The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
R
E
Vol. 76, No. 5
W
September/October 1994
The Sectoral Composition of Job Creation
and Destruction
Boom or Bust? The Economic Effects
of the Baby Boom
Realignments of Target Zone Exchange Rate
Systems: What Do We Know?
A Case Study in Monetary Control: 1980-82
THE
FEDERAL
J RESERVE
Jtk BANK of
A r ST. LOUIS
R
E
V
I
E
W
President
T h o m a s C . M e lze r
Review is published six times per year by the Research Department of the
Federal Reserve Bank of St. Louis. Single-copy subscriptions are available
free of charge. Send requests for subscriptions, back issues or address
Director of Research
W illia m G. D e w a ld
Associate Director of Research
C le tu s C. C o u gh lin
Research Coordinator and
Review Editor
W illia m T. G avin
changes to:
Federal Reserve Bank of St. Louis
Public Information Department
P.O. Box 442
St. Louis, Missouri 63166
The views expressed are those of the individual authors and do not neces
sarily reflect official positions of the Federal Reserve Bank of St. Louis, the
Federal Reserve System, or the Board of Governors. Articles may be reprinted
Banking
R. A lto n G ilb e rt
D a vid C. W h e e lo c k
International
C h ris to p h e r J . N e e ly
M ic h a e l R. P a k k o
P a tric ia S. P o llard
or reproduced only if the source is credited. Please provide the Public
Information Department with a copy of the reprinted materials.
All nonproprietary and nonconfidential data and programs for the articles
published in Review are available to our readers. This information can be
Macroeconomics
R ich ard G. A n d erson
J a m e s B. B u llard
M ic h a e l J . D u e k e r
Jo se p h A. R itte r
Jo h n A . T ato m
D a n ie l L. T h o rn to n
P e te r Yoo
obtained from three sources:
1. FR ED (F e d e ra l R e s e rv e E c o n o m ic D a ta ), an e le c tr o n ic b u l
le tin bo ard s e rv ic e . You can access FRED by dialing 314-621-1824
through a modem-equipped personal computer. Parameters should be
set to: no parity, word length = 8 bits, 1 stop bit, and the fastest baud rate
the modem supports (up to 14,400 bps). Information will be in directory
11 under the file name ST. LOUIS REVIEW DATA. For a free brochure
Regional
M ic h e lle A. C la rk
K e vin L. K lie s e n
A d am M . Z a re ts k y
on FRED, please call 314-444-8809.
2. T h e F e d e ra l R e s e rv e B a n k o f S t. Lo uis. You can request
data and programs on either disk or hard copy by writing to: Research
Director of Editorial Services
D a n ie l P. B ren nan
Managing Editor
C h a rle s B. H e n d erso n
Graphic Designer
B ria n D. E b e rt
Department, Federal Reserve Bank of St. Louis, Post Office Box 442,
St. Louis, MO 63166. Please include the author, title, issue date
and page numbers with your request.
3. In tra -u n iv e rs ity C o n s o rtiu m fo r P o litic a l an d S o c ia l
R e s e a rc h (IC P S R ). Member institutions can request data through
the CDNet Order facility. Nonmembers should write to: ICPSR, Institute
for Social Research, P.O. Box 1248, Ann Arbor, Michigan 48106, or
call 313-763-5010.
1
Federal Reserve Bank of St. Louis
Review
September/October 1994
In This Issue. ..
Job Creation and Destruction: The Dominance of Manufacturing
Joseph A. Ritter
Estimates of gross job creation and destruction give a deeper perspective
on the ebb and flow of labor markets than the headline-grabbing announce
ments of net employment growth. Overall employment growth may be the
result of lots of job creation cancelling much of the job destruction—or only
a little of each. Joseph A. Ritter examines how patterns of job creation and
destruction vary between goods-producing and service industries. He finds
that manufacturing and other goods-producing industries have contributed
disproportionately to changes in overall job creation and destruction, par
ticularly during recessions, but that this pattern may have changed recently.
13
Boom or Bust? The Economic Effects of the Baby Boom
P eter Yoo
Between 1947 and 1962, the population of the United States grew at an
average annual rate near 2 percent. This large but temporary increase in
the population growth rate, more familiarly called the baby boom, raises
an interesting and important question: How do such large changes in the
population growth rate affect a developed economy?
To answer this question, Peter Yoo turns to three models of economic growth
that incorporate different aspects of demographic changes. The three models
disagree about the speed and magnitude of such changes, but all show that
after a period of slow growth, per capita consumption increases. Best of all,
the models indicate such improvements in the standard of living occur even
as aggregate saving drops. This suggests, Yoo concludes, that the retirement
of the baby boomers need not imply diminishing standards of living.
23
Realignment of Target Zone Exchange Rate Systems: What Do We Know?
Christopher J. N eely
The recent suspension of the European Union’s Exchange Rate Mechanism
(ERM) has led to extensive discussion on the credibility of target zone
exchange rate systems. Researchers would like to understand the circum
stances associated with speculative attacks and the realignments of target
zones for several reasons. For example, monetary authorities would like
to maintain stable exchange rates and low inflation while retaining sufficient
flexibility to conduct countercyclical stabilization policy.
Christopher J. Neely surveys recent work on forecasting realignments and
estimating the credibility of target zone exchange rate systems. The literature
finds that realignments are somewhat predictable from readily available
SEPTEMBER/OCTOBER 1994
2
information such as interest rates and position of the exchange rate within
the band. The relationship between realignment expectations and macro
variables—such as output and prices—is weak and uncertain, however.
Neely concludes that further work on the formation of expectations would
make an important contribution to future research. Additionally, he finds
that the role of the U.S. dollar in ERM realignments is often noted but has
not yet been incorporated into the estimation techniques.
35
A Case Study in Monetary Control: 1980-82
R. Alton Gilbert
During the three years ending in the fall of 1982, the Federal Reserve
implemented the monetary policy decisions of the Federal Open Market
Committee (FOMC) by targeting nonborrowed reserves. Policymakers
described this change in the operating procedure as an attempt to improve
monetary control. This three-year experience with nonborrowed reserves
targeting has generated a great deal of analysis by economists.
R. Alton Gilbert investigates whether the record of policy actions during
this period reflected a consistent attempt to hit short-run objectives for
money growth, given the confidential information available then to
policymakers: staff projections of total reserves over periods between FOMC
meetings, and staff estimates of the level of total reserves that would be
consistent with the objectives of the FOMC for money growth.
FEDERAL RESERVE BANK OF ST. LOUIS
3
Joseph A. Ritter
Joseph A. Ritter is an economist at the Federal Reserve Bank
of St. Louis. Heidi L. Beyer provided research assistance.
Job Creation and Destruction:
The Dominance of Manufacturing
E s t im a t e s o f g r o s s )o b c r e a t i o n and
destruction (gross flows) give a deeper perspec
tive on the ebb and flow of labor markets in a
market economy than do the headline-grabbing
announcements of net employment growth.
Gross flow data give insight into the uniformity
of employment growth across different parts of
the economy. The path of total employment
may be the total of many industries with similar
growth experiences or of many industries with
extremely diverse experiences; overall employ
ment growth may be the result of lots of job
creation canceling lots of job destruction or
only a little of each.
In addition, the mix between job creation
and destruction can and does vary dramatically
over the business and seasonal cycles in the
economy. Considerable attention has been
devoted recently to the behavior of gross flows
in the labor market (Blanchard and Diamond,
1990; Davis and Haltiwanger, 1990, 1992; Ritter,
1993), and stylized facts from these descriptive
analyses have begun to generate theoretical
research (Mortensen and Pissarides, 1993).
Little attention, however, has been devoted to
the question of whether these facts characterize
all parts of the economy or only particular seg
ments. This paper addresses that question using
the method for measuring gross flows developed
in Ritter (1993). It examines gross job creation
and job destruction in three broad sectors: goods
production, trade, and service production
excluding trade.
The main conclusion is that job creation and
destruction behave much differently in the goodsproducing sector than in the rest of the economy.
Manufacturing and other goods-producing
industries, which make up only a quarter of
private nonfarm payrolls, contribute dispropor
tionately to changes in overall job creation and
destruction, particularly during recessions.
Given systematic differences between goodsand service-producing sectors, it is misleading
to draw sweeping conclusions (that is, “stylized
facts”) about the economy from aggregate gross
flows (Blanchard and Diamond, 1990; Ritter,
1993) or from manufacturing gross flows (Davis
and Haltiwanger, 1990, 1992). Anderson and
Meyer (1994), studying labor turnover, also
concluded that manufacturing was “atypical
in a large number of dimensions.”
In addition, the dynamics of job creation
and destruction in manufacturing appear to
have changed during the most recent recession.
Combined with the declining share of goods
production in overall employment, this suggests
that the dynamics of job creation and destruction
for the economy as a whole may be substantially
different in the future.
SEPTEMBER/OCTOBER 1994
4
CONSTRUCTING GROSS FLOW DATA
The raw data used to construct gross job cre
ation and destruction are monthly employment
levels in several hundred industries in the private
nonfarm sector of the economy. The payroll
or establishment survey, on which the employ
ment data are based, currently covers more than
370,000 establishments, including all firms with
more than 250 employees and a subset of smaller
firms. These data are benchmarked annually using
yet more comprehensive information. The survey
excludes agricultural workers, unpaid family
workers, domestic workers in private homes,
and self-employed persons. To focus on job
creation and destruction driven primarily by
market forces, the data used for this paper also
exclude government workers, though the survey
includes them.1
employment changes in industries in which
employment is increasing:
/C, = £<S< +)A£
i=1
where <5j(+) is 1 if employment is increasing in
industry i and 0 otherwise; Eit is employment in
industry i; and N is the number of industries in
the sector under consideration. Job destruction
is defined as the sum of absolute values of
employment changes in industries in which
employment is decreasing:
JDt = f t ( l - 8 \ ; ) )\AEi t \ = J Ct - f , * E i t .
1= 1
2= 1
Job creation and destruction rates used below
divide creation and destruction levels by total
employment in the sector’s N industries:
The details of constructing job creation and
destruction series (and caveats about them) are
described in Ritter (1993), but the main idea
is as follows. First, the breadth of coverage is
defined by the set of industries for which con
tinuous employment data are available since
1972. The 1972 start date was chosen because,
for a large fraction of industries outside manu
facturing, disaggregated employment data are
not available for earlier years. Thus, the data
cover a comprehensive cross-section of the non
farm business sector. In January 1972, employ
ment was 58.1 million for all private nonfarm
payrolls, with 97.6 percent in the industries used
in the job creation and destruction calculations.
By March 1994, total employment was 93.4 million
for all private nonfarm payrolls with 95.3 percent
included in the present calculations. Second, a
set of nonoverlapping industries is created using
the finest level of detail available. These are threeand four-digit industries as well as the parts of
two- and three-digit industries that are not more
finely classified into three- and four-digit indus
tries. The exact set of industries varies over time
as the Bureau of Labor Statistics (BLS) refines
the industrial classification scheme.
In several different years, the standard indus
trial classification (SIC) used by BLS to allocate
employment among industries is revised. In
general, the revision results in a finer breakdown
of industries already included, but sometimes
it adds coverage of entirely new industries. As
previously mentioned, the job creation and
destruction series are constructed so that the
breadth of industrial coverage does not change
from the first period to the last. A finer breakdown
within a larger industry is exploited, however,
by using an adjustment at the “birth” of a new
(three- or four-digit) industry that accounts for
the fact that the start of data on the industry does
not indicate job creation, but reclassification.
Since new three- and four-digit industries are
generally created to subdivide growing industries,
this procedure tends to limit the extent to which
job creation and destruction net out within
industries.2
Third, for a month t when there is no change
in the industrial classification (most months),
gross job creation is defined as the sum of
This paper presents data on three sectors: (1)
goods production, which includes manufactur
ing, construction and mining; (2) wholesale and
1 Including government workers in subsequent calculations
does not significantly change aggregate patterns of job cre
ation and destruction.
2 The exact procedure followed in months when a finer break
down of an industry appears in the data is described in the
appendix to Ritter (1993).
FEDERAL RESERVE BANK OF ST. LOUIS
JCR,
JDR,
JC t
JDt
5
Figure 1
Job Creation and Destruction Rates for All Private
Nonfarm Industries
5-month, centered moving average, seasonally adjusted
retail trade; and (3) service production except
trade. The third category includes services, trans
portation, utilities, communications, finance,
insurance and real estate. Trade is usually counted
as a service-producing industry, but is initially
treated here as a separate category because its
close tie to goods production (through purveyance
of goods) could make its gross flow dynamics
more similar to manufacturing than to services.
One problem with using industry data to mea
sure gross flows is that the unit of measurement
(an industry) is quite large. Substantial netting
of job creation and destruction could take place
within each industry. This point is discussed
extensively in Ritter (1993), but the problem is
magnified by the present attempt to disaggregate
the gross flows. Although 573 industries are
used in constructing gross flow measures for the
private nonfarm economy, 338 are in goods pro
duction, but only 97 are in trade and 138 are in
other service production. As a result, the average
sizes of industries in 1993 were 69,239 workers
in goods production, 264,679 in trade and
278,034 in service production.
GROSS FLOWS BY SECTOR
Job creation and destruction rates for the entire
nonfarm sector are shown in Figure 1. The figure
illustrates two features of gross flow data which
have been noted in previous work: (1) There is
always a great deal of both creation and destruc
tion; at their lowest points the five-month moving
averages of monthly creation and destruction rates
were still 0.5 percent and 0.4 percent of private
nonfarm employment per month. Because of
intraindustry netting, these figures understate the
extent of ongoing job creation and destruction.3
(2) Net employment change during recessions is
dominated by rises in job destruction, rather than
falls in job creation. As noted in Ritter (1993),
these features are shared by gross flow data pro
duced from the Current Population Survey,
3 Ritter (1993) compared job creation and destruction rates in
manufacturing constructed from establishment-level data
with those constructed from industry employment data. The
former were more than three times higher on average.
SEPTEMBER/OCTOBER 1994
6
Figure 2
Job Creation and Destruction Rates in Goods Production
5-month, centered moving average, seasonally adjusted
0.025
0.020
0.015
0.010
0.005
0
1972
74
76
78
80
82
84
86
88
90
92
1994
92
1994
Figure 3
Job Creation and Destruction Rates in Trade
5-month, centered moving average, seasonally adjusted
0.014 0.012
-
0.010
-
0.008 0.006 0.004 0.002
-
0 1972
74
76
FEDERAL RESERVE BANK OF ST. LOUIS
78
80
82
84
86
88
90
7
Figure 4
Job Creation and Destruction Rates in Service Production*
5-month, centered moving average, seasonally adjusted
* Excluding trade
** Spikes in creation and destruction during 1983 are caused by a large strike in
the telephone communications industry. See footnote 4.
which tracks individuals, and by gross flow data
produced by Davis and Haltiwanger (1990, 1992)
from the Census of Manufactures, which tracks
employment at single establishments.
Figures 2, 3 and 4 show job creation and
destruction rates for the goods-producing, trade
and service-producing sectors.4 Three points
about these charts stand out. First, the gap
between creation and destruction for the trade
and service-producing sectors during the 1980s
indicates the well-known fact that these sectors
produced substantial net employment gains during
the decade. In fact, in the service-producing
sector, job creation exceeded job destruction
during all but a few months since 1972. Trade
experienced more frequent employment declines,
but even during recessions these drops were not
4 The large spikes in destruction and creation during 1983 in
Figure 4 reflect the beginning and end, respectively, of a
large strike in the telephone communications industry (SIC
4813). A comparison of BLS data on new work stoppages
(which starts in 1981) and the job destruction series shown
in Figure 1 reveals that a few small spikes in job destruction
particularly large or prolonged. By contrast, fol
lowing the recovery from the 1982 recession, job
creation and destruction were closely balanced
in the goods-producing sector until the onset of
the 1990 recession.
Second, goods production shows a sharp
asymmetry between creation and destruction
during recessions; destruction is considerably
more volatile. Neither trade nor service produc
tion shows evidence of this asymmetry, however.
Finally, despite trade’s close link with goods
production, gross flows in the trade sector do
not exhibit patterns that closely resemble those
in goods production.
Job creation and destruction rates for different
sectors are compared directly in Figure 5, which
during the 1980s correspond to relatively large strikes, but
the telephone communications strike is the only one that has
a noticeable impact on the series.
SEPTEMBER/OCTOBER 1994
8
isolates a striking fact: Both creation and destruc
tion rates are far more volatile in goods-producing
industries than in trade or other service-producing
industries. Goods production thus contributes
disproportionately to fluctuation in aggregate
gross flows, particularly job destruction.
ment data are compared to those created from
establishment data by Davis and Haltiwanger, the
size of fluctuations is again very similar, though
the levels of the series differ dramatically (see
Ritter, 1993).
Figure 5 does not tell the whole story about the
relative importance of gross flows in goods pro
duction because this sector made up 25 percent
of private nonfarm employment in 1993 (down
from 39 percent in 1972). Figure 6 displays the
contributions of goods-producing and serviceproducing (now including trade) industries to
total job creation and destruction levels.
THE CHANGING ROLE OF
MANUFACTURING
Figure 6 appears to show that goods production
contributes a disproportionate share of overall job
creation and destruction levels. This is probably
misleading, however. The manufacturing sector
is more finely divided, so there is probably less
intraindustry netting of job creation and destruc
tion in the goods-producing sector than in the
service-producing sector. This would impart a
substantial upward bias to the relative contribu
tion of goods production to the level of overall
job creation and destruction.
The relative contributions of goods- and service-producing industries to cyclical changes
in overall job creation and destruction are shown
more reliably in Figure 6. Goods production has
typically accounted for more of the cyclical move
ments than the industries that make up the other
75 percent of employment. This is particularly
evident in the lower panel of Figure 6, which
shows much more dramatic cyclical swings
in total job destruction than in service produc
tion alone.
Two pieces of evidence suggest that intrain
dustry netting does not substantially bias the
contribution of goods-producing industries to
changes in job creation and destruction. First, if
four-digit industries are ignored in constructing
the job creation and destruction series (thus
increasing the average size of industries used in
the calculation and the extent of intraindustry
netting), both series shift down, but the ampli
tude of fluctuations is not significantly changed.5
Second, in manufacturing, if job creation and
destruction series created from industry employ
5 Regressing job creation constructed without four-digit indus
tries on job creation constructed with four-digit industries (or
vice versa) produces a coefficient very close to 1.0 and an
R2 greater than 0.99. The same is true of the job destruction
series.
