffe c t of th e Ja c k so n A d m in istra
tio n ’s an ti-U .S. B an k cam p aig n on a fin a n cia lly
u n so p h istica ted p u blic had b een to e x a c e rb a te
its dou bts ab ou t and su sp icion of all b a n k s and
all p a p er c u r r e n c y .111
The Jackson and Van Buren Administrations
[Martin Van Buren, who was elected president in
1836, had been Jackson’s vice president] responded
to the first problem w ith tw o policy initiatives.
The first was an executive order called the Specie
Circular, issued in 1836, shortly b efore Jackson
left office. This ord er directed the U.S. Treasury
to accept no cu rren cy other than specie in pay
m ent of debts to the federal governm ent.112 The
second policy initiative was the Independent T rea
sury Act, a product of the Van Buren Administra
110For accounts of the Bank War, see Catterall (1902), Ham
mond (1957) and Schlesinger(1953), among many others.
111For a description of the Loco Focos (formally, the Equal
Rights Party), a party of antibanking radicals which arose
during the Bank War and was ultimately instrumental in the
rise of Free Banking and the Independent Treasury, see
Redlich (1951), pp. 188-90, Hammond (1957), pp. 493-99,
and Schlesinger (1953), chapters XV-XVI.
" 2For analyses of the motivation behind, and/or the impact of,
the Specie Circular, see Hammond (1957), pp. 455-57,
Schlesinger (1953), pp. 129-31, Smith (1953), p. 185, Temin
Digitized for FEDERAL
FRASER RESERVE BANK
OF ST. LOUIS
tion. This Act w ithdrew the U.S. governm ent’s
cash deposits from the large state banks (some
times called pet banks) w here they had been placed
after the demise of the U.S. Bank. H enceforth, the
Treasury would act as its own banker. The net
effect of these tw o actions was profoundly anti
banking. The federal governm ent would no longer
deal with the state banks, or encourage (or even
recognize) their note circulation.113
The states’ response to the problem o f bank
regulation reflected tw o features of public atti
tudes tow ards banking w hich had grow n out of
the Bank W ar: suspicion of concentrated financial
pow er, and p referen ce fo r "hard m oney.” Under
the bank chartering system that existed in most
states prior to the late 1830s, the issuance of a
ch arter required a special act of the state legisla
ture. T he process of securing legislative assent
was lengthy, cum bersom e, uncertain and occa
sionally corrupt; in addition, banks that already
possessed ch arters generally lobbied vigorously
against the issuance of new ones. The upshot was
that most states had a relatively small num ber of
banks, and that these banks possessed, or w ere
believed to possess, considerable m arket p ow er.114
The "fre e ” in the free banking laws reflected the
desire to reform the chartering process in the
direction of free entry. Each state would form u
late a standardized ch arter, and any individual or
group w hich was able and willing to m eet the
term s of this ch arter could organize a bank in that
state. The free banks would be regulated by the
state auditor, or by a state banking agency created
for that purpose; the legislature would be involved
only indirectly. This system was intended to
greatly increase the num ber of state banks.115
T h ere is another sense in which the term "free
banking” is a m isnomer, how ever. Free banking
was not in any sense la issez -fa ire or unregulated
banking. The standardized ch arters imposed
num erous and relatively stringent restrictions on
the banks’ capitalization and reserves, the condi-
(1969), pp. 120-28, Timberlake (1960) and Timberlake (1978),
chapter 5.
113For discussions of the role of the Independent Treasury see
Hammond (1957), pp. 542-45, Taus(1943) and Timberlake
(1978), chapter 6.
114See Hammond (1963), pp. 7-8, (1957), pp. 577-80, Redlich
(1951), pp. 188-90, and Schlesinger (1953), p. 286.
115For discussions of the rationale behind Free Banking see
Redlich (1951), pp. 187-204, and Hammond (1957), pp.
572-80.
55
tions under w hich they could issue notes, and the
types of assets they could hold. These restrictions
were designed, in the “hard money” spirit, to ensure
that the banks' notes would always be convertible
on demand, and that they would be relatively
immune from losses associated w ith declines in
the value o f the banks' assets. One restriction was
so common that it has com e to be regarded as
characteristic of free banking: the notes of the
free banks had to be 100 p ercen t backed by hold
ings of state or federal governm ent secu rities.116
Free banking experienced some problem s, espe
cially during its early years. It seems in particular
to have been characterized by a relatively high
rate of bank failures. These failures, and a few
notorious instances of fraud, have given the sys
tem a bad reputation among historians.117 There
w ere also many complaints that the need to keep
track of the m arket values of the many different
types of bank notes in circulation—some of which
w ere counterfeit or issued by failed or insolvent
banks—materially reduced the effectiveness of
bank curren cy as a medium of exchange.118
Recent research has revealed that the losses to
noteholders associated with free bank failures
w ere actually quite small on average.119 Ironically
enough, many of the losses and failures that did
occu r w ere caused by defaults on state govern
m ent bonds, w hich w ere w idespread during the
1840s and 1850s.120 Entrepreneu rs responded to
the diversity of bank curren cies by publishing
“b an k n o te r e p o r te r s ” in w h ich th e value o f
different bank curren cies w ere recorded, insol
vent banks identified, and the appearance of
com m on counterfeits described.121 The system
w orked well enough that, until the Civil W ar
broke out, th ere seem s to have been little political
support for any federally dominated alternative.
Additional evidence that contem porary legislators
116See Rockoff (1975), pp. 2, 81-87, and Rolnick and Weber
(1984), (1988).
l17For unsympathetic descriptions of the record of Free
Banking, or references to such descriptions, see Cagan
(1963), pp. 19-21, Friedman (1960), p. 6, Hammond (1957),
pp. 723, 741-42, (1963), Rockoff (1975), pp. i-ii, and Rolnick
and Weber (1983).
119See Cagan (1963), pp. 19-20, Hammond (1957), pp. 722-23,
(1963), p. 14, and Rockoff (1975), pp. 26-29.
' 19Rockoff (1975) and Rolnick and Weber (1983,1988) use data
drawn from state auditors’ reports to provide careful esti
mates of the costs of free bank failures. Rockoff estimates
that the total losses endured by free bank noteholders from
1836 through 1860 were less than $2 million—about the cost,
viewed free banking as viable is provided by the
fa c t th a t n ew states co n tin u e d to adopt fre e
banking laws through the early 1860s. Rockoff
(1975) notes that “on the eve of the Civil W ar over
half the states, including the most populous states,
had free banking law s.”122
T h e Civil W ar a n d th e D em ise o f
State B a n k C u r re n c y
S y n o p s is : The changes in the American cu r
rency system produced by the Civil W ar w ere
almost as profound as those produced by the Rev
olutionary W ar. W hen the Civil W ar began, the
U.S. had a relatively decentralized system in which
paper cu rren cy was issued by state-chartered and
-regulated banks; the federal governm ent had no
role in the provision of paper money, and there
seemed little prospect that it would acquire one.
W hen the w ar ended, the nation had a relatively
centralized system in w hich paper cu rren cy was
issued by federally ch artered and regulated banks;
it was no longer possible for state banks to issue
paper money. In addition, the federal governm ent
had acquired and used the power to issue paper
money and to m ake it legal tender. This huge,
rapid transform ation was possible for tw o rea
sons. First, the w ar produced a dram atic shift in
the balance of pow er betw een the m ajor political
parties, and betw een state and federal govern
ments. Second, the w ar produced an unp rece
dented need for federal governm ent revenue.
The demise o f free banking, and m ore broadly
of the system of decentralized, state-regulated
provision of curren cy, was caused less by any
problem s this system may have experienced than
by the outbreak of the U.S. Civil W ar. This extra
ordinary political event created a pair of peculiar
circum stances w hich helped determ ine the future
in 1860, of a single year of 2 percent inflation. The estimates
provided by Rolnick and Weber are slightly higher (see their
1983 paper, p. 1089). Cagan (1963), by contrast, reports with
apparent endorsement an estimate by Jay Cooke that the
losses were $50 million per year. Cagan’s use of this esti
mate in his influential article served to reinforce the conven
tional view that free banking was a national disaster.
120See Rockoff (1975) and Rolnick and Weber
(1982,1983,1984,1985,1988).
121See Hammond (1963), p. 14, Temin (1969), p. 50, and
Rockoff (1975), pp. 23-25.
122Rockoff(1975), pp. 2-4.
SEPTEMBER/OCTOBER 1991
56
evolution of the U.S. cu rren cy and banking sys
tem. The first was that it gave the Republican
party, the political successor of the old Whig party,
almost com plete control over the federal govern
ment. The m ajority o f the Dem ocratic congress
men and senators w ere southerners who abandoned
their seats and defected to the Confederacy at the
outbreak of hostilities. Since the Republicans had
inherited from the W higs a p referen ce for m one
tary and financial centralization, while the Demo
crats rem ained the party of decentralization, the
defection o f the D em ocrats greatly increased the
prospects fo r centralizing change in the m onetary
system.
The second determ ining circum stance was the
federal governm ent’s need for enorm ous new
sources of revenue. Though taxes played some
role in Civil W ar finance, particularly in the w ar’s
later years, it was clear almost from the beginning
that the bulk of the expenditure burden would be
m et by borrow ing.123 As the w ar’s cost mounted,
how ever, the federal governm ent began to experi
ence difficulty obtaining the sums required on the
open securities m arket, and turned for assistance
to the large Eastern banks.124 Unfortunately, the
sum required by the governm ent far exceeded the
banks' aggregate specie holdings; when the specie
borrow ed by the governm ent failed fo r various
reason to retu rn to the banks as rapidly as they
had anticipated, they w ere forced to suspend spe
cie paym ents.125 T he suspension seemed to close
o ff the possibility of fu rth er bank loans, leaving
the federal governm ent desperate for new sources
of funds and for new ways to increase the demand
for its debt.