FEDERAL RESERVE BANK OF ST. LOUIS
Figure 2 reveals that gross flows in the goodsproducing sector were less volatile during the
1990 recession than during previous recessions.
This warrants closer attention to manufacturing,
which makes up more than three-quarters of
goods-producing employment. Figure 7 shows
that the phenomenon is even more pronounced
in manufacturing. When the gross flow data for
manufacturing are extended back to 1947 (which,
unfortunately, cannot be done reliably for non
manufacturing industries), all previous recessions
show much more dramatic swings in job creation
and destruction than 1990. If manufacturing is
split into durables and nondurables, both show
patterns very similar to Figure 7. Gross flows
for mining and construction (the remainder of
the goods-producing sector) did not seem to
follow the same pattern as manufacturing during
the 1990 recession. The very low levels of job
creation and destruction during the 1990 reces
sion are, therefore, clearly due to developments
in the manufacturing sector.
As measured by drops in either industrial
production or manufacturing employment, the
1990 recession was mild. Manufacturing employ
ment, however, declined almost continuously
from the beginning of 1989 until late 1993. It
appears that, rather than the usual sharp cyclical
response, manufacturing firms have experienced
a longer-term contraction over these five years.
Though it is clear that something different hap
pened during the 1990 recession, it is impossible
to know whether the old pattern of sharp increases
in job destruction will reassert itself in future
downturns. If the fluctuations of gross flows in
manufacturing remain subdued during future
recessions, the movement of overall gross flows
will be significantly damped. The declining
share of employment found in manufacturing
reinforces this effect by lowering the weight
attached to the most volatile sector.
9
Figure 5a
Job Creation Rates in Goods Production,Trade and
Service Production*
5-month, centered moving average, seasonally adjusted
* Excluding trade
Figure 5b
Job Destruction Rates in Goods Production,Trade and
Service Production*
5-month, centered moving average, seasonally adjusted
* Excluding trade
SEPTEMBER/OCTOBER 1994
10
Figure 6a
Job Creation in Goods Production and Service Production
1972
74
76
78
80
82
84
86
88
90
92
1994
Figure 6b
Job Destruction in Goods Production and Service Production
Thousands
1972
5-month, centered moving average, seasonally adjusted
74
76
RESERVE BANK OF ST. LOUIS
FEDERAL
78
80
82
84
86
88
90
92
1994
11
Figure 7
Job Creation and Destruction Rates in Manufacturing
5-month, centered moving average, seasonally adjusted
0.025
0.020
0.015 -
0.010
-
0.005 -
1972
74
CONCLUSIONS
Job creation and destruction behave much
differently in the goods-producing sector than in
the rest of the economy. Job creation and destruc
tion have historically been much more volatile
in manufacturing and other goods-producing
industries, so that they have contributed dispro
portionately to fluctuations in overall job creation
and destruction. Further, there does not appear
to be a cyclical asymmetry between creation and
destruction outside of manufacturing. The stylized
fact, cited by several authors (Blanchard and
Diamond, 1990; Davis and Haltiwanger, 1990,
1992; Ritter, 1993), that job destruction tends to
dominate employment changes during recessions
thus appears to be generated by manufacturing
industries. In addition, job creation and destruc
tion in manufacturing were noticeably damped
during the most recent recession. Combined
with the fact that goods production makes up a
declining share of employment, this suggests
1994
that the dynamics of job creation and destruction
may be substantially different in the future.
REFERENCES
Anderson, Patricia M., and Bruce D. Meyer. “The Extent and
Consequences of Job Turnover,” Brookings Papers on
Economic Activity (Microeconomics 1994).
Blanchard, Olivier Jean, and Peter Diamond. ‘The Cyclical
Behavior of the Gross Flows of U.S. Workers,” Brookings
Papers on Economic Activity (1990, No. 2), pp. 85-155.
Davis, Steven J., and John Haltiwanger. “Gross Job Creation,
Gross Job Destruction, and Employment Reallocation,”
Quarterly Journal of Economics (August 1992), pp. 819-63.
_____ , a n d _____ . “Gross Job Creation and Destruction:
Microeconomic Evidence and Macroeconomic Implications,”
NBER Macroeconomics Annual (1990), pp. 128-86.
Mortensen, Dale, and Christopher Pissarides. “Job Creation
and Job Destruction in the Theory of Unemployment,” The
Review o f Economic Studies { M y 1994), pp. 397-416.
Ritter, Joseph A. “Measuring Labor Market Dynamics:
Gross Flows of Workers and Jobs,” this Review
(November/December 1993), pp. 39-57.
SEPTEMBER/OCTOBER 1994
13
P eter Yoo
Peter Yoo is an economist at the Federal Reserve Bank of
St. Louis. Richard D. Taylor provided research assistance.
Boom or Bust? The Economic
Effects of the Baby Boom
B
1ETWEEN 1947 AND 1962, the population of
the United States grew at an average annual rate
near 2 percent, a large increase from the average
annual growth rate near 1 percent during the 20
years prior to World War II. Moreover, since
1962, the average population growth rate has
fallen to its pre-war level. This large but tempo
rary increase in the population growth rate,
more familiarly called the baby boom, raises an
interesting and important question: How do
such large changes in the population growth rate
affect a developed economy? Undoubtedly, the
baby boom has already had a large effect on the
U.S. economy, especially on the composition
of goods and services produced by the market
place and the government. But the economic
effects of the baby boom are more basic than the
optimal mix of convertibles and minivans, or
the number of school buildings vis-a-vis nursing
homes, because such large changes in the popu
lation growth rate affect aggregate consumption
and saving. Specifically, a large influx of workers
requires more capital to maintain the same level
of labor productivity, which in turn affects indi
vidual living standards.
Questions about growth of per capita income
and consumption per capita are not limited to the
entrance of the baby boomers into the economy
but extend to its aging as well. In a life-cycle
framework, individuals retire and consume their
savings. This implies that if a large fraction of
the population is retired, society will save less,
perhaps even “dissave,” and lower aggregate
saving leads to a slower rate of capital formation.
This possibility has caused a great deal of con
cern about the impending retirement of the baby
boom generation. Lower saving, however, need
not impose a drag on the economy. Just as the
entry of the baby boom increases the demand
for capital, the baby boomers’ retirement decreases
the demand for capital since their retirement
decreases the labor supply. Thus, the mere
retirement of the baby boom generation need
not imply slower growth since the economy
requires less capital. So what is the likely impact
of the baby boom on the rate of capital accumu
lation and, thus, on the growth of income per
capita and consumption per capita?
To answer this question, I turn to three models
of economic growth that incorporate different
aspects of demographic changes. Although the
models cannot possibly capture all aspects of
economic behavior that may affect the answer
to the question posed above, they can provide
insights about the fundamental relationship
between population growth and the growth of
output per capita. The models presented here,
and models of economic growth in general,
depend on accumulation of capital as the engine
of growth of output per worker and standards
SEPTEMBER/OCTOBER 1994
14
of living. At any given time, agents either con
sume or invest their resources, so their savingconsumption decisions are critical determinants
of how fast labor productivity will grow.
All three models presented here predict that a
temporary and unexpected increase of population
growth rate raises aggregate saving, but such an
increase in saving is not necessarily large enough
to maintain pre-boom rates of growth per capita
income and standards of living. Once a baby
boom has completely entered an economy,
capital intensity tends to rise and the economy
gradually returns to its pre-boom status. The
three models disagree about the speed and mag
nitude of such changes, but all show that after a
period of slow growth, per capita consumption
increases. Best of all, the models indicate such
improvements in the standard of living occur as
even aggregate saving drops. This suggests that
in isolation, the retirement of the baby boom
need not imply diminishing standards of living.
The paper proceeds as follows. The first
section presents a brief description of the baby
boom’s effect on the U.S. population. Next, I
present three growth models and their predictions
about the response of the economy to the baby
boom. The models focus on the relationship
between the population growth rate and capital
accumulation since all other economic factors
depend on saving and the resultant path of capital.
The third section examines the recent performance
of the U.S. economy to check the consistency
of the models’ qualitative predictions with
observed economic data. The final section
draws some conclusions about the baby boom
and the economy.
THE BABY BOOM
Figure 1, top panel, shows Bureau of Census
estimates of the annual growth rate of U.S. resi
dent population since 1930 and its middle pro
jections of the annual growth rate from 1994 to
2050.1 The figure underscores the demographic
importance of the baby boom. The baby boom
was well under way by 1947 and lasted some 15
years. During the baby boom, the population
growth rate was nearly double the 1 percent
average annual growth rate during the 20 years
prior to 1947. Once the baby boom ended, the
1 The Census Bureau regularly publishes three projections—
lowest, middle and highest. They represent different
assumptions about fertility, net immigration and life
expectancy. See Current Population Reports, P25-1104,
pp. xxxv-xxxix.
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
population growth rate returned to an average
annual rate near 1 percent. The top panel also
shows the annual growth rate of the working-age
population, all individuals ages 18 to 65, again
based on Bureau of Census’ estimates and pro
jections. The size of the working-age population
reflects the impact of the baby boom with a lag
of 18 years.
The top panel does not, however, adequately
reflect one of the key economic issues associated
with the passage of a baby boom: What happens
when the baby boom retires? From a life-cycle
viewpoint, the baby boomers’ retirement will
dramatically increase the number of dissavers
vis-a-vis savers, as well as the number of con
sumers relative to workers. One way to measure
the relative sizes of the two segments of the pop
ulation is the dependency ratio, which I define
as the ratio of the number of consumers to the
size of the potential labor force. The bottom
panel of Figure 1 shows the dependency ratio
for the United States between 1930 and 2050
based on the estimates and projections from the
Bureau of Census. The ratio rises at the start of
the baby boom since children only consume,
falls as they pass into adulthood, and finally,
rises again as they retire.
THREE MODELS
In this section, I present three exogenous
growth models to analyze the effects of a baby
boom on the U.S. economy:2 the neoclassical
model of Ramsey (1928); the dependency-ratio
model of Cutler, Poterba, Sheiner and Summers
(1992); and the o v erlap p in g g en era tio n s (OLG)
model of Yoo (1994). Each model provides a
framework to examine the relationship between
changes in the population growth rate and the
capital-labor ratio, which in turn determines
per capita income and consumption per capita.
I present each model with its simulation results
and then highlight the differences and similari
ties among the three models.
Neoclassical Growth M odel
The simplest model that relates population
growth rate to economic growth is the neoclassical
growth model of Ramsey. The model has a
benevolent social planner who, with perfect
2 Also see Auerbach, Kotlikoff, Hagemann and Nicoletti
(1989) and Auerbach, Cai and Kotlikoff (1991).
15
Figure 1
U.S. Population Characteristics
Total and Working-Age Population Growth Rates
Dependency Ratios
SEPTEMBER/OCTOBER 1994
16
foresight, maximizes the discounted utility
function of a representative agent subject to the
economy’s resource constraints.3 The solution
to the social planner’s problem is equivalent,
under appropriate assumptions, to the competi
tive equilibrium in which individuals and firms
maximize their utility and profits. The model
also assumes each individual inelastically pro
vides one unit of labor.
This assumption also applies to all three models in
this paper. In the steady state, the equilibrium
capital-labor ratio yields the modified golden rule,
which states that the marginal product of capital
in steady state equals the sum of the subjective
discount rate and the population growth rate.
Formally, the central planner maximizes the
utility function of a representative agent:
(5) f ' ( k ' ) = 8 + n\
(1) max [/(cf) =
J u(ct )e Stdt,
subject to the budget constraint that output in
each period equals consumption, net investment
and capital for new entrants:4
(2) yt = ct + k t+ nt k t,
where U[ct) is the instantaneous utility of a
representative agent, 8 is the subjective discount
rate with 0 < 8 < 1, ct and yt are consumption and
output per unit of labor, k t is the capital-labor
ratio, and nt is the population growth rate. I
assume that the net production function of the
economy is Cobb-Douglas to simplify the simu
lation:
(3) y = f [ k , ) = k ° ,
where a is the output elasticity of output of cap
ital, and 0 < a < 1.
The solution for the maximization problem is
(4) £l = I[/'(*
c,
P
where
is the coefficient of relative risk aversion. I
assume that instantaneous utility is isoelastic
with a constant coefficient of relative risk aver
sion, so that
3 The assumption of perfect foresight does not extend to the
timing of the beginning or end of the baby boom. Rather, I
assume that both the start and end of the baby boom are
unanticipated shocks to the population growth rate. This
assumption about the timing and the duration of the baby
boom applies to all three models. This assumption affects
the dynamics of the economy’s response to the baby boom.
If the timing of the baby boom were anticipated, the economy
FEDERAL RESERVE BANK OF ST. LOUIS
where stars denote steady-state values of each
variable. The corresponding optimum per capita
consumption equals
(6) c' = f [ k ' ) - r i k\
To determine the dynamics of the economy
near the steady state, I linearize equations 2 and
4 using a Taylor’s series expansion. Solving the
resulting system of second-order differential
equations and ruling out the divergent path, the
following equations describe the path of the
economy near the steady state:5
(7) k t= k ' + [k 0 - k ' ) e M,
where
A = < S - i x <S2 +4/3
2 '
p=
n * v
p
and k 0 is the initial capital-labor ratio.
To simulate the economic effects of the baby
boom, I assume that the U.S. economy starts at
the steady state for slow population growth and
introduces the baby boom. The economy then
moves toward the new steady state associated
with the faster population growth rate. Once the
population growth returns to its pre-boom rate,
the economy reverses direction and moves to
the pre-boom steady state. Table 1 shows the
would react earlier to the beginning and the end of the
baby boom.
4 Multiplying the budget constraint by the size of the labor
force gives the accounting identity Y = C + / + G, with G
equal to zero.
5 See Blanchard and Fischer (1989, chapter two) for more
details.
17
Table 1
Simulation Parameter Values
Parameters
Description
Value
T
Lifespan (OLG model only)
r
Working life (OLG model only)
n
Initial population growth rate
0.01
e
Size of baby boom
0.01
T
Duration of baby boom
8
Subjective discount rate
P
Coefficient of relative risk aversion
a
Capital’s share of output
60
45
15
0.01
2
0.33
parameters required to simulate this and the
two other models.6 Rather than using the actual
population growth rates, which would unneces
sarily complicate the simulations, the simulations
use a stylized baby boom. As the top panel of
Figure 1 indicates, the baby boom lasted approx
imately 15 years with an average growth rate of
nearly 2 percent per annum, whereas the growth
rate before and after the baby boom averaged
nearly 1 percent per annum. I therefore assume
that the pre- and post-baby boom population
growth rate is 1 percent, the population growth
rate during the baby boom is 2 percent, and the
baby boom lasts for 15 years. Since the Ramsey
model assumes all individuals in the economy
provide one unit of labor inelastically, I also
ignore childhood, pushing the start of the baby
boom by 18 years to 1965.
Figure 2a shows three variables—the capitallabor ratio, the saving rate and per capita con
sumption—normalized by their respective paths
in an economy without the baby boom. The first
figure shows that an increase in the labor force
depresses capital intensity; the higher population
growth rate depresses the modified golden rule
capital-labor ratio, which causes capital intensity
to drop for 15 years until the entry of the baby
boomers stops and the capital-labor ratio is some
10 percent below the pre-baby boom level. There
after, capital intensity converges to the pre-boom
level but does so very slowly. Figure 2a also
shows saving measured as fraction of output,
again normalized by the no-baby boom economy.
The Ramsey model shows a concentrated spike
6 See Auerbach and Kotlikoff (1987, chapter four) for a dis
cussion about the selection of the preference and production
parameters.
in saving, almost 20 percent higher than the
no-baby boom saving rate. Once the population
growth rate returns to pre-baby boom level, saving
falls and eventually returns to its previous level.
The last graph in 2a shows the path of consump
tion per capita normalized by the path of con
sumption in the economy without a baby boom,
and it shows an initial drop in per capita con
sumption of 10 percent, but once the population
growth rate returns to 1 percent, per capita con
sumption gradually returns to its original level.
The Dependency-Ratio Growth Model
One obvious problem with the Ramsey model
is its inability to address the problem of the baby
boomers’ retirement because the model assumes
that agents are homogeneous and that they are
infinitely lived. Once an individual enters an
economy, he or she is no different than any other
individual at that time, and then has an infinitely
long life. A recent paper by Cutler, Poterba,
Sheiner and Summers introduces agent hetero
geneity by incorporating a dependency ratio into
the Ramsey model. This captures the effects of
the retirement of the baby boom on the economy,
albeit in a rather ad hoc manner. Cutler and
others solve the model from a social planner’s
point of view with all individuals alive in each
period weighted equally in the social welfare
function. Unlike the Ramsey model, the com
mand and decentralized solutions are not equal.
The dependency-ratio model, therefore, gives a
path for the economy that does not correspond
to a market equilibrium.7
The command optimization problem is
(8) max t/= | u[ct )Nt e Std t ,
subject to resource constraint similar to equation 2,
(9) yt = Y tc t + k t + ntk t ,
where ct is per capita consumption, yt, k t, 8 and
nt are as previously defined, Nt is the population
size, and yt is the dependency ratio at time t and
equals
CON.
Yt ~ LFt
where LF, is the labor force and CONt is the
number of consumers.
7 The simulations presented in Cutler and others differ from
the one presented here because they incorporate an agedependent labor productivity profile into their simulations.
SEPTEMBER/OCTOBER 1994
18
Figure 2a
Simulation Results: Ramsey Model
Capital-labor ratio
Saving rate
Per capita consumption
Figure 2b
Simulation Results: Dependency Ratio Model
Capital-labor ratio
Saving rate
1.02
2.21
1960 70
10 20
40 2050
Per capita consumption
1960 70 80 90 00 10 20 30
40 2050
Figure 2c
Simulation Results: Overlapping Generations Model
Capital-labor ratio
Saving rate
The solution from the first-order conditions
of the planner’s problem is
(10) C^ = - [ f ' [ k t )-6 }.
c, p
In the steady state, equation 11 implicitly
defines the optimum capital-labor ratio:
(11) f'[ k ') = d .