One way in w hich the governm ent might "b o r
row ” was by issuing paper currency. Early in 1862,
at the urging of Secretary of the Treasury Salmon
123For the government’s reluctance to raise taxes during the
early years of the war, see Mitchell (1903), pp. 16-19, p. 37,
pp. 72-73.
124The federal government’s financing problems were greatly
exacerbated by its reluctance to sell its securities, which typi
cally returned interest at the rate of 6 percent of face value, at
a discount—its reluctance, that is, to borrow at interest rates
in excess of 6 percent. See Mitchell (1903), pp. 48-50, 64-65.
125For accounts of the suspension see Mitchell (1903), pp. 37-43
and Hammond (1970), pp. 150-59.
126The provisions of the first and the two subsequent Legal
Tender Act(s) are summarized by Mitchell (1903), pp. 44-118.
127ln 1875,10years after the end of the war, Congress enacted
legislation making the greenbacks convertible in specie,
according to the prewar definition of the dollar, beginning in
1879. This legislation became known, somewhat mislead
Digitized forFEDERAL
FRASER RESERVE BANK OF ST.
LOUIS
P. Chase, Congress passed the first Legal Tender
Act. The Act authorized the T reasury Departm ent
to issue $300 million in paper cu rren cy .126 This
curren cy was not convertible in specie, nor re
deem able in specie at any fixed future date; it was,
however, made legal tender in paym ent of public
a n d p r iv a te d e b ts .127 L a te r in th e w ar, c o n s id e r
ab le ad d itional q u an titie s of th e se g re e n b a ck s
w ere issued.
The Legal Tender Acts m arked the first time in
(post-revolutionary) U.S. history that the federal
governm ent had issued fiat cu rren cy —curren cy
which was entirely irredeem able, and was legal
tender for private debts. The Constitution did not
give the federal governm ent any explicit right to
issue paper cu rren cy (fiat or otherwise), or to
make paper cu rren cy legal tender; indeed, it con
tained language which was widely interpreted as
implicitly denying the governm ent these rights.128
This made the legality of the Acts seem very
doubtful. In the event, how ever, no attem pt to
challenge them managed to reach the Suprem e
Court until several years after the Civil W ar had
ended. T h ere ensued one of the m ore bizarre
episodes in U.S. m onetary history. Chase, who was
now Chief Justice, voted with the cou rt m ajority
to strike down legislation whose form he had ap
proved, and whose passage he had recom m ended,
w hen he was Secretary of the Treasury! The
greenbacks w ere saved w hen Congress, w hich
rem ained dominated by the Republicans, voted to
increase the num ber o f Supreme Court justices by
two; President Grant acted quickly to fill the re
sulting vacancies with judges who supported the
Acts' constitutionality. The enlarged court voted
5-4, w ith Chase dissenting, to uphold the A cts.129
This decision set a precedent that was later used
to justify fu rth er steps on the part of the federal
governm ent to regulate or control the curren cy
ingly, as the “ Resumption Act.” [The history and provisions
of the Resumption Act are summarized by Friedman and
Schwartz (1963), pp. 44-50. See also Timberlake (1978),
chapter 8.]
128For discussions of the constitutionality of legal tender federal
currency, see Breckinridge (1903), pp. 114-37, Mitchell
(1903), pp. 51-55, Hurst (1970), pp. 182-86, Christainsen
(1988), Timberlake (1989), (1991), chapters 8-10.
' 29See Breckinridge (1903), pp. 127-137, Friedman and
Schwartz (1963), pp. 46-47, Mitchell (1903), pp. 71-74, and
Timberlake (1989), (1991), chapter 9.
57
sy ste m —n otably, the c re a tio n o f th e F ed eral
Reserve System.
The greenbacks m arked a change in the U.S.
curren cy system which w ent beyond governm ent
issuance of currency. As has already been noted,
the fiscal crisis of late 1861 caused the private
banking system to suspend specie payments. On
earlier occasions w hen there had been a national
or regional suspension, most of the surviving
banks had refused to redeem their notes in any
way until they could resum e redeem ing them in
specie on demand. The fact that the greenbacks
w ere legal tender, how ever, gave bank debtors
the legal right to use them to repay their debts.
Since greenbacks began trading at substantial
discounts shortly after they w ere first issued,
m oreover, debtors hastened to take advantage of
this opportunity. The banks consequently felt enti
tled, and indeed compelled, to redeem their notes
in greenbacks.130 T h ereafter, the banking system
redeem ed its notes and deposits almost exclusively
in legal tend er paper currency, regardless of
w hether it was convertible in specie. This ensured
that goods and assets would be priced in legal
tend er paper—in other words, that governm ent
cu rren cy would replace specie as the nation’s unit
of account.
Because neither Congress nor the Administra
tion was willing to risk a repetition of the Revolu
tionary hyperinflation, greenbacks could be used
to finance at most a small portion of the w artim e
deficit.131 T he balance of the deficit had to be
financed by the issuance of conventional, interestbearing debt.132 Secretary Chase responded to this
situation by developing a strategy fo r m onetary
reform w hich promised to simultaneously achieve
both cu rren cy centralization and debt demand
enhancem ent. This program , w hich was ultim ate
ly embodied in the National Banking Act, called
for the creation of a "National Banking System ”
(NBS)—a system of private, federally chartered
banks which would be nationally regulated ana
logues of the state-chartered “free banks.” The Act
imposed reserve, capital, convertibility and other
requirem ents that w ere generally similar to those
>30See Mitchell (1903), pp. 144-49.
131The limited financial resources of the southern states made it
difficult for the Confederate government to borrow large
sums. As a result, it was forced to cover a very substantial
fraction of its deficit through currency creation. This strategy
ultimately produced a hyperinflation. See Lerner (1956), and
Timberlake (1978), pp. 102-03.
132Mitchell (1903), pp. 119-31, discusses the reasons the federal
government issued no greenbacks from 1863 to the end of
the Civil War.
imposed on the free banks. These requirem ents
w ere to be adm inistered, and the national banks
regulated, by a new federal agency called the
O ffice of the Com ptroller of the C urrency.133 For
the first time in U.S. history, the federal govern
m ent had moved to create a system of private
banks (rather than a single, centrally administered
private bank) under its direct regulatory control.
From the perspective of the cu rren cy system,
the key features o f the National Bank Act involved
the notes the national banks w ere to issue. These
notes w ere to be printed by the Treasury Depart
ment and issued to the banks, rath er than printed
by the banks directly; they w ere to look entirely
uniform , except for an indication of the identity of
the issuer. In order to obtain a given value of bank
notes, a national bank had to deposit U.S. govern
ment securities of essentially equal value (state
governm ent securities would not do) with the
Com ptroller of the Currency. Thus, national bank
notes w ere to be 100 percent backed by U.S. gov
ernm ent securities.
The requirem ent th at the notes be backed by
fe d e r a l governm ent securities was designed to
create a “captive” demand fo r federal debt on the
part of banks of issue. Since notes w ere the prin
cipal liabilities of contem porary banks, and since
the fram ers of the Act evidently expected most of
the state banks to apply for federal charters, there
was every reason to expect that the Act would
force the banking system to purchase Treasury
securities in large quantities. This, it was hoped,
would materially ease the federal governm ent’s
borrow ing problem s.134
Congress, anticipating heavy demand for nation
al bank notes, included provisions in the Act estab
lish in g a m axim um q u an tity w h ich could be
issued and allocating this quantity across the
various regions.135 Contrary to expectation, how
ever, during the year or so after the Act was passed
the num ber of ch arte r applications was small,
and the volume of U.S. bonds deposited as note
backing was far low er than anticipated. Congress
133Forthe history and provisions of the National Banking Act,
see Redlich (1951), pp. 99-113 and Hammond (1970),
chapters 10-11.
134Discussions of the budgetary motivations for the National
Banking Act can be found in Hammond (1957), p. 725,
(1970), pp. 135-36, 286-95, and Mitchell (1903), pp. 37,
44-45, 103, 109.
135These provisions are summarized by Redlich (1951), p. 118.
SEPTEMBER/OCTOBER 1991
58
responded by amending the Act to impose a puni
tive tax on state bank notes—a tax rate so high (10
percent) that it made note issue by state banks
entirely u nprofitable.136 This decision eliminated
state bank notes from circulation, and m arked the
final demise of state curren cy systems.
Of course, a system under w hich any curren cy
that was not issued directly by the federal govern
m ent was printed by the federal governm ent, fully
backed by U.S. Treasury securities, and issued in
quantities and locations closely regulated by the
governm ent, might be said to have differed very
little from a system under w hich the federal gov
ernm ent directly issued all paper cu rren cy .137
Indeed, it could be argued that the only really
significant differences betw een the NBS and a
direct note issue system w ere that under the NBS,
the governm ent (1) had little short-run influence
over the total quantity of notes (which indeed
proved relatively unresponsive to short-run influ
ences of any kind), and (2) assigned the responsi
bility fo r clearing the notes (and thus fo r ensuring
their convertibility in specie) to the issuing banks.
W hen 50 years o f experience with the system
seemed to suggest that these features w ere serious
liabilities, the federal governm ent used the broad
m onetary pow ers it had acquired during the Civil
W ar to establish a system of direct issue—the Fed
eral Reserve System.
CONCLUDING REM ARKS
The changes in the U.S. curren cy system that
resulted from the Legal T en d er and National Bank
ing acts stand, along w ith the m onetary clauses of
the U.S. Constitution, as classic examples of cases
in which the basic stru ctu re of the system was
strongly influenced by extraordinary political
events with largely non-econom ic (or at least, non
monetary) causes. If, as we have seen, the American
colonies could have obtained their independence
from Great Britain without fighting a long, expensive
and divisive revolutionary war, the monetary his
tory of the next 90 years might have been very
different: historical evidence suggests that the
136See Friedman and Schwartz (1963), pp. 18-19, Hammond
(1957), pp. 732-34, and Redlich (1951), p. 113.
137This point is made by Friedman and Schwartz (1963), p. 21.
138See note 89 above.