The model has the interesting property that
the steady-state capital-labor ratio is independent
of all parameters except the subjective discount
rate and the parameters of the production func
tion. Thus, unlike the Ramsey model (or the
overlapping-generations model), the capital-labor
FEDERAL RESERVE BANK OF ST. LOUIS
Per capita consumption
ratio does not adjust to changes in the population
growth rate. Rather, consumption must respond
to any unexpected changes in the population
growth rate or to the dependency ratio, and
furthermore, the response to such changes
is instantaneous.
Although the dependency-ratio model of
Cutler and others incorporate some agent hetero
geneity into the problem, they do not consider
the saving decisions of individuals, especially
saving for retirement and, furthermore, Cutler
and others solve the model from a social plan
ner’s viewpoint. These two facts produce a sim
ple solution, but the solution requires substan
tial redistributions as the baby boom enters and
exits the economy. Cutler and others use the
19
existence of the Social Security system to justify
their modeling choice and the resultant redistri
bution. But the redistributions required by the
social optimum are not the redistribution scheme
embodied by Social Security. In the model, a
large unexpected increase in the population
growth rate requires a large cut in consumption
to finance a large increase in investment to
maintain the constancy of the capital-labor ratio.
Moreover, the end of the baby boomers’ entry
into the economy diminishes the rate of capital
formation, causing a sharp increase in consump
tion. The transfers involved are opposite those
provided by Social Security; the dependencyratio model’s solution transfers resources to the
new entrants, whereas Social Security transfers
wealth from the young to the elderly.
Figure 2b shows the results of the simulation
from the dependency-ratio model. As before,
I have normalized the results by the no-baby
boom economy. As shown by the first graph
and equation 11, the baby boom has no effect
on the capital-labor ratio. The second graph
in 2b shows saving as fraction of output, again
normalized by the no-baby boom economy. Any
changes in the growth of the labor force must
be offset by changes in saving because the model
requires a constant capital-labor ratio. Therefore,
a doubling of the population growth rate requires
a doubling of the saving rate to provide enough
capital for the faster rate of population growth.
Once the population growth rate reverts to the
initial rate, saving returns to the baseline. Since
output is either saved or consumed, per capita
consumption reflects the path of saving. Figure
2b also shows the path of consumption per capi
ta normalized by the path of consumption in
the economy without a baby boom. Since the
dependency-ratio model shows doubling of saving,
consumption falls by 50 percent and indeed the
third graph of 2b reflects such a drop. Once
the boom is over, the increase in the number of
workers supporting retirees implies less has to
be saved and more can be consumed, although
this does not last forever.
and an explicit retirement period maximizes
his or her utility subject to a lifetime budget
constraint. I then aggregate each individual’s
decisions with the decisions of an optimizing
firm to obtain a general equilibrium solution
for the path of an economy confronted with an
unanticipated baby boom. Unlike the other two
models, the model uses discrete time periods,
although this quantization is materially insignif
icant.
The individual born in period t faces the
problem
(12) m a x ^ ( l + 5 )1_su(ct+s_l s ),
S=1
subject to the lifetime budget constraint that his
or her discounted expenditures be no greater
than the person’s available lifetime resources:
(13) £
w,
f + S —1 ,S
t ( l + rt
where ct s is the consumption in period f of an
agent s years old, T is the lifetime of an individual,
T' represents the number of periods working
and wt and rt are the real wage and the real
returns to capital in period t.
The explicit solution comes from recursively
solving the associated Euler equations, and it
produces, under the assumption of static expec
tations, the following two equations, which
describe the optimal saving-consumption deci
sions of an individual:8
w,
(1 4 ) ct+s_ls = es
i= s
(1 +
r t
)
7l7 + (1 + r>
(15)
^t + s - l , s
(l + r( )a1+s- 2,s-i +lv(+s-i - < W , . S i f s < T '
(1 + rt )at+s_2 s_j - c t+s_l s
if s > T ' ,
where
1+
rt
I
p
1+8
An Overlapping-Generations
Growth Model
The model used by Yoo confronts some of the
problems of the Ramsey and the dependencyratio models by using the overlapping-generations
framework. An individual with a finite lifetime
and a t s is the asset level of an agent s years old
in period t which he or she holds as physical
capital.
The sum of all individual savings equals the
capital stock, and the number of working-age
8 Static expectations imply that agents assume that future fac
tor prices equal today’s prices.
SEPTEMBER/OCTOBER 1994
20
individuals equals the labor force of the economy:
T
(16) Lf = £ p ,( s )
S=
1
(t7) * t = E a , f* ( s ) .
S= 1
where cp((s), the age distribution, is the number
of individuals age s in period t. I also assume
markets are competitive and firms minimize
costs so that factor prices equal their marginal
product:
(18) r, = f ' ( k t )
(19) wt = f ( k ,) ~ f ' [ k , ) k t .
Given a set of parameters, modeling the
effects of a baby boom requires specifying the
path of <pt(s) to reflect changes in the population
growth rate. Once I have specified the parameters
and <pt(s), calculating the effects of the baby boom
becomes a series of iterations. First, equations 14
and 15 determine individual behavior, then given
their saving-consumption decisions, equations 16
through 19 determine output and factor prices
which become the basis for the next iteration,
which again begins with 14 and 15.
Figure 2c shows the impact of the baby boom,
simulated by the OLG model. An increase in the
labor force depresses capital intensity, and the
model shows declining capital relative to labor
for a long period of time, nearly 30 years, in which
the minimum is approximately 4 percent lower
than the no-baby boom baseline.9 Figure 2 c also
shows saving gradually increasing until all baby
boomers are dead, reaching a peak near 2010
approximately one-third higher than the no-baby
boom economy. The third figure shows the path
of consumption per capita, and it indicates that
consumption falls gradually, 5 percent below
baseline. Consumption then rises for the following
four decades until it reaches its initial level.
Comparing the Simulation Results
Comparing the nine graphs in Figure 2 indi
cates several similarities as well as several
points of divergence. The figures indicate that
the magnitude and the timing of the economic
effects of the baby boom are the major points of
divergence among the three models. Although
9 The relative smoothness of the OLG model is partially attrib
utable to the static expectations assumption.
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
the Ramsey and OLG models both show declining
capital-labor ratios, the drop is much larger in
the Ramsey model, 10 percent versus 4 percent.
Furthermore, the Ramsey model predicts the
trough will occur more than 10 years earlier than
the OLG model, despite the fact that the Ramsey
model requires substantially more time to return
to the pre-baby boom steady state. The paths of
saving also indicate responses of different mag
nitudes and timing, although the signs of the
responses are the same. Both infinite horizon
models show declining saving at the end of the
baby boom, whereas the OLG model continues
to increase until the first of the baby boomers
are near retirement. Peak savings in the Ramsey
and OLG models are similar in magnitude, and
the much higher saving of the dependency-ratio
model is attributable to the constancy of the
marginal product of capital. The behavior of
consumption per capita is very similar to that
of saving, both in timing and magnitude; the two
infinite-horizon models indicate that per capita
consumption in the United States should have
already rebounded from the depressed state
induced by the entry of the baby boomers, with
the dependency-ratio model suggesting a signifi
cantly bigger response to the baby boom. The
OLG model, in contrast, suggests that we should
be now near the trough of the fall in consumption.
The most striking point of agreement among
the three models is the response of the con
sumption and saving relationship to the passage
of the baby boom. All three models predict that
an unexpected baby boom causes a temporary
increase in saving and an associated temporary
drop in per capita consumption. Most impor
tantly, the return to the pre-baby boom saving
rate that occurs in all three models coincides
with an increase in consumption. This counter
intuitive result arises because the demand for
capital diminishes as the population growth rate
slows. Moreover, the overlapping-generations
model shows that even with the baby boomers
dissaving in retirement, consumption per capita
continues to increase. These results suggest that
current concerns about an economic decline fol
lowing the retirement of the baby boomers may
be unfounded.
U.S. EXPERIENCE THUS FAR
Figure 3 shows a series of comparisons
between observed data and simulation results.
21
Figure 3
Observed Data vs. Simulation Results
Panel b: Real returns to capital
Panel a: Real wages
6-
1.08
■1.01 10D e p e n d en c y ratio
-1
4-
/
»-
R am sey
1.06
6-
1.04
-0.99
2
4-
-
-0.98
0-
1960
\
70
1
80
'p — ' ' '
90
00
R am sey
10
20
30
U 3'
40
/
------
1.02
2-
1
0-
-0.96 -21960
2050
u
70
D e p e n d en c y ratio
80
90
00
10
20
30
40
0.98
2050
Panel d: Per capita consumption
Panel c: National saving rate
6
1.2
-
D e p e n d en c y ratio
2
-
-
0.8
0.6
0.8
-2
1960
70
80
90
00
10
20
30
40
0.4
2050
Scale: observed data; left, simulation; right
It is important to note that the actual data is
not normalized; therefore, the magnitudes of
the actual data and the simulation results are
not directly comparable. Panels a and b show
real wages and real returns to capital rather than
capital-labor ratios. Since the two factor prices
are monotonic transforms of the capital-labor
ratio, they should provide a reasonable alternative
to directly comparing observed and simulated
capital-labor ratios. Panel a shows the annual
growth of real wages, as measured by hourly
compensation, compared to the wages from the
three models, which I have also normalized by
the no-baby boom wages. Growth of real wages
has been on a downward trend that is consistent
with the predictions of the Ramsey and OLG
models. Panel b shows the real returns to capital,
measured by long government yields less CPI
inflation, compared to returns to capital from the
three models, also normalized by the no-baby
boom baseline. Although the rise of real long
government bond yield during the 1980s is con
sistent with the OLG model, its relationship to
the simulated returns to capital is ambiguous.
Panels c and d provide direct comparisons
between observed and simulated paths of saving
and consumption. Once again, I have normalized
the simulated results by the baseline economy
with no-baby boom. As shown in panel c, the
observed saving rate, measured by the national
saving rate, has fallen recently, as predicted by
the Ramsey and the dependency-ratio models,
but the drop does not correspond to a reversion
to pre-baby boom rates. The observed behavior
of the real annual growth of consumption per
capita is more consistent with the paths from
the three models’ predictions. The growth of
consumption has gradually slowed since the
start of the baby boom as predicted by all three
models, especially the OLG model, since the
Ramsey and dependency-ratio models indicate
that consumption should have already returned
to near pre-baby boom levels.
SEPTEMBER/OCTOBER 1994
22
PROGNOSIS
REFERENCES
As the models show, demographic factors
can play an important role in macroeconomic
performance, mostly at low frequencies. Given
the simple and stylized simulations reported in
this paper, the correspondence between simula
tion and observed low-frequency movements in
several important macroeconomic variables is
noteworthy. Slow wage growth and diminished
consumption growth are consistent with the pre
dictions of the models, especially the OLG model.
The evidence from saving rates and the real
returns to capital is less clear.
Auerbach, Alan J., Jinyong Cai, and Laurence J. Kotlikoff.
“U.S. Demographics and Saving: Predictions of Three
Saving Models,” Carnegie-Rochester Conference Series on
Public Policy, vol. 34 (1991), pp. 73-101.
What does the baby boom imply for future
growth and welfare? The models suggest a
faster rate of consumption growth, along with
declining real returns to capital and higher wages
that accompany higher labor productivity.
Moreover, these benefits occur throughout the
remainder of the baby boom generation’s lifetime,
including retirement. Thus, even as they dissave,
according to the OLG model, consumption per
capita will continue to increase.
RESERVE BANK OF ST. LOUIS
FEDERAL
_____ , and Laurence J. Kotlikoff. Dynamic Fiscal Policy
(Cambridge University Press, 1987).
_____ , Laurence J. Kotlikoff, Robert P. Hagemann, and
Giuseppe Nicoletti. ‘The Economic Dynamics of an Aging
Population: The Case of Four OECD Countries,” OECD
Working Paper No. 62 (1989).
Blanchard, Olivier Jean, and Stanley Fischer. Lectures on
Macroeconomics. MIT Press, 1989.
Cutler, David M., James Poterba, Louise M. Sheiner, and
Lawrence H. Summers. “An Aging Society: Opportunity or
Challenge?" Brookings Paper on Economic Activity, vol. 1
(1990), pp. 1-73.
Ramsey, Frank P. “A Mathematical Theory of Saving,”
Economic Journal (December 1928), pp. 543-59.
Yoo, Peter S. ‘The Baby Boom and Economic Growth,”
Federal Reserve Bank of St. Louis Working Paper No. 94001A (February 1994).
23
Christopher J. N eely
Christopher J. Neely is an economist at the Federal Reserve
Bank o f St. Louis. Kelly M. Morris provided research assistance.
Realignments of Target Zone
Exchange Rate Systems:
What Do We Know?
Chief Witch: Yes, that’s right.
MacBeth: I understand you can foretell the future.
— From a BBC Radio Program, June 1968
During the French revolution such people were known as agioteurs (speculators)
— and they were guillotined.
— Michel Sapin, French Minister of Finance, speaking of currency traders1
C_JlNCE MARCH 1979, most of the nations of the
European Union have participated in a “target
zone” system of exchange rate management known
as the Exchange Rate Mechanism (ERM) of the
European Monetary System (EMS). Although the
target zones of the ERM have weathered many
adjustments since their inception, speculative
currency attacks in September 1992 and August
1993 led to the de facto suspension of the system.
The United Kingdom and Italy suspended their
participation in the ERM on September 17, 1992.
After August 1993, the bands were broadened
sufficiently to functionally alter the character of
the system. These recent crises have focused
attention on the stability of not only the ERM,
but of target zone systems generally.
' Macleod (1992).
2 See Corbae, Ingram and Mondino (1990) for a theoretical
development of one justification for target zones.
A target zone is a hybrid exchange rate regime,
a compromise between floating and completely
fixed exchange rates. In a target zone system,
monetary authorities pledge to keep the exchange
rate with a particular foreign currency, or basket
of currencies, within given margins around a
central parity. At times, the authorities may also
choose to realign the central parity. Advocates
argue that target zones blend the advantages of
fixed exchange rates and flexible exchange rate
systems.2 Krugman and Miller (1992) point out
that the original justification for constraining
EMS exchange rates within target zones was to
reduce exchange rate volatility, which contributes
to uncertainty and risk in international trade and
investment.3 More recently, a desire to “borrow”
3 Engel and Hakkio (1993) and Neely (1993) study the volatility
of exchange rates under target zones from different per
spectives.
SEPTEMBER/OCTOBER 1994
24
the low inflation reputation of a foreign central
bank (for example, the Bundesbank) has been
frequently cited as an advantage of target zones.
Compared to completely fixed rates, target zones
allow central banks greater scope for monetary
independence.4 Paradoxically, the exercise of
independence may contribute to expectations
of realignment, which produce a “speculative
attack,” in which speculators refuse to hold one
of the currencies at any exchange rate in the target
zone. A successful speculative attack necessitates
a realignment of the central parity, thus thwarting
the goal of stability of the exchange rate.5
Researchers would like to understand the
circumstances associated with speculative attacks
and the realignments of central parities within a
target zone for several reasons. If financial market
participants could forecast realignments, they
could profit from the large changes in asset prices.
For example, it is estimated that investor George
Soros made $1 billion speculating against the
pound and the lira as a result of the crisis of 1992.
Monetary authorities have a different rationale
for analyzing realignments: They wish to be able
to manage the economy more effectively. Ideally,
they would like to maintain stable exchange rates
and low inflation while also retaining sufficient
monetary flexibility to conduct countercyclical
stabilization policy. Although there is no con
sensus on the microeconomic benefits of exchange
rate stability versus the macroeconomic benefits
of domestic stabilization policy, realignments
produce uncertainty about the value of interna
tionally held assets/investments which policy
makers would like to avoid.
Economists have had little success in fore
casting exchange rates at short horizons. Yet, there
is evidence (Mizrach, 1993c) that we can forecast
target zone realignments over a short interval
using information from interest rates, inflation,
and the position of the exchange rate in the target
zone. This article surveys the recent research
on forecasting realignments and estimating the
credibility of target zones. To facilitate under
standing of the functioning of exchange rate target
zones, the next section of this article presents a
simple monetary model of exchange rate deter
4 In this context, independence means freedom to use monetary policy for internal, rather than external, goals. The limits
of this type of monetary independence in a target zone are
explored by Kool (1993). Pollard (1993) examines the bene
fits of freeing central banks from political pressures.
5 The theoretical literature on speculative attacks on fixed
exchange rate systems is well-developed. Salant and
FEDERAL RESERVE BANK OF ST. LOUIS
mination. Section three discusses the functioning
of target zone systems. The empirical literature
on realignments and credibility of target zones
is surveyed in section four. The final section
summarizes the conclusions of the literature
and suggests future research.
EXCHANGE RATE DETERMINATION
Target zones are created to stabilize exchange
rates. It is necessary to understand exchange
rates and the market forces that determine them
to understand the forces behind realignments of
target zones. To give the reader an idea of what
an exchange rate within a target zone looks like,
the top panel of Figure 1 depicts the log of the
deutsche mark per franc exchange rate from
March 1979 to July 1993. As the relative price
of money, the exchange rate is determined by
market “fundamentals,” that is, output, price
levels, money supplies and interest rates. In the
short run, a relation called uncovered interest
parity (UIP) is thought to control exchange rates.
In the long run, theory suggests that the relative
prices of goods determine exchange rates through
a relation called purchasing power parity (PPP).
U ncovered Interest Parity
Markets for financial instruments have low
transactions costs and very good information,
so small changes in expected asset returns cause
large movements of capital. Expected asset returns
drive exchange rate movements because investors
must exchange currencies to purchase foreign
financial instruments or repatriate earnings from
international investments. For example, if French
interest rates exceed those of Germany, a German
investor might choose to exchange deutsche marks
for francs at the current exchange rate, buy French
financial assets (such as government bonds)
that pay a higher interest rate, and then repur
chase deutsche marks with the francs when
the bond matures.
Of course, if French bonds pay a higher interest
rate, why would any investor choose to buy
German bonds? The answer is that there are
two forms of returns from international invest
ments, the return on the investment itself and
Henderson (1978), Flood and Garber (1984) and Obstfeld
(1984 and 1986) have made important contributions.