139Redlich (1951) comments that had Secretary Chase
promoted the National Banking Act less vigorously, the early
Digitized for FEDERAL
FRASER RESERVE BANK
OF ST. LOUIS
states might have retained the right to issue their
own currencies, and that these curren cies might
not have been convertible in specie. Subsequently,
if the American states could have resolved their
sectional disputes without fighting a long, expen
sive and divisive civil war, the m onetary history of
the ensuing 50 years might also have been very
different: historical evidence suggests that the
states might have retained the right to ch arter
banks o f issue, and that the federal governm ent’s
role in the U.S. cu rren cy system might have
rem ained relatively limited.
It should also be remembered that the federal
government chartered the second United States
Bank in response to financial dislocations associated
with the W ar of 1 8 1 2 .138 In the absence o f Second
Bank opposition, the inconvertible banking sys
tem s that arose in the w estern and southern states
after the Panic of 1819 might have survived and
becom e entrenched; without a Second Bank for
the Jacksonians to fight, the "hard m oney" prin
ciples of the Free Banking Era might never have
becom e popular. Here again a sequence of essen
tially political disputes played a key role in dictat
ing the evolution of U.S. currency arrangements.139
C u r re n c y S y stem E v o lu tio n : An
A lterna tiv e View
As we have seen, U.S. m onetary history has
been punctuated by a sequence of rath er abrupt
transitions from one curren cy system to another
with very different features. These transitions are
often in terp reted as p a rt of a process of Darwinian
advancement—a process, that is, through which
old and relatively inefficient systems w ere replaced
by new and m ore efficient successors. The modern
cu rren cy system em erged out of this process as
the most efficient system yet devised.
W hile this historical interpretation certainly
sounds plausible, it is one that we should accept or
reject on the basis of evidence concerning the rela
tive efficiency of past and presen t cu rren cy sys
tems. Unfortunately, the prestige of Darwinism
Civil War suspension also might have led to the development
of a currency system based on inconvertible bank notes
(p. 95).
59
has becom e so great that econom ists tend to
reverse the logical process by using the various
system s’ orders in the historical sequence as the
basis for efficiency com parisons. (If only the fittest
systems survive, then the systems that survived at
each stage must have been the fittest.) The disap
pearance of older systems is regarded as com pel
ling evidence that they w ere less efficient than
their successors.
The claim that the cu rren cy system, if left to
itself, tends to progress (slowly) in the direction of
greater efficiency is not disputed in this article;
indeed, several examples of this sort of progres
sion have been presented above. W hat the article
has argued is that the U.S. cu rren cy system has
not been left to itself, and that its evolution has
been anything but an orderly and inevitable
progression tow ard econom ic efficiency. Instead,
it has been dominated by political decisions that
w ere largely uninfluenced by efficiency consider
ations. Many of these decisions w ere made in
response to political pressures of a particularly
urgent sort—pressures growing out of the U.S.
governm ent’s (and earlier, the British and/or
colonial governm ents’) involvement in prolonged
and expensive wars.
It is, of course, possible that we have been
fortunate, and that the political process has given
us a cu rren cy system that is very efficient, or at
least m ore efficient than the historical alterna
tives. It is also possible that w e have not been quite
so fortunate; the question is a complex one, and
cannot be answ ered here. This article is content to
point out that the m odern cu rren cy system has
not developed because of any clear advantage in
efficiency. The possibility that history provides
attractive alternatives cannot be ruled out, and the
question of which system is best rem ains both
open and interesting.
REFERENCES
Arnold, Samuel G. History of the State of Rhode Island and
Providence Plantations, vols. I and II, (D. Appleton, 1878).
Bancroft, George. “ Bank of the United States,” North
American Review (January 1831), pp. 21-64.
Bates, Frank Greene. “ Rhode Island and the Formation of
the Union,” Columbia University Studies in History,
Economics and Public Law, vol. X, no. 2 (The MacMillan
Company, 1898).
Berry, Thomas S. Western Prices Before 1861 (Harvard
University Press, 1943).
Billias, George A. “ The Massachusetts Land Bankers of
1740,” University of Maine Studies, Second series, Number 74
(Maine University Press, 1959).
Breckinridge, S. P. Legal Tender: A Study in English and
American Monetary History (University of Chicago Press,
1903).
Brock, Leslie V. The Currency of the American Colonies
1700-1764 (Arno Press, 1975).
Bullock, Charles J. Essays on the Monetary History of the
United States (The MacMillan Company, 1900).
Cable, John R. “ The Bank of the State of Missouri,” Studies
in History, Economics and Public Law, vol. 102 (Columbia
University Press, 1923).
Cagan, Phillip. “ The First Fifty Years of the National Banking
System—An Historical Appraisal,” in Deane Carson, ed.,
Banking and Monetary Studies (Richard D. Irwin Inc., 1963).
Calomiris, Charles W. “ Institutional Failure, Monetary Scarci
ty, and the Depreciation of the Continental,” Journal of
Economic History (March 1988), pp. 47-68.