25
the return on the exchange rate. Generally
speaking, the expected return for international
assets should be the same for all assets.6 A simple
example illustrates the manner in which the
asset returns and expected exchange rate move
ments interact.
The expected gross return in deutsche marks
for a German investor who invests DM 1000 in
German bonds during period t, for t years, com
pounded annually, is simply
(1) E xpected gross return fo r investing in
German Bonds = 1000-(l + i “ )r ,
where itGe is the annual rate of interest on a
German bond.7 If the same investor exchanged
deutsche marks for francs, bought and held
French bonds, then exchanged the earnings in
francs for deutsche marks, the expected gross
return would be:
(2) E xpected gross return fo r investing in
French Bonds = 1525. (! +
et
)* . E (e
)
= 1000■ (1 + J,Fr )r {E, [ ^ ]).
e,
Define the log of the expected return on the
exchange rate (deutsche marks per franc) from
period “t ” to period “f+ t ” by 8
(3) Et [Ast+I] = \n{Et l ^ } )
et
= l n { E , [ e t +r ] ) - l n { e t ).
For expected returns to be equalized, a higher
French interest rate must be offset by an expected
depreciation in the exchange rate (fewer deutsche
marks per franc in the future). If nominal interest
rates are not too large, equating the right sides of
equations 1 and 2 and using definition 3 gives
6 This is, of course, a simplification. A more accurate statement
would be that the after-tax, risk-adjusted return for different
assets must be the same. Koedijk and Kool (1993) compare
the profitability of investment strategies in different ERM
currencies.
7 If it were not necessary to consider intervals other than a
year, r could be set equal to 1 for simplicity.
us an approximation to the expectation of the
exchange rate change next period:
(4) E L[AS^ A ^ i G s_i Fr>
r
where t is the number of years per period. If
the periods are months, for instance, r = 1/12.
Economists call this relationship UIP.9 Nations
with consistently high inflation rates tend to have
higher nominal interest rates (to compensate
investors for loss of purchasing power) and
depreciating currencies.
Empirical studies have failed to find much
support for the UIP hypothesis among flexible
exchange rate systems (Froot and Thaler, 1990).
This may be due to unrealistic assumptions. UIP
assumes that investors are risk-neutral when, in
fact, there seem to be time-varying risk premia
in the data. Also, there are frequently capital
controls in the real world that prevent investors
from adjusting their portfolios in response to
changes in interest rates or expected exchange
rates. Despite the fact that it has a poor record
of empirical support among flexible exchange
rate systems, UIP is a useful way of thinking
about target zone exchange rates. In contrast
to previous studies on flexible rate systems,
Mizrach (1993a) finds support for UIP in the
well-integrated capital markets of the EU.
Purchasing Pow er Parity
One can buy goods and services as well as
financial assets with money. A higher price
level in France means that one can buy fewer
goods with a given quantity of francs; each franc
is less valuable. PPP says the exchange rate will
adjust downwards to reflect higher prices. That
is, if France maintains a 10 percent higher inflawhich, together with equations 1 and 2, would imply that
E([1/As(+J = 1/E,[Asf+T]. Since, in general, £,[1/As(+T] #
1 /E ([A s (+t ], UIP cannot hold simultaneously in discrete time
for two currencies. This is known as Siegel’s paradox.
Siegel’s paradox was shown to be irrelevant in empirical
work by McCulloch (1975).
8 We will take advantage of the fact that for -.2 < x < .2, a rea
sonable approximation is ln( 1 +x) = x. An immediate appli
cation of this is ln( 1 + ip e) ~ ip e. This means that for small
percentage changes, the log difference of a variable is
approximately the percentage change in the variable.
Define s, = ln(et). Using the approximations and the defini
tions, [(et+1/e t) - 1] « ln{et+1le t) = ln(et+1) - ln(et)
= st+1 - st= A sl+1.
9 If we were to repeat this example from the point of view of a
French investor, we would find an analogous UIP condition
SEPTEMBER/OCTOBER 1994
26
tion rate than Germany, its exchange rate will
depreciate 10 percent per year in the long run.
A variable useful for measuring changes in
relative purchasing power is called the “real
exchange rate.” The real exchange rate in peri
od t[rxt) is defined to be:
e P Fr
& rx< = - p ^ ’
where PtFr and PtGe denote the price levels in
France and Germany in period t, and et denotes
the nominal exchange rate in that period. An
increase in the real exchange rate means that
the franc becomes more valuable, imports will
be cheaper to French consumers but the price of
French exports to Germany rises. French goods
will become less competitive on the world market.
If PPP holds, the real exchange rate will tend to
be mean-reverting; it will tend to return to some
constant level.10 Empirically, evidence supporting
PPP is limited, but PPP remains useful for thinking
about long-run tendencies in exchange rates.11
Both UIP and PPP suggest that a nation which
has a consistently more expansionary monetary
policy will have a currency that will tend to
depreciate. The depreciation will occur through
the inflation premium built into the nominal
interest rate according to UIP, and through rising
prices of domestic goods which require that the
home currency lose value relative to foreign cur
rencies to keep the real exchange rate constant
according to PPP.
TARGET ZONE EXCHANGE RATE
SYSTEMS
A target zone is a hybrid exchange rate regime,
a compromise between managed floating and
completely fixed exchange rates. In a managed
float, monetary authorities may or may not, at
their discretion, intervene to control the rate of
exchange. If monetary authorities fix the exchange
10 Roughly speaking, a random variable, such as the real
exchange rate, that can be forecasted accurately far into
the future is said to be mean-reverting. A mean reverting
process is one that will tend to return its usual value in the
long run.
11 Barriers to trade, transportation costs, differing baskets of
goods across countries, imperfect competition, nontraded
goods and differentiated goods may all contribute to weak
ening the effects of PPP. For an investigation of PPP within
the EMS, see Edison and Fisher (1991). Coughlin and
Koedijk (1990) review the literature on the determination
of the real exchange rate in the long run. Dueker (1993)
investigates PPP with the more recent econometric technique
of fractional integration.
FEDERAL RESERVE BANK OF ST. LOUIS
rate, they willingly buy or sell their own currency
in unlimited quantities at the fixed rate. A target
zone exchange rate system has elements of each.
Monetary authorities pledge to intervene in the
market to keep the domestic exchange rate with a
particular foreign currency, or basket of currencies,
within narrow margins around a central parity.
Realignments occur when central banks are un
willing (or find it too costly) to conduct the inter
ventions necessary to preserve the target zone.
The ERM
The most important target zone, the ERM,
has operated since March 1979 to prevent what
was perceived to be the excessive volatility in
exchange rates that had prevailed in the 1970s.12
The target zones for each currency were initially
established at ±2.25 percent around the bilateral
central parities for most of the currencies, ± 6 per
cent for the more volatile currencies such as the
Italian lira, Spanish peseta, British pound and
Portuguese escudo.
It is common to divide the period of the ERM
into three sub-periods. The first period extends
from the inception of the ERM in March 1979 until
the end of 1983. The target zones were charac
terized by lack of credibility and frequent deval
uations during this period. The second period
lasted from 1984 to the end of 1991 and coincided
with increasing confidence in the ERM and greater
convergence in the economic fundamentals of
the member nations. Figure 1 illustrates four
devaluations of the French franc relative to the
deutsche mark in the first period and only two
in the second period.13 It was widely thought in
1989 and 1990 that the target zones had become
permanent and would never be realigned but
would simply lead into monetary union, a system
of permanently fixed exchange rates with one
monetary authority. This would effectively
mean one currency. Events would prevent this
smooth transition.
12 For more information on the history and practices of the
EMS, see Fratianni (1988), Lingerer, Hauvonen, LopezClaros and Mayer (1990), Zurlinden (1993), Edison and
Fisher (1991), Bean (1992) and Higgins (1993).
13 The data in Figure 1 ends shortly before the widening of
the target zones to ±1 5 percent for all rates except the
guilder/deutsche mark in August 1993, which was a de facto
realignment and the practical suspension of the system. See
Zurlinden (1993) for a full description of the evolution of the
bilateral central parities in the ERM.
27
Figure 1
Deutsche Mark Per Franc Exchange Rate
(March 1979 through July 1993)
In levels of normalized exchange rates
French-German 3-Month Interest Rate Differentials
(March 1979 through July 1993)
The third period for the system was the time
leading to the crises and suspension of the system.
German unification and the recession in Europe
are widely accepted as the underlying causes of
the crises of September 1992 and August 1993.,4
Reunification opened up major investment oppor
tunities in the undeveloped East, increasing the
demand for deutsche marks and required the
German government to spend a great of money
to subsidize the East and bring it up to western
standards. The government also agreed to con
vert East German ostmarks to West German
deutsche marks on a very generous 1:1 basis.15
This one-time expansion of the money supply
raised fears of inflation. High German interest
rates put upward pressure on the deutsche mark.
At the same time, a recession was ravaging Europe,
striking Britain and Italy particularly hard.
Pressure mounted on the Bank of England and
the Bank of Italy to lower interest rates to fight
their recessions, while the Bundesbank resisted
lowering money market interest rates due to fear
of inflation. Furthermore, the Danish rejection
of the Maastricht treaty in June 1992 put the
European Monetary Union (EMU) in jeopardy.
This was the catalyst for the speculative attack of
September 1992, which drove the British pound
and the Italian lira from the ERM.16 The pressure
mounted over the next year as speculation against
the remaining weaker currencies continued.
Finally, in August 1993, the ERM was effectively
suspended as bilateral bands were widened from
±2.25 percent to ±15 percent for all the rates
except the Dutch guilder/deutsche mark rate.
14 Higgins (1993) and Zuriinden (1993) examine the events
leading to the collapse of the ERM in more detail.
16 See Zuriinden (1993) for a detailed description of the experi
ences of the British pound in the ERM.
15 The exchange of deutsche marks for ostmarks was not
unlimited on a 1:1 basis. Bofinger (1990) provides a more
detailed account of these events.
SEPTEMBER/OCTOBER 1994
28
THE CREDIBILITY OF TARGET
ZONES: FORECASTING REALIGN
MENTS
Realignments have been a common feature
of target zone systems. This section surveys
the research on realignments of target zones
conducted in the last several years. This litera
ture has focused on a number of related issues
such as the credibility of a particular target zone,
the probability of a realignment and the expected
size of a realignment. Economists have had little
success in forecasting financial variables such as
exchange rates.'7 Target zone exchange rates may
be different, however. Central banks manage
exchange rates to promote full employment or
low inflation or some other economic goal;
they do not conduct monetary policy for profit.
Knowledge of economic variables may be used
to forecast their policies. Expectations that the
monetary authorities will prefer to realign rather
than defend the target zone will lead investors
to demand an interest rate premium to hold the
weak currency. Therefore, clear expectations
of a devaluation will be accompanied by a high
interest rate differential between the currencies.18
The Simplest Test o f Target Zone
Credibility
This test is constructed to evaluate a weak
currency that is expected to stay the same or
depreciate. Recall that we developed a forecast
for expected future exchange rate changes based
on interest rate differentials, UIP:
f r l
E t [ A S f+r ] _
( o j ------------------- - J ,
T
.fie
-1 ,
-Ft
■
The intuition behind equation 6 is that investors
must be compensated by a higher interest rate
for holding assets denominated in a currency
that is expected to lose value (depreciate).
In a target zone, the most that the exchange
rate could depreciate without a realignment is
the distance from the exchange rate to the lower
bound. Denote this distance in percentage
terms (it must be a nonpositive number):
17 There is a good reason for this. If someone could predict
the future movement of an asset price (for example, an
unusual increase in a stock price) based on public informa
tion, that person would borrow money to buy as much stock
as possible immediately, driving the price up right away.
This is a simple version of the “efficient markets hypothesis.”
If price changes could be easily anticipated, they would
already have happened.
FEDERAL RESERVE BANK OF ST. LOUIS
(7) d t = -= --l = s - s ( ,
et
where e is the lower bound of the target zone,
s = ln[e) and st = ln(et). If the target zone is per
fectly credible (no probability of a realignment),
the expected depreciation in the exchange rate
can be no greater than the distance from the
exchange rate to the bottom of the band. That
is, for all period lengths we must have
(8) E t [Asf+1 ] > d t .
In a perfectly credible target zone, at a forecast
horizon of length (1/r), we must have
(9)
T
(i(Ge- i (Fr)> d ( .
As rgoes to zero, that is, as the forecast horizon
becomes arbitrarily short, equation 9 must hold;
the right side is less than or equal to zero and
the left side is going to zero. If equation 9 fails
to hold, we can conclude the target zone is not
perfectly credible; devaluation is considered
possible.
The converse is not true, however. There
could be significant realignment expectations with
equation 9 still holding. For example, suppose
that the deutsche mark per franc rate is currently
at central parity so dt = -2 .2 5 percent, i,Fr = 4
percent and itGe = 2 percent. Further, investors
know it to be equally likely that either there will
be no realignment and the exchange rate will be
exactly the same a year (r = 1) from now or that
there will be a realignment and th e ex ch a n g e rate
will be exactly 3 percent lower. That means that
in this case, equations 8 and 9 would hold but
the target zone is not perfectly credible since
there is a 50 percent chance of a devaluation
(realignment downward).
Formal tests of target zone credibility or
realignment probabilities are usually based on
the information content of interest rate differen
tials. The greater the risk of devaluation, the
higher the difference in interest rates. An exam
ple of the relation between exchange rates and
18 There are other methods for determining the credibility of
target zones, such as those in Koedijk and Kool (1993), but
this article will focus on those methods using interest rate
differentials.
29
Figure 2
Deutsche Mark/Franc Within the Band Minus Adjusted
Interest Differentials
Percent
8
6
4
,
i
t h
2
^
j | p
l
fWj
0
..
.
--- i~^— r----- 1
-----1
-----r —
— h -----1
1979
81
83
85
interest rate differentials is shown in Figure 1.
The top panel shows the time series of the
exchange rate with the devaluations and the
bottom panel shows the corresponding series
of the French three-month interest rates minus
German three-month interest rates.19 The interest
rate differential was always greater than 0; the
expectation was always that the French franc
would depreciate. The bottom half of Figure 1
shows that interest rate differentials tend to widen
before realignments (vertical lines).
Figure 2 displays the time series of the deutsche
mark per franc exchange rate within the target
zone minus the adjusted three-month interest
rate differential. This series is equivalent to the
guaranteed excess return from investing in French
securities over German securities conditional on
the band remaining intact. In the notation used
above, it is
(10) dt -r-(i(Ge - i(Fr).
This variable indicates a lack of credibility at the
i
87
i
i
89
i
i
91
i
1993
three-month horizon for the target zone when it
is greater than zero. This is the “simplest test”
of target zone credibility. Thus, Figure 2 shows
the target zone lacked credibility most of the time
in the early 1980s, gradually falling below zero
later in the decade as French inflation fell.
In “The Simplest Test of Target Zone Credibility,”
Lars Svensson (1991) uses equation 9 to examine
if interest rates were high enough to conclude
that there must be some devaluation expectation
for the Swedish target zone from 1987 to 1990.
The data are monthly. During the period of
Svensson’s study, Sweden had a unilateral target
zone with a trade weighted “basket” (or weighted
average) of the currencies of its 15 largest trading
partners. Hence, the relevant exchange rate is
now measured in basket units per krona and the
respective interest rates are in basket units and
krona. The width of the band was 1.5 percent
during this period. Svensson plots the return
available on domestic securities (for 12-month
maturities) against the maximal return (in Swedish
krona) on the weighted basket of foreign securities,
19 The periods of realignments are marked in the bottom panel
by vertical lines.
SEPTEMBER/OCTOBER 1994
30
assuming the target zone would remain intact.
He found the Swedish target zone lacked credi
bility with the ECU for securities with a 12-month
horizon from the third quarter of 1989 until the
end of the sample in 1990.
The hypothesis of UIP is used to investigate
credibility in the same way as the “simplest test.”
Recall that UIP expressed the expected movement
in (basket units per Swedish krona) exchange
rates as:
(11) £,[A s,„] = T - ( i " - i f " ) .
This expression is also called the expected rate
of devaluation. By using the interest rates for
securities of different maturities, Svensson is able
to construct a series of forecasts for the future
value of the exchange rate. For example, the
forecast for the exchange rate in two years was
constructed using the 24-month Euro-currency
interest rates for the basket of currencies and
the Swedish krona in equation 11 to get the
expected change in the weighted exchange rate
over that period. If the forecasted exchange rate
fell outside the target zone for a particular matu
rity at some point, the target zone was said to
lack credibility at that forecast horizon.
Svensson used maturities of 12, 24 and 60
months over the sample period to conclude that
while the market generally found the Swedish
target zone to be credible in the short run, there
was strong evidence that the market also always
believed that devaluation within a longer horizon
(24 to 60 months) was a distinct possibility.
Expected exchange rates always fell outside the
target zone for those maturities for the sample
period.
M ean Reversion Within the
Target Zone
A major problem with using UIP to estimate
the credibility of target zones is that it predicts
movements in the exchange rate, not the central
parity. The movement of the exchange rate within
the band, especially at short horizons, could
account for much or all of the interest rate dif
ferential. At longer horizons, the interest yield
for securities gets larger (as more interest accrues
over time) but the exchange rate within the band
is still bounded. For example, if the target zone
is 2.25 percent wide (as were the ERM target
20 (12/3)*.0225 = .09
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
zones before August 1993) and the exchange rate
is at central parity, the simplest test tells us that
the interest rate differential on 12-month securities
would have to exceed 2.25 percentage points
(the width of the band) before we could reject
the idea that the target zone is perfectly credible.
But, the same test tells us the annualized interest
differential for three-month securities would
have to exceed 9 percentage points before we
could reach the same conclusion.20
To more accurately estimate the credibility of
the target zone, at short horizons, it is necessary
to estimate the movement of the exchange rate
within the band. Investigating this matter, Rose
and Svensson (1991) find that daily deutsche
mark per franc rates within the band tend to be
mean-reverting, that is, they tend to come back
to central parity if they are away from it. The
mean reversion is due to the fact that monetary
authorities will usually defend the target zone
by intervening to move the exchange rate back
to the center of the target zone if it approaches
the edges.
To explain how movements of the exchange
rate within the band are forecasted, define the
log of the position of the exchange rate within
the band as
(12) x ( = s, —c t ,
where ct is the log of the central parity of the
band at time t. Note that xt may be positive or
negative. Of course, one may rewrite the
exchange rate as the sum of the central parity
and the position within the band as
(13) s, = x t + c t,
and by taking differences (percentage changes)
of this equation over time we get
(14) As, = Ax, + Ac,.
Using the UIP condition stated earlier and rear
ranging terms, we may express the expected
change in central parities (the expected realign
ment) as
(15) E t [Ac,+r ] = T ■(i?°sket ~ if"') - E t [Axt+r}.
Equation 15 illustrates that to more accurately pre
dict changes in the central parity (realignments), it
31
is necessary to predict the way exchange rates
might move within the band.
Rose and Svensson (1991) make the additional
assumption that the future movements of the
exchange rate within the band might be predicted
from present position and other ERM exchange
rates with the deutsche mark. They use an ordi
nary least-squares regression to predict the
changes in the exchange rate within the band
for the next month (i?f[Axf+T]). They find that future
changes in the exchange rate are dominated by
current position within the band. If the exchange
rate is near the edges, it will tend to come back
to the middle. Other variables, including other
ERM exchange rates, lagged changes and higherorder terms were found to be statistically or eco
nomically insignificant.
In order to predict the rate of expected
realignment (if([Act+T]) they substitute the fore
cast for the change in the exchange rate within
the band (£'([Ax(+T]) into equation 15 to predict
realignments. They report some success, but
suggest that since the expected rate of realign
ment consistently “overpredicts” realignments,
private agents may not anticipate realignments
very well. Since their model is based on market
expectations—high interest rate differentials—
misprediction by private agents may degrade
its performance.
Expectations
The question of why private agents may fail
to anticipate realignments is puzzling to econo
mists. Kaminsky (1993) attributes this lack of
success in predicting exchange rate movements
in general to the fact that agents must “learn”
about the nature of the economy and the behavior
of the monetary authorities. W h ile th ey are
learning, they may make systematic mistakes
about the credibility of the authorities or the
nature of shocks hitting the economy. The
question of how private agents develop their
expectations and beliefs about the economy is
an important one. If central banks knew how
to influence expectations of devaluation, they
could prevent speculative attacks and stabilize
the exchange rate.
The UIP relation tells us something about
expectations; interest rate differentials forecast
expected movement, but the story is not as simple
as that presented in section two. Investors care
not only about expected profit, but also about
minimizing risk associated with the profit. For
instance, German investors buying domestic bonds
are sure of their nominal return, but if they buy
French bonds, they must also take the risk that
exchange rates will not move as predicted. If the
exchange rate depreciates more than expected,
they lose money. Because of this risk, investors
require a “risk premium” in the form of an espe
cially high interest rate to hold certain currencies.
This risk premium may also change over time as
economic conditions change and investors per
ceive more or less risk in the exchange rate. This
time-varying risk premium makes it difficult to
accurately estimate expectations from interest
rate differentials.
An obvious way to investigate agents’ expec
tations about the exchange rate is to ask them.
Frankel and Phillips (1991) use this method to
investigate the hypothesis of increasing EMS
credibility after 1987 (until 1991). With the survey
data method from the Currency Forecasters’ Digest
(CFD) as well as the UIP method, Frankel and
Phillips examine whether forecasts of future
exchange rates fall within the target zone for
monthly EMS exchange rates. They consider the
main advantage of survey data to be immunity
from error due to exchange rate risk premia. The
closer the forecast is to the central parity, the more
credible the target zone.21 Prior to 1990, estimates
of the expected annual rates of devaluation were
about 2-5 percent for most currencies. These
estimates tended to overpredict actual devalua
tions. Their study concludes that between 1987
and 1991, the EMS experienced a significant gain
in credibility using one- and five-year horizons.
That is, one- and five-year forecasts of the
ex ch a n g e rate move much closer to current
central parity after 1987.
UIP and survey data approaches are useful
to inform us as to the expectations of market
participants with respect to the exchange rate,
but they do not tell us how these expectations
are formed. Using Swedish data from 1982 to
1991, Lindberg, Svensson and Soderlind (1991)
consider this problem of explaining time-varying
market devaluation expectations in terms of
underlying factors. They first use a variant of
the “simplest test” to compute devaluation
expectations over time for one-, three-, six- and
21 Their methods are very similar to Svensson’s “simplest test”
discussed above.
SEPTEMBER/OCTOBER 1994
32
12-month forecast horizons. Generally, they
were unable to find much incidence of a lack
of credibility at short forecast horizons.22
Lindberg, Svensson and Soderlind (1991)
attribute the failure to find a lack of credibility at
shorter horizons to ignoring expected changes
within the band. As discussed in the context of
mean reversion, changes within the band may be
large relative to interest rate differentials at short
horizons. To get more precise estimates of deval
uation expectations, Lindberg, Svensson and
Soderlind (1991) required a specification for future
values of the exchange rate. Theory suggested
starting with a simple log linear specification:
(18) x, = P 0 + P 1 - x ^ .
Although they considered a variety of explana
tory variables and methods to estimate equation
18 and its variants, a simple OLS regression
with a Newey-West correction for conditional
heteroskedasticity to the errors worked best for
estimating changes within the band. The gains
to precision were described as “substantial” for
short horizons.
With the new devaluation expectations series,
Lindberg, Svensson and Soderlind examine the
circumstances around four specific periods of
high realignment expectations. The first period,
October 1982, was the only time that the target
zone was actually realigned. The market seemed
to have weakly anticipated it two to three months
before it occurred. The high realignment expec
tations in the spring of 1985 were ascribed to the
e le c tio n of a n ew g ov ern m en t and uncertainty
about the width of the band.23 The third period
of high realignment expectations was also asso
ciated with political events, the political crisis
and weak economy of the first three quarters of
1990. Finally, high realignment expectations in
the late fall of 1990 were also imputed to fears
that the government would change the target zone
before the general election of September 1991.
In a more formal investigation of how expec
tations are formed by political events and
macrovariables, Lindberg, Svensson and
Soderlind regressed devaluation expectations
on variables such as changes in the real exchange
rate, parliamentary elections, changes in foreign
22 There was a lack of credibility at all horizons before the only
actual devaluation (October 1982) and around the time of an
election (September 1985). In addition, the target zone fre
quently lacked credibility at the 12-month forecast horizon.
Digitized for FEDERAL
FRASER RESERVE BANK OF ST. LOUIS
exchange reserves, unemployment, money growth,
government borrowing and the current account.
Only changes in the real exchange rate, parlia
mentary elections and the current account proved
to be significant explanatory variables. The
coefficients on these significant explanatory
variables were unstable over subperiods, however,
perhaps indicating the shifting focus of market
participants as they develop their expectations.
Rose and Svensson (1993) extended the efforts
to learn about the causes and behavior of realign
ment expectations during the EMS. They regressed
realignment expectations on measures of relative
money, output, the real exchange rate, inflation,
the trade balance, reserves and exchange rate
volatility within the band. They found no
robust link between realignment expectations
and the macroeconomic variables. Use of a
vector autoregressive system had no more suc
cess. They conclude that there is “no apparent
relationship between macroeconomic variables
and credibility” (p. 16).
After examining the behavior of macroeco
nomic variables and political events before
the currency crises of 1992 and 1993, Rose
and Svensson find it difficult to convincingly
explain the cause and suddenness of the crises.
Although it is easy to claim ex post that the
macroeconomic fundamentals dictated a revalu
ation of the deutsche mark, “it remains a mystery
that the deepest financial markets in the world
yielded so remarkably few indications of an
imminent crisis” (p. 26). Furthermore, the
weak link between realignment expectations
and m a c ro e co n o m ic v a ria b les is trou b lin g.
Truncated Data
An often ignored problem in working with
data from target zone exchange rate systems is
that the data are “truncated. ” This is a problem
for statistical research on this data; much com
monly used statistical theory assumes the distri
bution of the random variable to be unbounded.
Chen and Giovannini (1992) suggest transforming
the exchange rate into the following unbounded
random variable:
(19) Z, = l n [ ^ ] ,
L —x t
23 The width of the target zone was not public information at
this time.
33
where L = ln {e/ct), e is the upper edge and c t is
the central parity of the target zone.
Working with the transformed random variable
zt, Chen and Giovannini investigate target zone
credibility in the usual ways using monthly data
from the ERM and the Bretton Woods system.24
With a linear prediction of the exchange rate
within the target zone, they estimate band credi
bility from the UIP relationship. Their confidence
intervals for the expected changes within the band
are actually constrained by the band (by con
struction) whereas the confidence intervals for
the untransformed variables frequently fall outside
the target zone. This property rules out nonsen
sical values for expected changes within the band
and means a better estimation of the process. As
in other studies, they are able to frequently reject
perfect credibility for ERM zones during the 1980s.
The Probability and Size of
Realignments
The simplest test of target zone credibility
only predicts the expected rate of devaluation
U ^.JA sJ) over a period of time. It does not
predict the probability of realignment over that
period, nor does it predict the size of a realignment
conditional on one occurring. The simplest test
is unable to differentiate between an almost cer
tain small realignment and a low probability of
a large realignment.
Recently, Mizrach (1993b and 1993c) has used
a hybrid Markov-Probit model to estimate the
probability of realignment and the expected size,
conditional on an occurrence. The probability
of realignment estimated by a probit model uses
the log of the position of the exchange rate within
the band, and the domestic yield curve as inde
pendent variables. The log of the exchange rate
within the band is again modeled as a linear
autoregression; lagged values of xt predict future
values. The expected size of an exchange rate
movement, conditional upon a realignment, is
allowed to depend on the real exchange rate.
Nonlinear least-squares were used to estimate the
model on daily data from the ERM, the FF/DM
and IL/DM exchange rates.
Mizrach found strong evidence of mean rever
sion within the band; the parameter estimates
suggest that any deviation from central parity
would be expected to be cut in half in a week or
two. The model forecasts systematically larger
realignments than actually occurred for both the
franc and the lira. The probit parameters all
were significant and had the appropriate sign.
Restrictions of constant realignment risk and
no mean reversion were strongly rejected.
It was found that, typically, probabilities were
at usual levels up until a month before a realign
ment and then began climbing upwards. The short
nature of the warning time provided by the model
leads Mizrach to conclude that realignments
“surprised” market participants and policymakers.
Mizrach concludes that his model supports the
hypotheses of mean reversion within the band
and produces credible estimates of time-varying
realignment risk.
The Role o f the Dollar
The empirical work discussed above does
not use a potentially important indicator of
realignments, weakness in the U.S. dollar. As
noted by Edison and Kole (1994) and others,
realignments tend to be associated with weak
ness in the U.S. dollar. The role of the dollar
and the deutsche mark as international stores
of value is the explanation for this. When the
dollar is weak, investors substitute into deutsche
mark-denominated assets. This increases the
value of the deutsche mark not only with respect
to the dollar but also to other ERM currencies.
This added pressure in times of crisis has fre
quently contributed to realignments.
CONCLUSIONS
This article has surveyed recent work on
forecasting realignments and estimating the
credibility of target zones. The literature has
found that realignments are predictable to some
extent within short intervals from readily avail
able information such as interest rates and the
position of the exchange rate within the band.
Most of the research surveyed here has taken
the formation of expectations for granted and has
used interest rate differentials which develop
from those expectations as starting points for fore
casting realignments. The relationship between
realignment expectations and macrovariables is
weak and uncertain. It is not clear how expecta-
24 While generally described as an adjustable-peg fixed-rate
system, the Bretton Woods system is more accurately
described as a narrow target zone system. The target
zones were ± 1 percent around dollar parities.
SEPTEMBER/OCTOBER 1994
34
tions are formed. Further, realignments are said
to “surprise” policymakers and market partici
pants; realignment expectations rise only a short
time before realignments. To some extent, this
is to be expected. Although there are false alarms
in which realignment expectations rise and then
fall back again, once realignments are seen as
likely, speculative pressure builds up that often
results in a self-fulfilling speculative attack.
Further research on the formation of expecta
tions would be an important contribution.
REFERENCES
Bean, Charles R. “Economic and Monetary Union in Europe,”
The Journal o f Economic Perspectives (fall 1992), pp. 31-52.
Bofinger, Peter, “The German Monetary Unification (Gmu):
Converting Marks to D-Marks,” this Review (July/August
1990), pp. 17-36.
Chen, Zhaohui, and Alberto Giovannini. “Estimating Expected
Exchange Rates Under Target Zones,” NBER Working
Paper No. 3955 (January 1992).
Corbae, Dean, Beth Ingram, and Guillermo Mondino. “On the
Optimality of Exchange Rate Band Policies,” University of
Iowa Working Paper 90-04 (March 1990).
Coughlin, Cletus C., and Kees Koedijk. “What Do We Know
About the Long-Run Real Exchange Rate?” this Review
(January/February 1990), pp. 36-48.
Dueker, Michael J. “Hypothesis Testing with Near Unit-Roots:
The Case of Long-Run Purchasing Power Parity,” this
Review (July/August 1993), pp. 37-48.
Edison, Hali J., and Eric O’ N. Fisher. “A Long-Run View of the
European Monetary System,” Journal o f International
Money and Finance (March 1991), pp. 53-70.
_____ , and Linda S. Kole. “European Monetary Arrangements:
Implications for the Dollar, Exchange Rate Variability and
Credibility,” Board of Governors of the Federal Reserve
System, Division of International Finance (April 1994).
Engel, Charles, and Craig S. Hakkio. “Exchange Rate
Regimes and Volatility,” Federal Reserve Bank of Kansas
City Economic Review (third quarter 1993), pp. 43-58.
Flood, Robert P., and Peter M. Garber. “Collapsing ExchangeRate Regimes: Some Linear Examples,” Journal of
International Economics (August 1984), pp. 1-13.
Frankel, Jeffrey, and Steven Phillips. “The European Monetary
System: Credible At Last?,” NBER Working Paper No. 3819
(August 1991).
Fratianni, Michele. ‘The European Monetary System: How
Well Has It Worked?” Cato Journal (fall 1988), pp. 477-501.
Froot, Kenneth A., and Richard H. Thaler. “Anomalies: Foreign
Exchange,” The Journal of Economic Perspectives (summer
1990) pp. 179-92.
Koedijk, Kees G., and Clemens J. M. Kool. “Betting on the
EMS,” Open Economies Review (1993), pp. 151-73.
Kool, Clemens J. M. “The Case for Targeting Domestic Money
Growth under Fixed Exchange Rates: Lessons from the
Netherlands, Belgium, and Austria: 1973-1992,” unpublished
manuscript (October 1993).
Krugman, Paul, and Marcus H. Miller. “Why Have a Target
Zone?” Centre for Economic Policy Research Discussion
Paper Series No. 718 (October 1992).
Lindberg, Hans, Lars E. O. Svensson, and Paul Soderlind.
“Devaluation Expectations: The Swedish Krona 1982-1991,”
NBER Working Paper No. 3918 (November 1991).
Macleod, Alexander, “After Currency Swings, Europe Seeks
Order,” The Christian Science Monitor, Wednesday
September 30, 1992.
McCulloch, J. Huston. “Operational Aspects of the Siegel
Paradox,” The Quarterly Journal o f Economics (February
1975), pp. 170-2.
Mizrach, Bruce. “Uncovering Interest Rate Parity in the ERM”,
Federal Reserve Bank of New York, unpublished manuscript
(October 1993a).
_____ . “Mean Reversion in EMS Exchange Rates”, Federal
Reserve Bank of New York, unpublished manuscript (June
1993b).
_____ . “Target Zone Models with Stochastic Realignments:
An Econometric Evaluation”, Federal Resen/e Bank of New
York, unpublished manuscript (revised, April 1993c).
Neely, Christopher J. “Target Zones and Conditional Volatility:
An ARCH Application to the EMS,” Federal Reserve Bank of
St. Louis Working Paper No. 94-008 (December 1993).
Obstfeld, Maurice. “Rational and Self-Fulfilling Balance-ofPayments Crises,” The American Economic Review (March
1986), pp. 72-81.
_____ . “Balance-of-Payments Crises and Devaluation,” Journal
o f Money, Credit and Banking (May 1984), pp. 208-17.
Pollard, Patricia S. “Central Bank Independence and
Economic Performance,” this Review (July/August 1993),
pp. 21-36.
Rose, Andrew K., and Lars E. O. Svensson. “European
Exchange Rate Credibility Before the Fall,” NBER Working
Paper No. 4495 (October 1993).
_____ , a n d _____ . “Expected and Predicted Realignments:
The FF/DM Exchange Rate During the EMS,” International
Finance Discussion Paper Number 395, Board of Governors
of the Federal Reserve System (April 1991).
Salant, Stephen W., and Dale W. Henderson. “Market
Anticipations of Government Policies and the Price of Gold,”
Journal o f Political Economy (August 1978), pp. 627-48.
Svensson, Lars E. O. “The Simplest Test of Target Zone
Credibility,” IMF Staff Papers (September 1991) pp. 655-65.
Higgins, Bryon. “Was the ERM Crisis Inevitable?” Federal
Reserve Bank of Kansas City Economic Review (fourth
quarter 1993), pp. 27-40.
Ungerer, Horst, Jouko J. Hauvonen, Augusto Lopez-Claros,
and Thomas Mayer, “The European Monetary System:
Developments and Perspectives,” IMF Occasional Paper
No. 73 (November 1990).
Kaminsky, G. Faciela. “Is there a Peso Problem: Evidence
from the Dollar/Pound Exchange Rate, 1976-1987,” The
American Economic Review (June 1993), pp. 450-72.
Zurlinden, Mathias. ‘The Vulnerability of Pegged Exchange
Rates: The British Pound in the ERM” this Review
(September/October 1993), pp. 41-56.
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
35
R. Alton Gilbert
R. Alton Gilbert is a vice president at the Federal Reserve Bank of
St. Louis. Christopher A. Williams provided research assistance.