Calomiris, Charles W., and Charles M. Kahn. “ The Role of
Demandable Debt in Structuring Optimal Banking Arrange
ments,” American Economic Review (June 1991), pp. 497-513.
Carothers, Neil. Fractional Money (Augustus M. Kelley, 1930),
(reprinted 1967).
Catterall, Ralph C. H. The Second Bank of the United States
(University of Chicago Press, 1902).
Checkland, S.G. Scottish Banking: A History, 1695-1973
(Glasgow: Collins, 1975).
Christainsen, Gregory B. “ Fiat Money and the Constitution: A
Historical Review,” in Thomas D. Willett, ed., Political Busi
ness Cycles (Duke University Press, 1988).
Clapham, Sir John. The Bank of England, A History, vols. I
and II (Cambridge University Press, 1944), (reprinted 1958).
Clough, Shepard B. and Charles W. Cole. Economic History
of Europe (Heath, 1941).
Cole, Arthur H. Wholesale Commodity Prices in the United
States, 1700-1861 (Harvard University Press, 1938).
Davis, Andrew McFarland. “ Currency and Banking in the
Province of Massachusetts-Bay,” American Economics
Association Publications, Third series, Part I: “Currency,”
Volume I, Number 4, December 1900, Part II: “ Banking,”
Volume II, Number 2, May 1901 (The Macmillan Company).
Dewey, Davis R. State Banking before the Civil War, National
Monetary Commission, (U.S. Government Printing Office,
1910).
Dorfman, Joseph. The Economic Mind in American Civiliza
tion, 1606-1865, vols. I and II (London: George G. Harrap &
Company LTD., 1947).
Dowd, Kevin. “ Option Clauses and the Stability of a Laissez
Faire Monetary System,” Journal of Financial Services
Research (December 1988), pp. 319-33.
Dowrie, George W. “The Development of Banking in Illinois,
1817-1863,” Studies in the Social Sciences, vol. XI, no. 4
(University of Illinois Press, December 1913).
Dwyer, Gerald P., and R. Alton Gilbert. “ Bank Runs and
Private Remedies,” this Review (May/June 1989), pp. 43-61.
Ernst, Joseph A. Money and F>olitics in America, 1755-1775
(The University of North Carolina Press, 1973).
Felt, Joseph B. An Historical Account of Massachusetts Cur
rency (Bun Franklin, 1839).
Fenstermaker, J. Van. The Development of American Com
mercial Banking: 1782-1837 (Kent State University Press,
1965a).
_______ . “ The Statistics of American Commercial Banking,
1782-1818,” Journal of Economic History (1965b), pp. 400-13.
SEPTEMBER/OCTOBER 1991
60
Ferguson, E. James. “ Currency Finance: An Interpretation of
Colonial Monetary Practices,” William and Mary Quarterly,
3rd series, vol. X (1953), pp. 153-80.
________The Power of the Purse (University of North Caroli
na Press, 1961).
Feavearyear, Sir Albert. The Pound Sterling: A History of
English Money (Oxford at the Clarendon Press, 1963).
Flanagan, Beecher. “A History of State Banking in North
Carolina to 1866,” Doctoral dissertation, George Peabody
College for Teachers, 1935.
Fraas, Arthur. “ The Second Bank of the United States: An
Instrument for an Interregional Monetary Union,” Journal of
Economic History (June 1974), pp. 447-67.
Franklin, Benjamin. Essays on General Politics, Commerce
and Political Economy, Volume II, Part II of Jared Sparks, ed.,
The Works of Benjamin Franklin (Augustus M. Kelley, 1971).
Friedman, Milton. A Program for Monetary Stability (Fordham
University Press, 1960).
Friedman, Milton, and Anna J. Schwartz. A Monetary History
of the United States: 1867-1960 (Princeton University Press,
1963).
Gallatin, Albert. The Writings of Albert Gallatin, Henry Adams,
ed., vol. Ill (J. B. Lippincott & Company, 1879).
Gouge, William M. A Short History of Paper-Money and Bank
ing in the United States (T. W. Ustick, 1833).
Gould, Stephen Jay. Wonderful Life: The Burgess Shale and
the Nature of History (W. W. Norton & Company, 1989).
Govan, Thomas P. “ Banking and the Credit System in
Georgia, 1810-1860,” Doctoral dissertation, Vanderbilt Univer
sity, 1937 (Arno Press, 1978).
Greene, Jack P., and Richard M. Jellison. “ The Currency Act
of 1764 in Imperial-Colonial Relations, 1764-1766,” William and
Mary Quarterly, Third series, Volume XVIII (1961), pp. 485-518.
Hammond, Bray. “ Long and Short term Credit in Early
American Banking.” Quarterly Journal of Economics, vol. XLIX
(1934), pp. 79-103.
________Banks and Politics in America from the Revolution
to the Civil War (Princeton University Press, 1957).
________“ Banking Before the Civil War,” in Deane Carson,
ed., Banking and Monetary Studies (Richard D. Irwin Inc.,
1963), pp. 1-14.
________Sovereignty and an Empty Purse: Banks and Poli
tics in the Civil War (Princeton University Press, 1970).