A Case Study in Monetary
Control: 1980-82
OR SEVERAL YEARS PRIOR to October 1979,
the Federal Reserve implemented monetary policy
decisions of the Federal Open Market Committee
(FOMC) by targeting the federal funds rate. Staff
of the Open Market Desk bought or sold govern
ment securities with the objective of keeping the
federal funds rate within a range specified by
the FOMC at its latest meeting.
procedure to promote better short-run control of
the monetary aggregates, to better control infla
tion.' Under the NBR operating procedure, the
objective of the staff of the Open Market Desk was
to keep the average level of NBR between FOMC
meetings at levels consistent with the short-run
objectives of the FOMC for growth of the mone
tary aggregates.
The effects of monetary policy on the economy
under a procedure of targeting the federal funds
rate depend on the willingness of policymakers
to move the funds rate target fast enough and far
enough when the pace of economic activity
changes. In the 1970s, the tendency of the Fed
to limit changes in the federal funds rate as the
growth of total spending accelerated produced
rapid money growth, resulting in accelerating
inflation in the late 1970s.
The Fed stopped targeting NBR in the fall of
1982; the operating procedure used since then is
similar to targeting the federal funds rate.2
In response to the accelerating inflation, the
Fed in October 1979 adopted a procedure of
targeting nonborrowed reserves (NBR). The
FOMC stated that it adopted the NBR operating
1 For a description of the decisions by the FOMC at its meeting
in October 1979, see Board of Governors (1979, p. 974).
2 For a general description of the mechanics of various oper
ating procedures, see Gilbert (1985). Thornton (1988) pro
vides evidence that targeting borrowed reserves has been
essentially the same as targeting the federal funds rate.
The NBR operating procedure generated a great
deal of interest and controversy among econo
mists. There is a large literature on the conduct
of monetary policy under that procedure and,
in recent years, economists have continued to
analyze the conduct of monetary policy during
the three years ending in the fall of 1982.3 Critics
of the NBR procedure contend that it caused a
high degree of interest rate volatility, as illustrated
in Figure 1. Some critics argue that the Fed
actually did not change its operating procedure
1986); Hoehn (1983); Lindsey (1982, 1983); Lindsey and
others (1984); McCallum (1985); Poole (1982); and Spindt
and Tarhan (1987). For recent additions, see Avery and
Kwast (1993), Goodfriend and Small (1993) and Pearce (1993).
3 The following are selected references to the literature on the
NBR operating procedure: Goodfriend (1983); Hetzel (1982,
SEPTEMBER/OCTOBER 1994
36
Figure 1
Weekly Federal Funds Rate: January 3,1979, to December 28,1983
Percent
1979
1980
1981
1982
1983
Note: Shaded area encompasses the period of nonborrowed reserves targeting
(10/3/79 through 9/29/82).
in any fundamental way in October 1979.4 Others
blame large errors in hitting money targets on
improper design of the operating procedure,
especially in combination with lagged reserve
accounting in effect at the time.5
Whatever the flaws in the NBR targeting
procedure as a method of monetary control, the
Federal Reserve did achieve its objective of sharply
reducing the rate of inflation during the period
in which it used that procedure (Figure 2). That
success in reducing the rate of inflation, however,
came at the price of a very sharp recession
(Figure 3).
This article extends the literature on NBR
targeting in two ways. First, it presents informa
tion relevant for interpreting policy actions that
was confidential until several years after the end
4 See Poole (1982).
5 See McCallum (1985). Gilbert and Trebing (1982) provide
a description of lagged and contemporaneous reserve
accounting.
6 The weekly reports of the Manager of the Open Market
Account, which included the projections and estimates of
TR, became public information five years after the dates of
the reports. Cook (1989a, 1989b) presents some, but not
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
of the period of NBR targeting: Federal Reserve
staff projections of total reserves (TR) over periods
between FOMC meetings, and staff estimates of
the levels of TR over the same periods that would
have been consistent with FOMC objectives for
growth of the monetary aggregates (the TR paths).6
In addition, this article extends the literature by
answering a question not answered by the other
studies: Did the pattern of policy actions under
the NBR operating procedure reflect a consistent
use of the procedure for hitting short-run targets
for growth of the monetary aggregates, given the
information available to policymakers on staff
projections of TR and estimates of the TR paths?
This article may have implications for the
choice of operating procedure in the future. If
the Federal Reserve chose once again to target a
all, of the information on the NBR operating procedure pre
sented in this article. In particular, Cook presents information
on the gap between projections of TR and the TR path, but
he does not present the levels of those projections and esti
mates. Feinman (1988) made extensive use of the data
from the weekly reports of the Manager of the Open Market
Account in an unpublished dissertation.
37
Figure 2
Rate of Growth in the GDP Deflator
1975
77
79
81
83
1985
Note: Rates of growth in the GDP deflator are two-quarter growth rates;
the shaded area encompasses the period of nonborrowed reserves
targeting (1979:Q4 through 1982:Q3).
Figure 3
Rate of Real GDP Growth
1975
77
79
81
83
1985
Note: Rates of real GDP growth are two-quarter growth rates; the shaded
area encompasses the period of nonborrowed reserves targeting (1979:Q4
through 1982:Q3).
SEPTEMBER/OCTOBER 1994
38
narrow monetary aggregate, the Federal Reserve
might consider a change in operating procedure,
perhaps to an NBR operating procedure. Several
prominent monetary economists have expressed
dissatisfaction with the lack of success of the
FOMC in hitting its targets for money growth
under NBR targeting.7 It is not possible to evaluate
NBR targeting as a method of monetary control
from the experience of 1979-82, however, without
knowing whether policy actions were consistent
with use of the procedure for monetary targeting.
procedure is unclear. On several occasions, the
FOMC widened the range on the federal funds
rate when the rate threatened to move outside
the range. On other occasions, the federal funds
rate was allowed to move outside its range for
short periods of time.9
TARGETING NONBORROWED
RESERVES
S taff Projections o fT R and Estimates
o f the TRPath
This section describes the nature of the NBR
operating procedure. Most members of the FOMC
at the special meeting on October 6, 1979, agreed
that the degree of monetary control under the
procedure of targeting the federal funds rate had
become unsatisfactory. They decided to adopt
instead a procedure that linked the supply of
NBR to their objectives for money growth, while
permitting larger fluctuations in the federal funds
rate than under the previous procedure of federal
funds rate targeting.8
After each FOMC meeting, the staff would
estimate the average level of TR that would be
consistent with the FOMC’s objectives for growth
of monetary aggregates until the next meeting.
This was called the “TR path.” The target for
the average level of NBR between FOMC meetings,
called the “NBR path,” was simply the TR path
minus the borrowings assumption of the FOMC.
The objective of the Open Market Desk was to
keep the average level of NBR between FOMC
meetings equal to the NBR path.10
Changes in the Nature o f FOMC
Decisions
Staff estimates of the TR path were based on
FOMC objectives for M l and M2 and estimates
of the following: (1) currency in the hands of
the public; (2) average reserve requirements on
deposit liabilities in M l and M2; (3) required
reserves on bank liabilities not included in M l
or M2; and (4) excess reserves. The staff generally
revised their estimate of the TR path each week,
based on new information about the factors that
affected the relationship between reserves and
the monetary aggregates.
Under the federal funds rate targeting proce
dure, the FOMC stated its objectives for growth
of each monetary aggregate between meetings as
a range of growth rates from a month before the
meeting to a month after the meeting. Beginning
with its meeting on October 6, 1979, the FOMC
began specifying its o b je ctiv e s for growth of the
monetary aggregates as specific growth rates over
periods between meetings. Under the federal
funds rate targeting procedure, in contrast, the
FOMC stated its objectives for money growth as
ranges of growth rates of the monetary aggregates.
Although the FOMC continued to specify
ranges for the federal funds rate under the NBR
operating procedure, the ranges were widened
substantially. For most periods, the range was
400 basis points, compared with ranges of 50 to
100 basis points under the federal funds rate
operating procedure. The role that the wider
ranges for the funds rate played in the operating
7 See Friedman (1984), McCallum (1985), Pierce (1984) and
Poole (1982).
8 See Board of Governors (1979, p. 974).
9 See Gilbert and Trebing (1981) and Thornton (1982, 1983).
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
At each meeting, the FOMC also made an
assumption about the average level of borrowed
reserves over the period until the next meeting.
The staff used this “borrowings assumption” in
deriving the target level for NBR.
Each time the staff estimated the TR path,
they also projected the average level of TR over
the same period. Projections of TR were based
on estimates of the actual levels of the monetary
aggregates between FOMC meetings and the four
estimates specified above that were made in
estimating the TR path. Each change in the gap
between the staff projection of TR and their esti
mate of the TR path during an intermeeting period,
therefore, reflected a change in the staff projec
tions of the monetary aggregates. Appendix 1
illustrates the process of projecting TR and esti10 The staff of the Open Market Desk converted the NBR path
for each intermeeting period into weekly and daily objectives
for NBR. See Levin and Meek (1981), Meek (1982) and
Stevens (1981).
39
Figure 4
Supply and Demand for Reserves
Interest rates
Reserves
mating the TR path for the first intermeeting
period in Table 1."
Graphical Representation o f NBR
Targeting
Since projections of TR and estimates of the
TR path reflected information about the same
four variables specified above, projections of
TR often were revised in the same direction as
the estimates of the TR path. In the three weeks
ending February 27,1980, for instance, the projec
tions of TR and the TR path were both reduced,
but by different amounts (Table 1). Changes in
projections of TR and TR paths over the 37 periods
in Table 1 had the same signs in all but eight of
the periods. These comparisons indicate that
changes in projections of TR over intermeeting
periods tended to reflect the same factors that
caused the staff to revise its estimates of the TR
path: changes in factors that affect the relationship
between reserves and the monetary aggregates.
Implementation of monetary policy under
this operating procedure is illustrated in Figure
4, using the concepts of supply and demand for
reserves and equilibrium in the market for reserves
described in Appendix 2.12 Levels of TR and
NBR on the horizontal axis refer to average levels
for the weeks between FOMC meetings. On the
vertical axis, rfl is the level of the discount rate
and rf is the level of the federal funds rate. The
TR path is illustrated as R*. The NBR path is N,
based on a borrowings assumption of R* minus
N. The objective of the Open Market Desk was
to keep the average level of NBR over intermeeting
periods close to the NBR path.
11 Although the Federal Reserve began using the NBR operat
ing procedure in October 1979, the reports of the Manager
of the Open Market Account did not include projections of
TR and TR paths on a consistent basis until February 1980.
Cook (1989b) discusses some of the difficulties in deriving
consistent information from the weekly Reports of Open
Market Operations on the conduct of monetary policy in the
first few weeks under the NBR operating procedure.
TR would be at the path level R* if the demand
12 Lindsey (1982,1983) describes how the procedure of target
ing NBR worked in practice by examining the timing of
money growth relative to FOMC objectives, borrowed
reserves, the federal funds rate and the discount rate. Meek
(1982) describes in detail the operations of the Open Market
Desk under NBR targeting.
SEPTEMBER/OCTOBER 1994
40
Figure 5
Tightening of Monetary Policy
Interest rates
Figure 6
Federal Funds Rate Targeting
Interest rates
FEDERAL RESERVE BANK OF ST. LOUIS
41
curve for reserves was D}. From that initial
position, consider the effects of an increase in
the demand for reserves, illustrated by a shift in
the demand curve to D2, which reflected an
increase in the demand for money.13 TR would
rise to Rlt which is above the TR path. Since the
staff of the Open Market Desk would keep NBR
at the level N, the rise in TR to R1 would involve
an increase in borrowed reserves. The federal
funds rate would rise from rh to r 2j, inducing the
higher level of borrowings. Without any addi
tional policy actions, the money stock would
tend to exceed the FOMC’s objectives because
TR would be above the path level.
During some intermeeting periods, the Federal
Reserve took no policy actions in response to
changes in the demand for reserves. In the case
illustrated in Figure 4, FOMC members consid
ered the rise in the federal funds rate from r^to
r 2^ an adequate response to the shift in demand
for reserves, even if growth of the monetary
aggregates exceeded objectives established at
the last FOMC meeting.
Experience eventually convinced some Federal
Reserve officials that rapid policy responses were
necessary to close the gap between actual money
growth and FOMC objectives once money growth
started to deviate substantially from FOMC objec
tives.14 During some periods between FOMC
meetings, the Federal Reserve adjusted the level
of the NBR path or the discount rate to reduce
the deviations of the money stock from desired
levels. The Federal Reserve took such policy
actions when the deviations appeared to reflect
more than transitory movements in the money
demand schedule, perhaps due to changes in
aggregate spending.15
In the situation illustrated in Figure 5, the
staff projects TR to be i?3, which is above the TR
path (R *). The policy action illustrated in Figure 5
is a reduction in the NBR path from N, to N2,
which involves an increase in the borrowings
assumption from R* minus Nj to R* minus N2.
Due to the inelastic demand for reserves over
intermeeting periods, the average level of TR
would decline to R2, still above the TR path, but
13 If the shift in demand for reserves resulted from an increase
in average reserve requirements on deposit liabilities or
excess reserves, the TR path would shift to the right. The
rise in the demand for reserves would not affect the federal
funds rate.
the reduction in NBR would produce a sharp
increase in the federal funds rate. The Fed could
have the same effect on the funds rate and TR by
keeping NBR at N1 and raising the discount rate
to r 2(j. In taking policy actions that reduced but
did not eliminate the gap between projections of
TR and path levels, Fed officials emphasized the
assumption that sharp increases in interest rates
would, over time, reduce the quantity of money
demanded. This article does not model the
assumed feedback mechanism based on money
demand as a function of lagged interest rates.16
One of the issues policymakers confronted
in determining whether to adjust the NBR path
or the discount rate when TR was projected to
deviate from path levels involved their confi
dence in the projections of TR and estimates of
the TR path. Studies conducted during the period
of NBR targeting indicated large errors in these
projections and estimates.17 These errors would
tend to be smaller later in intermeeting periods,
when actual observations were available for part
of the periods. Observations in Table 1 are con
sistent with the view that the projections and
estimates of TR were subject to large errors, and
that the errors affected the timing of policy
actions. Table 1 indicates that often there were
large revisions to the projections of TR and to
TR paths over intermeeting periods. Also, on
those occasions when policymakers took actions
between FOMC meetings, they generally acted
at least two weeks after an FOMC meeting, when
they might assume that the projections and esti
mates were more accurate.
Graphical Representation o f Targeting
the F ed era l Funds Rate
One way to highlight the nature of NBR tar
geting is to contrast the open market operations
for a given situation under NBR targeting and
under the procedure of targeting the federal funds
rate. Suppose the demand for reserves increases,
reflecting an increase in the demand for money.
Under the NBR targeting procedure, the staff of
the Open Market Desk would continue to target
the same average level of NBR over the interme
diate period (as in Figure 4). If the policymakers
16 For references to this feedback mechanism from changes in
interest rates to changes in the quantity of money demand
ed, see Axilrod (1981, p. A23) and Lindsey (1983).
17 See Levin and Meek (1981) and Pierce (1981).
14 See Axilrod (1981, pp. A23 - A24).
15 See Lindsey (1983, p. 5).
SEPTEMBER/OCTOBER 1994
42
wished to limit the deviation of money growth
from FOMC objectives, they would reduce the
target level of NBR (as in Figure 5). Under the
federal funds rate targeting procedure, in contrast,
the Fed would respond to an increase in the
demand for reserves by increasing the level of
NBR enough to keep the federal funds rate un
changed, as illustrated in Figure 6. This contrast
provides a standard for judging whether Fed
actions in the three years ending in the fall of 1982
were consistent with use of the NBR operating
procedure for targeting the monetary aggregates.
INTERPRETING FEDERAL RESERVE
ACTIONS
The framework of supply and demand for
reserves is used to interpret monetary policy
actions under the NBR operating procedure,
as recorded in Table I .'8
Policy Actions in S elected
Interm eeting Periods
This section illustrates use of the NBR operating
procedure for implementing monetary policy
during the first two intermeeting periods covered
in Table 1. These periods illustrate very different
patterns in use of the procedure. During the first
period, after the FOMC meeting on February 4-5,
1980, the Fed reduced the NBR path and raised
the discount rate when projections of TR began
to rise relative to the TR path. This period illus
trates aggressive use of the procedure for monetary
targeting. During the second period, after the
FOMC meeting on March 18, estimates of TR
declined sharply relative to the TR path, but the
Fed made no adjustments in the NBR path or
discount rate in response.
The period from the FOMC meeting on
February 4-5, 1980, until the next FOMC meeting
was divided into two periods of three weeks each
for purposes of projecting the average level of TR
and estimating the TR path.'9 As of February 7,
the staff projected an average level of TR for the
18 Information on the conduct of monetary policy in Cook
(1989a, 1989b) is similar to that in columns six through nine
of Table 1. One difference involves the dating of the differ
ence between projections of TR and the TR path (column
six) and policy actions (columns seven and eight). The
dates in Table 1 are those in the weekly Report of Open
Market Operations from the Federal Reserve Bank of New
York. Cook dates the gap between the projections of TR
and the TR path and dates policy actions as of weeks end
ing on Wednesdays, thus reflecting the changes that
occurred during each seven-day period. For this reason, the
dates in Table 1 and in Cook (1989a, 1989b) do not match.
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
three weeks ending February 27 that was only
$38 million below the initial estimate of the TR
path. By February 15, however, the projections
and estimates of TR had changed substantially,
with TR projected to be $313 million above the
path level. As of February 15, the Fed reduced
the target for NBR by $67 million relative to the
new estimate of the TR path. The reduction in
the NBR path was a restrictive policy action.
The staff of the Open Market Desk responded to
a reduction in the NBR path by adjusting its plans
for open market operations to hit a lower average
of NBR over the intermeeting period. The Fed
also raised the discount rate from 12 percent to
13 percent, effective February 16, another restric
tive policy action.
Even though the Fed took these restrictive
policy actions over the three weeks ending
February 27, the average level of TR was $272
million above the final estimate of the TR path.
These observations raise an issue about how to
interpret monetary policy actions under the NBR
operating procedure. One view of the conduct
of monetary policy during the three weeks ending
February 27 would be that policy actions were
inconsistent with hitting FOMC targets for mon
etary aggregates because TR was above the TR
path. Interpretation of these actions, however,
must account for the way that the Fed operated
under lagged reserve requirements, which were
in effect during the period of NBR targeting.
Required reserves for each week were determined
by deposit liabilities two weeks earlier. The Fed
operated under the constraint of supplying each
week enough reserves to meet required reserves,
either through open market operations or through
the discount window. For the three weeks ending
February 27, required reserves were based on
deposits over the three weeks ending February 13.
By the time the Fed took policy actions on
February 15, therefore, required reserves for the
three weeks ending February 27 were predeter
mined.
This article evaluates whether policy actions
19 When periods between FOMC meetings were longer than
five weeks, the staff divided the intermeeting periods into
two subperiods for purposes of setting TR paths and project
ing the average levels of TR. The staff divided these inter
meeting periods into subperiods to avoid setting weekly
objectives for NBR just after an FOMC meeting based on
estimates of variables for six or seven weeks into the future.
The staff considered their estimates that far into the future to
be so unreliable that revisions in their estimates over inter
meeting periods could generate unnecessary noise in week
ly objectives for NBR.
Table 1
Policy Actions Under the Nonborrowed Reserves Operating Procedure (amounts in millions of dollars)
FOMC
meeting
Period for
setting total
reserves path
Dates of
projections
and estimates
Projected
total reserves
Total
reserves path
Difference
Changes in the NBR path
between FOMC meetings to
limit the size of deviations
of TR from path
Discount rate
Change in the
federal funds rate,
in basis points
1980
February 4-5
1980
3 weeks ending
February 27
February
7
15
22
27
$ 43,182
43,083
43,311
43,042
-140
$ 43,220
42,770
42,770
42,770
-450
February
March
29
7
14
19
42,915
42,933
43,013
43,005
90
42,289
42,289
42,289
42,289
0
626
644
724
716
March
21
28
4
14
18
23
44,597
44,633
44,458
44,476
44,339
44,336
-261
44,571
44,571
44,771
44,771
44,771
44,771
200
26
62
-313
-295
-432
-435
April
May
25
2
9
16
21
44,543
44,379
44,410
44,377
44,352
-191
45,131
45,181
45,231
45,231
45,231
100
-588
-802
-821
-854
-879
May
23
30
6
13
18
43,821
43,714
43,548
43,592
43,535
-286
43,821
43,714
43,554
43,592
43,592
-229
20
27
7
9
43,299
43,354
43,377
43,509
210
11
23
28
1
8
13
41,602
41,558
41,538
41,512
41,639
41,645
43
Change
3 weeks ending
March 19
Change
March 18
5 weeks ending
April 23
April
Change
April 22
4 weeks ending
May 21
Change
May 20
4 weeks ending
June 18
June
SEPTEMBER/OCTOBER
Change
3 weeks ending
July 9
June
July
Change
July 9
5 weeks ending
August 13
July
August
1994
Change
$
- 38
313
541
272
As of 2/15:$ -67
As of 2/29: $ -300’
Through 2/15: 12%
As of 2/16: 13
February 13
20
27
84
123
-25
As of 3/14:
imposed 3% sur
charge
March
5
12
19
155
28
-21
No change
March
April
26
2
9
16
23
154
161
-35
-69
-79
■P*
CO
As of 5/7: eliminated
3% surcharge
April
May
30
7
14
21
-244
-216
-211
-14
0
0
-6
0
-57
As of 5/28: 12%
As of 6/13: 11%
May
June
28
4
11
18
-125
128
-106
-69
43,299
43,354
43,377
43,377
78
0
0
0
132
No change
June
July
25
2
9
9
33
-15
41,602
41,558
41,505
41,455
41,480
41,480
-122
0
0
33
57
159
165
As of 7/28: 10%
July
16
23
30
6
13
-28
-30
30
62
-75
As of 5/2: $ 100
August
FEDERAL RESERVE BANK OF ST. LOUIS
Table 1(continued)
FOMC
meeting
Period for
setting total
reserves path
Dates of
projections
and estimates
Projected
total reserves
Total
reserves path
Difference
Changes in the NBR path
between FOMC meetings to
limit the size of deviations
of TR from path
Discount rate
Change in the
federal funds rate,
in basis points
1980
1980
August 12
5 weeks ending
September 17
August
September
15
19
22
29
5
12
17
Change
Sept. 16
5 weeks ending
October 22
September
October
4 weeks ending
November 19
October
November
5 weeks ending
December 24
November
December
January
Change
3 weeks ending
February 4
January
February
Change
128
128
282
362
285
380
375
24
31
7
14
19
42,004
41,996
41,639
41,745
41,753
■251
41,795
41,795
41,420
41,445
41,445
-350
209
201
219
300
308
21
25
1
5
39,988
40,224
40,382
40,392
40,381
40,395
40,514
526
39,691
39,821
40,041
40,131
40,171
40,171
40,171
480
297
403
341
261
210
224
343
40,948
40,991
40,971
41,168
41,199
251
40,948
41,048
41,148
41,338
41,338
390
41,740
41,509
41,427
41,520
41,371
-369
42,041
41,841
41,841
41,934
41,934
-107
Change
December
$
382
495
323
442
438
398
12
4 weeks ending
January 14
39,816
40,111
40,111
40,261
40,311
40,311
40,311
495
41,199
41,199
41,199
41,299
41,299
41,299
100
23
24
December
18-19
$
41,581
41,694
41,522
41,741
41,737
41,697
116
Change
Nov. 18
39,944
40,239
40,393
40,623
40,596
40,691
40,686
742
19
26
3
10
17
22
Change
Oct. 21
$
23
29
5
9
14
16
23
30
2
4
No change
August
20
27
3
Sept.
10
17
As of 9/5: $-150
As of 9/26: 11%
Sept.
24
1
8
As Of 10/3: $ -200
50
68
44
-25
42
22
21
153
21
5
9
15
As of 11/17: basic
rate 12%; 2% sur
charge
October
Nov.
29
5
12
19
62
82
66
57
As of 12/5: basic
rate 13%; 3% sur
charge
Nov.
Dec.
26
3
10
17
24
221
29
110
101
-39
0
-57
-177
-170
-139
No change
Dec.
January
31
7
14
-99
161
-42
21
28
February 4
-29
-123
-93
-301
-332
-414
-414
-563
No change
As of 11/7: $-100
As of 11/14: $-50
As of 12/1: $-170
January
■C*
Table 1(continued)
FOMC
meeting
Period for
setting total
reserves path
Dates of
projections
and estimates
Projected
total reserves
Total
reserves path
Difference
Changes in the NBR path
between FOMC meetings to
limit the size of deviations
of TR from path
Discount rate
Change in the
federal funds rate,
in basis points
1981
February 2-3
1981
4 weeks ending
March 4
February
March
6
17
25
27
4
Change
4 weeks ending
April 1
March
4 weeks ending
June 17
March
40,006
40,165
40,132
40,132
40,132
126
0
-33
97
-10
-105
No change
May
1
8
15
20
40,959
40,736
40,683
40,679
-280
40,407
40,362
40,294
40,294
-113
552
374
389
385
May
22
29
5
12
17
40,011
40,104
40,141
40,078
40,069
58
40,011
40,098
40,204
40,138
40,138
127
19
30
6
8
40,464
40,674
40,743
40,879
415
40,643
40,808
40,907
40,907
264
SEPTEMBER/OCTOBER
June
February
As of 2/25: $-166
40,006
40,132
40,229
40,122
40,027
21
July
1994
No change
3
10
20
24
29
April
Change
Change
-169
-327
-351
-484
-390
No change
June
3 weeks ending
July 8
$
-481
-472
-349
-402
-296
Change
May 18
39,796
39,998
39,973
39,973
39,973
177
40,300
40,135
40,010
40,010
40,010
-290
Change
3 weeks ending
May 20
$
39,819
39,663
39,661
39,608
39,714
-105
Change
4 weeks ending
April 29
39,627
39,671
39,622
39,489
39,583
-44
6
13
20
27
1
April
March 31
$
March
April
As of 5/1: $ -2502
As of 5/8: $ -234
April
11
18
25
4
-68
-70
-85
77
11
18
25
1
-20
-140
-65
145
8
15
22
29
50
-10
22
73
As of 5/5: basic rate
14%; 4% surcharge
May
6
13
20
263
-70
68
0
6
-63
-60
-69
No change
May
June
27
3
10
17
-18
-31
93
-23
-179
-134
-164
-28
No change
June
July
24
1
8
10
-36
109
CJl
FEDERAL RESERVE BANK OF ST. LOUIS
Table 1(continued)
FOMC
meeting
Period for
setting total
reserves path
Dates of
projections
and estimates
Projected
total reserves
Total
reserves path
Difference
Changes in the NBR path
between FOMC meetings to
limit the size of deviations
of TR from path
Discount rate
Change in the
federal funds rate,
in basis points
1981
1981
July 6-7
3 weeks ending
July 29
0
32
-8
139
No change
40,782
40,954
40,982
40,982
200
-155
-139
-158
-152
No change
40,510
40,483
40,515
40,535
40,589
79
40,668
40,683
40,833
40,833
40,833
165
-158
-200
-318
-298
-244
No change
18
25
2
7
40,715
40,721
40,847
40,821
106
41,162
41,140
41,226
41,226
64
-447
-419
-379
-405
As of 9/22:
3% surcharge
Oct.
9
20
23
28
40,997
40,812
40,799
40,751
-246
40,997
40,883
40,868
40,868
-129
0
-71
-69
-117
As of 10/12:
2% surcharge
Oct.
Nov.
30
6
13
17
18
40,673
40,661
40,600
40,617
40,662
-11
40,817
40,855
40,754
40,771
40,771
-46
-144
-194
-154
-154
-109
20
30
4
14
18
23
23
41,209
41,277
41,305
41,620
41,488
41,488
41,533
324
41,209
41,252
41,252
41,525
41,389
41,389
41,389
180
0
25
53
95
99
99
144
July
10
17
24
29
$ 41,359
41,136
41,126
41,273
-86
$ 41,359
41,104
41,134
41,134
-225
July
Aug.
31
6
14
19
40,627
40,815
40,824
40,830
203
Aug.
21
28
4
15
16
Change
3 weeks ending
August 19
Change
August 18
4 weeks ending
September 16
Sept.
Change
3 weeks ending
October 7
Sept.
Oct.
Change
October 5-6
3 weeks ending
October 28
Change
3 weeks ending
November 18
Change
November 17 5 weeks ending
December 23
Nov.
Dec.
Change
$
July
15
22
29
-117
29
-51
Aug.
5
12
19
-29
4
-10
Aug.
Sept.
26
2
9
16
-78
-52
-39
-41
23
30
7
-76
-33
46
Oct.
14
21
28
-53
39
-45
Nov.
4
11
18
-8
-78
-84
Nov.
Dec.
25
2
9
-75
6
-44
22
17
Sept.
Oct.
As of 11/6: $56
As of 11/2:
basic rate 13%
As of 11/17:
surcharge eliminated
As of 12/4: 12%
16
23
-C»
CT)
Table 1(continued)
FOMC
meeting
Period for
setting total
reserves path
Oates of
projections
and estimates
Projected
total reserves
Total
reserves path
Difference
Changes in the NBR path
between FOMC meetings to
limit the size of deviations
of TR from path
Discount rate
Change in the
federal funds rate,
in basis points
1982
December
21-22,1981
1982
6 weeks ending
February 3
Dec.
Jan.
Feb.
28
4
8
15
22
29
3
Change
February 1-2
4 weeks ending
March 3
Feb.
SEPTEMBER/OCTOBER
6 weeks ending
June 30
5
12
19
26
31
39,102
39,094
38,988
39,002
39,035
-67
39,376
39,239
39,159
39,159
39,159
-217
-274
-145
-171
-157
-124
April
2
9
16
23
28
39,536
39,537
39,582
39,498
39,474
-62
39,536
39,449
39,414
39,334
39,334
-202
0
88
168
164
140
April
May
30
7
14
19
39,679
39,658
39,786
39,810
131
39,702
39,702
39,821
39,821
119
-23
-44
-35
-11
May
21
28
4
11
18
28
30
39,401
39,409
39,368
39,478
39,487
39,472
39,507
106
39,401
39,385
39,355
39,428
39,373
39,373
39,373
-28
0
24
13
50
114
99
134
June
Change
1994
0
206
324
486
517
614
662
March
Change
May 18
$
0
-95
-81
-116
-40
Change
3 weeks ending
May 19
42,684
42,573
42,536
42,534
42,459
42,351
42,351
-333
41,270
41,309
41,158
41,181
41,181
-89
Change
4 weeks ending
April 28
$
41,270
41,214
41,077
41,065
41,141
-129
Change
March 29-30
42,684
42,779
42,860
43,020
42,976
42,965
43,013
329
5
16
19
26
3
March
4 weeks ending
March 31
$
No change
Dec.
Jan.
30
6
13
20
27
3
11
44
-56
54
102
79
10
17
24
3
42
42
-175
21
March
10
17
24
31
28
54
-41
51
April
7
14
21
28
16
-47
33
-29
May
5
12
19
81
-56
-30
May
June
26
2
9
16
23
30
-97
-27
17
64
-7
64
As of 1/15: $-187
Feb.
No change
Feb.
March
No change
No change
-si
FEDERAL RESERVE BANK OF ST. LOUIS
Table 1(continued)
FOMC
meeting
Period for
setting total
reserves path
Dates of
projections
and estimates
Projected
total reserves
Total
reserves path
Difference
Changes in the NBR path
between FOMC meetings to
limit the size of deviations
of TR from path
Discount rate
Change in the
federal funds rate,
in basis points
1982
1982
June 3 0 July 1
4 weeks ending
July 28
July
2
9
16
23
28
Change
4 weeks ending
Aug. 25
3 weeks ending
Sept. 15
$
-208
-255
-252
-231
-232
Aug.
Sept.
27
3
10
15
39,510
39,609
39,767
39,812
302
39,510
39,573
39.663
39.663
153
0
36
104
149
17
24
1
6
40,227
40,279
40,348
40,386
159
39,933
39.784
39.784
39.784
-149
294
495
564
602
8
15
22
27
40,454
40,579
40,583
40,578
124
40,454
40,598
40.587
40.587
133
0
-19
-4
-9
Sept.
Oct.
Change
1 The three weeks ending March 19,1980, is the second subperiod between FOMC meet
ings on February 4-5 and March 18. The NBR path was reduced by $300 million relative
to the TR path at the beginning of this subperiod to limit the size of the deviation of TR
from path.
2 The three weeks ending May 20 is the second subperiod between FOMC meetings on
March 31 and May 18. The NBR path was reduced by $250 million relative to the TR path
at the beginning of this second subperiod to limit the size of the deviation of TR from path.
0
-84
-97
-83
-109
40.411
40.411
40,391
40.343
40.343
-68
Change
3 weeks ending
Oct. 27
39,978
40,078
40,114
40.085
40.085
107
40,203
40,156
40,139
40,112
40,111
-92
Oct.
Oct. 5
$
30
6
13
20
25
Change
3 weeks ending
Oct. 6
39,978
39,994
40,017
40,002
39,976
-2
July
Aug.
Change
Aug. 24
$
As of 7/20: 11.50%
July
7
14
21
28
-34
-129
-104
-112
As of 8/2: 11%
As of 8/16: 10.50%
Aug.
4
11
18
25
13
-25
-79
-107
Sept.
1
8
15
111
-1
13
22
29
6
4
-19
65
13
20
27
-117
-7
-9
As of 7/16: $ 85
As of 7/30: $100
As of 8/27: 10%
Sept.
As Of 9/24: $ 248
Oct.
As of 10/12: 9.50%
Oct.
00
49
were consistent with use of the NBR procedure
for monetary control by examining the direction
and magnitude of policy actions in relation to the
gaps between the projections of TR and estimates
of the TR path at the time of the policy actions.
From this perspective, policy actions during the
three weeks ending February 27, 1980, were
consistent with use of the NBR operating proce
dure for monetary control.20
As of February 29, the staff projected that
TR would be $626 million above path level in
the second intermeeting period (the three weeks
ending March 19). That day, the Fed reduced its
target for NBR by $300 million relative to the TR
path to limit the size of this deviation of TR from
the path. As a result of that reduction in the NBR
path, banks were forced to obtain more of the
reserves from the discount window to meet their
required reserves. The federal funds rate rose by
155 basis points in the week of this policy action.
Projections later in the period indicated that
the gap between TR and the path level was con
tinuing to grow. On March 14, the Fed imposed
a surcharge of 3 percent on discount window
borrowings by banks with deposits of $500 million
or more that borrowed frequently, as part of
President Carter’s program of credit controls and
monetary restraint.21 During this first intermeeting
period examined in Table 1, the Fed took four
policy actions that were appropriate for monetary
control with TR projected to exceed the path
level: two reductions in the NBR path and two
increases in the discount rate.
The FOMC met again on March 18, four days
after President Carter announced a program of
credit controls and monetary restraint. In support
of the President’s program, the FOMC tightened
20 The last observation for TR over each intermeeting period
reflects the information available to Fed staff as of the end of
the period. For instance, the last estimate of TR for the
three weeks ending February 27, 1980, was the staff esti
mate as of February 27. The data for TR over intermeeting
periods reflect the information available to policymakers at
the time, not subsequent revisions to TR.
21 For more details on the discount rate surcharge, see Board
of Governors (1980, pp. 315-18). For a description of the
credit control program, see Gilbert and Trebing (1981).
22 This article does not include among the policy actions some
adjustments to the supply of NBR which might properly be
classified as policy actions. Levin and Meek (1981) mention
that on some occasions the staff of the Open Market Desk
based open market operations on movements in the federal
funds rate, rather than their numbers on factors affecting
NBR. As they describe those actions, the objective was to
use the federal funds rate as an indicator of errors in their
numbers on factors affecting NBR. They do not indicate that
these open market operations based on movements in the
monetary policy by increasing the borrowings
assumption substantially (Table 4). With given
objectives for growth of the monetary aggregates,
a larger borrowings assumption implies a lower
NBR path and, therefore, a more restrictive mon
etary policy.