Heath, Milton Sidney. Constructive Liberalism: The Role of the
State in Economic Development in Georgia to 1860 (Harvard
University Press, 1954).
Hepburn, A. Barton. A History of Currency in the United
States (Augustus M. Kelley, 1903), (reprinted 1967).
Holder, Brantson B. “The Three Banks of the state of North
Carolina: 1810-1872,” Doctoral dissertation, University of
North Carolina, 1937.
Holdsworth, John T., and Davis R. Dewey. The First and
Second Banks of the United States (U.S. Government Printing
Office, 1910).
Horsefield, J. Keith. British Monetary Experiments, 1650-1710
(Harvard University Press, 1960).
Hoyt, Elizabeth S. “ Reactions in North Carolina to Jackson’s
Banking Policy, 1829-1832,” North Carolina Historical Review
(1948), pp. 167-78.
Huntington, A. T., and Robert J. Mawhinney. Laws of the
United States Concerning Money, Banking, and Loans,
1778-1909, National Monetary Commission (U.S. Government
Printing Office, 1910).
Digitized forFEDERAL
FRASER RESERVE BANK
OF ST. LOUIS
Hurst, James W. A Legal History of Money in the United
States 1774-1970 (University of Nebraska Press, 1973).
Kemmerer, Donald L. “The Colonial Loan-Office System in
New Jersey,” Journal of Political Economy (December 1939),
pp. 867-74.
King, Robert G. “ On the Economics of Private Money,”
Journal of Monetary Economics (July 1983), pp. 127-58.
Klein, Benjamin. “The Competitive Supply of Money,”
Journal of Money, Credit and Banking (November 1974), pp.
423-53.
Knox, John Jay. A History of Banking in the United States
(Rhodes & Company, 1903).
Lerner, Eugene M. “ Inflation in the Confederacy,
1861-1865,” in Milton Friedman, ed., Studies in the Quantity
Theory of Money (Uriwerstiy of Chicago Press, 1956).
Libby, Orin G. The Geographical Distribution of the Vote of
the Thirteen States on the Federal Constitution 1787-8 (Univer
sity of Wisconsin Press, June 1894).
Macaulay, Thomas B. A History of England from the Ac
cession of James II, vols. IV,V (Longmans, Green, Reader &
Dyer, 1877).
Macon, Hershal L. “A Fiscal History of North Carolina,
1776-1860,” Doctoral dissertation, University of North Caro
lina, 1932.
Metz, William D. “ Politics and Finance in Massachusetts,
1713-1741,” Doctoral dissertation, University of Wisconsin,
1945.
Mitchell, Wesley C. A History of the Greenbacks (University of
Chicago Press, 1903).
Nettels, Curtis P. The Emergence of a National Economy,
1775-1815 (M. E. Sharpe, Inc., 1962).
________The Money Supply of the American Colonies Before
7720 (University of Wisconsin Press, 1934).
Nevins, Allan. The American States During and After the
Revolution: 1775-1789 (Augustus M. Kelley, 1924).
North American Review. “ Bank of the United States”
(April 1831), pp. 524-63.
Phillips, Henry Jr. Historical Sketches of the Paper Currency
of the American Colonies (W. E. Woodward, 1865).
Redlich, Fritz. The Molding of American Banking: Men and
Ideas, Parts I and II (Johnson Reprint Corporation, 1951).
Rockoff, Hugh.“The Free Banking Era: A Re-examination,”
Journal of Money, Credit and Banking (May 1974), pp. 141-67.
________The Free Banking Era: A Re-examination (Arno
Press, 1975).
________ “ Institutional Requirements for Stable Free Bank
ing Regime,” Cafo Journal (Fall 1986), pp. 617-34.
Rolnick, Arthur J., and Warren E. Weber. “ Free Banking,
Wildcat Banking, and Shinplasters,” Federal Reserve Bank of
Minneapolis Quarterly Review (Fall 1982), pp. 10-19.
_______ . "New Evidence on the Free Banking Era,” Ameri
can Economic Review (December 1983), pp. 1080-91.
_______ . “ The Causes of Free Bank Failures: A Detailed Ex
amination,” Journal of Monetary Economics (November 1984),
pp. 267-91.
________“ Banking Instability and Regulation in the U.S.
Free Banking Era,” Federal Reserve Bank of Minneapolis
Quarterly Review (Summer 1985), pp. 2-9.
________“ Explaining the Demand for Free Bank Notes,”
Journal of Monetary Economics (January 1988), pp. 47-71.
________“ Gresham’s Law or Gresham’s Fallacy?” Journal
of Political Economy (February 1986), pp. 185-99.
61
Rothbard, Murray. The Panic of 1819 (Columbia University
Press, 1962).
Russell, Steven. “ Some Implications of the Colonial Currency
Experience,” Working paper, April 1988.
________‘“ Free Banking’ as Inconvertible Banking: North
Carolina, 1819-1828,” Working paper, September 1989a.
_______ . “ Inconvertible Banking in the Early Nineteenth
Century United States: A Legal Restrictions Approach,”
Working paper, October 1989b.
_______ . “ The Pricing of North Carolina Bank Notes,
1818-1827: An Empirical Investigation,” Working paper, May
1989c.
Santoni, G.J. “A Private Central Bank: Some Olde English
Lessons,” this Review (April 1984), pp. 12-22.
Schlesinger, Arthur M. Jr. The Age of Jackson (Little, Brown
and Company, 1953).
Schweitzer, Mary M. “ State-Issued Currency and the
Ratification of the U.S. Constitution,” Journal of Economic
History (June 1989), pp. 311-22.
Scott, William Robert. The Constitution and Finance of
English, Scottish and Irish Joint-Stock Companies to 1720,
Three volumes (Cambridge: The University Press, 1910-12).
Smith, Adam. The Wealth of Nations, Edited and introduced
by Edwin Cannan (Random House, Inc., 1937, originally
published in 1776.)
Smith, Bruce D. “ Money and Inflation in Colonial Mas
sachusetts,” Federal Reserve Bank of Minneapolis Quarterly
Review (Winter 1984), pp. 1-14.
________“American Colonial Monetary Regimes: The
Failure of the Quantity Theory and Some Evidence in Favour
of an Alternative View,” Canadian Journal of Economics
(August 1985a), pp. 531-65.
________“ Some Colonial Evidence on Two Theories of
Money: Maryland and the Carolinas,” Journal of Political
Economy (December 1985b), pp. 1178-211.
Smith, Walter B. Economic Aspects of the Second Bank of
the United States (Harvard University Press, 1953).
Smith, Walter B., and Arthur H. Cole. Fluctuations in
American Business, 1790-1860 (Harvard University Press,
1935).
Sumner, William G. A History of Banking in All the Leading
Nations, Volume I (The Journal of Commerce and Commer
cial Bulletin, 1896).
Sylla, Richard, John B. Legler, and John J. Wallis. “ Banks
and State Public Finance in the New Republic: The United
States: 1790-1860,” Journal of Economic History (June 1987),
pp. 391-403.
Taus, Esther Rogoff. Central Banking Functions of the United
States Treasury, 1789-1941 (Columbia University Press, 1943).
Taylor, Robert J. Western Massachusetts in the Revolution
(Brown University Press, 1954).
Temin, Peter. The Jacksonian Economy (W. W. Norton &
Company, Inc., 1969).
Thayer, Theodore. “The Land-Bank System in the American
Colonies,” Journal of Economic History (Spring 1953), pp.
145-59.
Timberlake, Richard H. Jr. “ The Specie Circular and Distri
bution of the Surplus,” Journal of Political Economy (April
1960), pp. 109-17.
_______ “ Denominational Factors in Nineteenth-Century Cur
rency Experience,” Journal of Economic History (December
1974), pp. 835-50.
_______ . The Origins of Central Banking in the United States
(Harvard University Press, 1978).
________“ The Significance of Unaccounted Currencies,”
Journal of Economic History (December 1981), pp. 853-66.
_______ . “The Government’s License to Create Money,”
Cafo Journal (Fall 1989), pp. 301-21.
________Gold, Greenbacks, and the Constitution (George
Edward Durrell Foundation, 1991).
U.S. Constitution.
Viner, Jacob. Studies in the Theory of International Trade
(Harper & Brothers, 1937).
Wallace, Neil. “A Legal Restrictions Theory of the Demand
for ‘Money’ and the Role of Monetary Policy,” Federal
Reserve Bank of Minneapolis Quarterly Review (Winter 1983),
pp. 1-7.
White, Lawrence H. Free Banking in Britain: Theory, Ex
perience, and Debate 1800-1845 (Cambridge University Press,
1984).
________“Accounting for Non-Interest-Bearing Currency: A
Critique of the Legal Restrictions Theory of Money,” Journal of
Money, Credit and Banking (November 1987), pp. 448-56.
West, Robert Craig. “ Money in the Colonial American Econo
my,” Economic Inquiry (January 1978), pp. 1-15.
Wilburn, Jean Alexander. Biddle’s Bank: The Crucial Years
(Columbia University Press, 1967).
SEPTEMBER/OCTOBER 1991
F e d e ra l R e s e rv e B a n k of St. L o u is
Post Office Box 442
St. Louis, Missouri 63166
The R e v ie w is published six
times p er year b y the Research
and Public Inform ation
Department o f the Federal
R eserve Rank o f St. Louis.
Single-copy subscriptions are
available to the public f r e e o f
charge. Mail requests f o r
subscriptions, back issues, or
address changes to: Research
and Public Information
Department, Federal R eserve
Rank o f St. Louis, P.O. Roy 442,
St. Louis, Missouri 63166.
The views expressed are those
o f the individual authors and do
not necessarily reflect official
positions o f the Federal R eserve
Rank o f St. Louis o r the Federal
R eserve System. Articles herein
may b e reprinted provided the
source is credited. Please provide
the Rank’s Research and Public
Inform ation Department with a
copy o f reprinted material.
@FedFRASER