As of the beginning of the period after the
March FOMC meeting (that is, the five weeks
ending April 23, 1980), TR was projected to be
approximately equal to the TR path. Later in that
period, the projection of TR was reduced and the
TR path increased, producing a widening gap
between projected TR and the path level. The Fed,
however, took no policy actions to limit the size
of that gap. The actual level of TR ended up
$435 million below the final estimate of the
TR path.
G eneral Patterns in Policy Actions
Examination of policy actions in Table 1 for
the entire period from February 1980 through
October 1982 indicates several patterns:22
Variable Pattern in the Use of Policy Tools —
For given staff projections and estimates of TR,
policy actions were highly variable. As noted for
periods examined above, widening gaps between
projections of TR and path levels induced prompt
and substantial adjustments of policy tools in
some periods but not in other periods. To iden
tify relevant periods when the Fed did not take
policy actions, it is necessary to specify a criterion
for identifying relatively large deviations of TR
from the TR path. This paper uses $200 million
or more as the size of a large deviation, based on
the following reasoning. Over the period of NBR
targeting, TR was approximately $40 billion. A
gap of $200 million is one-half of 1 percent of
federal funds rate interfered with hitting targets for NBR over
intermeeting periods.
Other adjustments to the supply of NBR raise more ques
tions about adjustments to the supply of NBR that should be
labeled as policy actions. At times, the staff adjusted the
supply of NBR to prevent large movements in borrowings
and in the federal funds rate just prior to FOMC meetings.
Weekly Reports on Open Market Operations mention that at
times the staff did not make the full adjustments to the TR
path that were indicated by their information on factors
affecting the relationship between reserves and the mone
tary aggregates, and the reports refer to occasions when the
staff deliberately allowed NBR to deviate from its path level,
to avoid forcing large changes in borrowed reserves just
before FOMC meetings. Table 1 limits its list of policy
actions to those identified clearly as policy actions in the
Report on Open Market Operations.
SEPTEMBER/OCTOBER 1994
50
$40 billion. An error of approximately one-half
of 1 percent in hitting a target for an aggregate
over a month, compounded over a year, would
be an error of 6 percent, which could be inter
preted as a substantial error. TR deviated from
the TR path by at least $200 million, and the Fed
took no policy actions in response, in each of the
periods after the FOMC meetings on March 18,
1980, and December 18-19, 1980.
Directions of Policy Actions Were Appropriate
for Monetary Control — Prior to the fall of 1982,
the direction of each policy action between FOMC
meetings was appropriate for monetary control.
When TR was projected to be above the path level,
policy actions included reductions in the target
for NBR relative to the TR path or increases in the
discount rate. The Fed took the opposite types
of policy actions when TR was projected to be
below the path level.23
The only exception to this pattern occurred
on February 25, 1981. The Fed reduced the NBR
path by $166 million when the staff projected TR
to be $351 million below the TR path. At that
time, the growth of M2 and M3 exceeded FOMC
objectives, whereas M l was growing more slowly
than the target set by the FOMC at its meeting on
February 2-3, 1981. TR was below the TR path
because required reserves predominately reflected
the required reserves on deposits in M l. In
February 1981, the FOMC decided to put more
weight on its objectives for M2 and M3 than on
M l. Therefore, the FOMC decided to reduce the
supply of NBR to limit the growth of M2 and M3.
This reduction in the NBR path on February 25,
1981, was consistent with use of the NBR proce
dure for monetary targeting, even though TR was
projected to be below the path at the time of the
policy action.
The change in the NBR target on September 24,
1982, in contrast, illustrates a policy action that
was inconsistent with use of the NBR operating
procedure for monetary control. It is generally
23 Some changes in the gap between the NBR path and the
TR path were labeled “technical adjustments” to the supply
of NBR, not policy actions. The purpose of these technical
adjustments was to offset the effects on interest rates of
changes in the relationship between borrowings and the
spread between the federal funds rate and the discount rate
for TR. At times, the staff concluded that there were persis
tent changes in the quantity of reserves borrowed by banks
for given spreads between the federal funds rate and the
discount rate. In terms of Figures 1 and 4, there appeared
to be shifts in the slope of the supply curve of reserves. At
those times, the staff adjusted the supply of NBR to offset
possible effects on interest rates of such changes in the
FEDERAL RESERVE BANK OF ST. LOUIS
recognized that by the fall of 1982, the Fed had
abandoned use of the NBR operating procedure
in favor of smoothing short-term interest rates.24
For operational purposes, however, the staff
continued to calculate the numbers that had been
important for conducting policy under the NBR
procedure. After the FOMC meeting on August
24,1982, projections of TR were increased gradu
ally relative to estimates of the TR path, and by
September 24, the gap had reached $495 million.
A policy action appropriate for monetary targeting
would have been a reduction in NBR. Instead,
the Fed in creased the target for NBR, to limit the
rise in short-term interest rates in response to the
rise in demand for reserves. This action, the kind
of policy action illustrated in Figure 6, provides
one way to date the end of the NBR operating
procedure.
Size of the Policy Actions — Table 2 lists the
changes in the NBR path between FOMC meetings
that the Fed classified as policy actions. These
changes in the NBR path generally were about
half or less of the gap between TR projected by
the staff and the TR path at the time of the policy
actions. These observations indicate that even
at those times when the Fed adjusted the NBR
path as a policy action, the Fed was willing to
tolerate large deviations of TR from the path over
intermeeting periods. The emphasis in the policy
was bringing the levels of the monetary aggregates
closer to FOMC objectives over time. The policy
did not call for actions to force immediate shifts
of the levels of the aggregates back to the levels
specified in FOMC directives.
Policy Actions Did Not Cause All of the Sharp
Fluctuations in Interest Rates — The federal
funds rate was more variable during the period
of NBR targeting than in surrounding periods
(Figure 1). These large fluctuations generated a
lot of complaints from market participants and
from economists critical of the procedure. In
evaluating NBR targeting as a method of imple
menting monetary policy, it would be useful to
behavior of banks. Table 1 does not include these adjust
ments to the supply of NBR because the purpose of this arti
cle is to examine patterns of policy actions under the NBR
operating procedure. Reports by the Manager of the Open
Market Account distinguish between technical adjustments
and changes in the supply of NBR labeled policy actions.
24 See Thornton (1983, 1988).
51
Table 2
Size of Changes in the Nonborrowed
Reserves Path
Date
Change in
the NBR path
(millions of dollars)
Change in the NBR
path as a percentage
of the most current
staff projection of
the gap between
TR and the TR path
1980
-6 7
21.4%
2/29
-300
47.9
5/2
+100
12.5
9/5
-150
52.6
10/3
-200
61.9
11/7
-100
45.7
2/15
$
11/14
-5 0
16.7
12/1
-170
49.9
1981
-166
N/A1
5/1
-250
45.3%
5/8
-234
62.6
11/6
+ 56
28.9
2/25
$
1982
-187
38.5%
7/16
+ 85
87.6
7/30
+ 100
48.1
1/15
$
1 The NBR path reduced at a time when TR were
projected to be below the TR path.
know whether the relatively large fluctuations
in interest rates under NBR targeting reflected
frequent, aggressive policy actions to hit short-run
money targets. Perhaps fluctuations in the federal
funds rate under a NBR targeting procedure would
be substantially smaller than the experience of
1980-82 if the Fed used the procedure less aggres
sively in attempting to hit short-run money targets.
In contrast, many of the relatively large weekly
changes in the federal funds rate may have
25 Cook (1989a, 1989b) conducted a similar analysis of the
timing of policy actions and changes in the federal funds rate
during the period of NBR targeting. Cook investigated the
degree to which changes in the federal funds rate over peri
ods between FOMC meetings could be explained in terms of
policy actions. Cook concluded that roughly two-thirds of
occurred simply because the Fed placed less
weight on limiting interest rate fluctuations
under the NBR operating procedure than other
operating procedures.
It is possible to determine whether the rela
tively large weekly fluctuations in the federal
funds rate reflected the effects of policy actions by
examining their timing and the timing of policy
actions.25 Table 3 examines the pattern of policy
actions during the weeks in which the federal
funds rate changed by 100 basis points or more.
Changes in weekly average levels of the federal
funds rate of 100 basis points or more were rela
tively common during the three years ending
in September 1982. For example, Table 3 list
29 weekly occurrences. During the three years
ending in September 1979, in contrast, there were
no weeks when the federal funds rate changed
by as much as 100 basis points. During the three
years ending in September 1985, the three years
following the period of NBR targeting, the federal
funds rate changed by 100 basis points or more
in only five weeks.
Seven of the relatively large changes in the
federal funds rate in Table 3 occurred in the weeks
just after FOMC meetings. For instance, the fed
eral funds rate rose 154 basis points in the week
ending March 26, 1980, the first week after the
FOMC meeting on March 18. The decisions
of the FOMC at its meeting on March 18, 1980,
can be characterized as a tightening of monetary
policy. Table 4 illustrates the shift in monetary
policy at the FOMC meeting on March 18 in terms
of an increase in the borrowings assumption
relative to the level set at the prior meeting:
from a level of $1.25 billion set at the meeting
on February 4-5 to a level of $2.75 billion set
on March 18. The rise in the federal funds rate
in the week ending March 26 is consistent with
a tightening of monetary policy at the FOMC
meeting on March 18.
The federal funds rate fell by 244 basis points
in the week ending April 30,1980, which was the
first week after the FOMC meeting on April 22.
At its meeting on April 22, the FOMC decided to
reverse the tightening of monetary policy at its
prior meeting. Table 4 illustrates the easing of
monetary policy at the meeting of April 22 with
the changes in the federal funds rate were due to judgmen
tal actions of the Federal Reserve. This article, in contrast,
examines the timing of relatively large weekly changes in the
federal funds rate and policy actions.
SEPTEMBER/OCTOBER 1994
52
Table 3
Association Between Weekly Changes in the Federal Funds Rate
of 100 Basis Points or More and Policy Actions
Week
ending
Change in the
federal funds rate
from the prior week,
in basis points
Change in the
NBR target
Change in the
discount rate
or surcharge
First week after an
FOMC meeting
X indicates occurrence in the week
1980
2/20
+123
X
X
3/5
+155
3/26
+154
4/2
+161
4/30
-244
5/7
-216
5/14
-211
5/28
-125
6/4
+128
6/11
-106
10/1
+153
11/26
+221
12/10
+110
12/17
+101
X
X
X
X
X
X
X
X
X
X
1981
1/7
+161
1/28
-123
3/18
-140
4/1
+145
5/6
+263
7/8
+109
7/15
-117
X
X
X
1982
1/27
+102
2/24
-175
7/14
-129
7/21
-104
7/28
-112
8/25
-107
9/1
+111
X
X
10/13
-117
X
X
FEDERAL RESERVE BANK OF ST. LOUIS
X
X
53
Table 4
Initial Assumptions for Borrowed
Reserves Set by the FOMC, 1980-82
Date of FOMC meeting
Initial assumption for
borrowed reserves
(millions of dollars)
1980
January 8-9
$
1,000
February 4-5
1,250
March 18
2,750
April 22
1,375
May 20
100
July 9
75
August 12
75
September 16
750
October 21
1,300
November 18
1,500
December 18-19
1,500
1981
February 2-3
$
1,150
May 18
2,100
July 6-7
1,500
August 18
1,400
October 5-6
850
November 17
400
December 21-22
300
1982
March 29-30
$
1,500
1,150
May 18
800
June 30-July 1
800
August 24
350
October 5
300
November 16
250
December 20-21
200
the decline in the initial borrowings assumption
to $1,375 billion.
Comparison of Tables 3 and 4 illustrates this
consistent pattern: On those occasions when
the federal funds rate changed by over 100 basis
points in the first week after an FOMC meeting,
increases in the federal funds rate coincided
with increases in the initial borrowings assump
Of the 29 weeks in Table 3 in which the federal
funds rate changed by 100 basis points or more,
15 were not the first week after an FOMC meeting
or weeks of changes in the NBR path or the dis
count rate. Many of the relatively large weekly
changes in the federal funds rate, therefore,
reflected the relatively low weight the Fed attached
to limiting fluctuations in the federal funds rate
under the NBR operating procedure. Also, the
economy was very volatile during the period
of NBR targeting. Influences other than the con
duct of monetary policy may have contributed
substantially to the variability of interest rates
over this period.
1,300
March 31
February 1-2
tions at the FOMC meetings, and relatively large
decreases in the federal funds rates were associ
ated with reductions in the initial borrowings
assumptions. This pattern prevailed until the fall
of 1982, when the Fed had largely abandoned use
of NBR targeting. Thus, some of the relatively
large changes in the federal funds rate reflected
policy actions initiated at the time of FOMC
meetings.
CONCLUSIONS
The conduct of monetary policy in the United
States from October 1979 through the fall of 1982
has important implications for the design of pro
cedures for targeting monetary aggregates today.
This is the only period in which daily open market
operations were tied directly to objectives of the
FOMC for growth of the monetary aggregates. It
is our closest approximation to short-run monetary
control in the United States. Some critics of the
conduct of monetary policy in this period have
concluded that errors in hitting the money targets
of the FOMC reflected problems inherent in the
design of the procedure.
This article presents information on the
conduct of monetary policy in this period of
nonborrowed reserves (NBR) targeting not avail
able in other published studies. This information
includes Fed staff projections of the actual levels
of total reserves (TR) over periods between FOMC
meetings and staff estimates of the average levels
of TR between meetings that would have been
consistent with FOMC objectives for money
growth (the TR paths). Using this information,
we can examine the timing and size of policy
actions in relation to the information available
to Fed policymakers at the time.
Examination of policy actions during the
period of NBR targeting yields the following
SEPTEMBER/OCTOBER 1994
54
observations. First, the pattern of policy actions
does not reflect consistent use of the procedure
over time for monetary targeting. During some
intermeeting periods in which the staff projected
that TR would deviate substantially from the TR
path, the Fed took no policy actions, whereas
in other periods the Fed took aggressive actions
consistent with monetary targeting. Second,
when the Fed did take policy actions, they were
in the directions appropriate for monetary control,
given the staff projections and estimates available
at the time. This observation contradicts asser
tions that there was no change in the operating
procedure in October 1979. Third, the magnitude
of policy actions often was small in relation to
the gap between the projection of TR and the path.
These three observations have implications for
interpreting the three years ending in the fall of
1982 as an experiment in monetary targeting. The
commitment of policymakers to hitting short-run
money targets varied over those three years. Any
conclusions derived from data for those three
years concerning NBR targeting as a method of
monetary control should account for variation
over time in the commitment of policymakers to
take actions appropriate for monetary control.
______ . “Overview of Findings and Evaluation,” New
Monetary Control Procedures, Federal Reserve Staff Study,
vol. I (February 1981).
The fourth observation concerns the degree of
interest rate variability under a procedure of NBR
targeting. While several of the relatively large
weekly changes in the federal funds rate coincided
with the timing of policy actions, the Fed took no
policy actions at the time of some relatively large
fluctuations in the federal funds rate. Interest rate
fluctuations during the period of NBR targeting
reflect use of an operating procedure which left
the federal funds rate largely unconstrained within
wide bands. It is difficult to extrapolate from this
experience to the degree of weekly interest rate
variability that would exist under use of an NBR
procedure now. This experience, however, is
consistent with the view that targeting NBR for
purposes of short-run monetary control would
tend to increase weekly interest rate variability.
Goodfriend, Marvin. “Discount Window Borrowing, Monetary
Policy, and the Post-October 6, 1979 Federal Reserve
Operating Procedure,” Journal o f Monetary Economics
(September 1983), pp. 343-56.
REFERENCES
Avery, Robert B., and Myron L. Kwast. “Money and Interest
Rates Under a Reserves Operating Target,” Federal
Reserve Bank of Cleveland Economic Review (second
quarter 1993), pp. 24-34.
Board of Governors of the Federal Reserve System.
“Announcements,” Federal Reserve Bulletin (April 1980),
pp. 314-24.
______ . “Record of Policy Actions of the Federal Open Market
Committee,” Federal Resen/e Bulletin (December 1979),
pp. 972-8.
Cook, Timothy. “Determinants of the Federal Funds Rate:
1979-82,” Federal Reserve Bank of Richmond Economic
Review (January/February 1989a), pp. 3-19.
______ . “Determinants of the Federal Funds Rate: 1979-1982,”
Federal Reserve Bank of Richmond Working Paper 88-7
(March 1989b).
Feinman, Joshua. “An Analysis of the Federal Reserve’s
Nonborrowed Reserves Operating Procedure.” Ph.D.
dissertation. Brown University, May 1988.
Friedman, Milton. “Lessons from the 1979-82 Monetary Policy
Experiment,” The American Economic Review (May 1984),
pp. 397-400.
Gilbert, R. Alton. “Operating Procedures for Conducting
Monetary Policy,” this Review (February 1985), pp. 13-21.
______ , and Michael E. Trebing. “The New System of
Contemporaneous Reserve Requirements,” this Review
(December 1982), pp. 3-7.
______ , a n d ______ . ‘The FOMC in 1980: A Year of Reserve
Targeting,” this Review (August/September 1981), pp. 2-22.
______ , and David H. Small, eds. Operating Procedures and
the Conduct o f Monetary Policy: Conference Proceedings,
Finance and Economics Discussion Series, Board of
Governors of the Federal Reserve System (March 1993).
Hetzel, Robert L. “Monetary Policy in the Early 1980s,”
Federal Resen/e Bank of Richmond Economic Review
(March/April 1986), pp. 20-32.
______ . “The October 1979 Regime of Monetary Control and
the Behavior of the Money Supply in 1980,” Journal of
Money, Credit and Banking (May 1982), pp. 234-51.
Hoehn, James G. “Recent Monetary Control Procedures and
Response of Interest Rates to Fluctuations in Money
Growth,” Federal Reserve Bank of Dallas Economic Review
(September 1983), pp. 1-10.
Levin, Fred, and Paul Meek. “Implementing the New
Procedures: The View from the Trading Desk,” New
Monetary Control Procedures, Federal Reserve Staff Study,
vol. I (February 1981).
Axilrod, Stephen H. “U.S. Monetary Policy in Recent Years:
An Overview,” Federal Reserve Bulletin (January 1985),
pp. 14-24.
Lindsey, David E. “Nonborrowed Reserve Targeting and
Monetary Control,” in Laurence H. Meyer, ed., Improving
Money Stock Control: Problems, Solutions, and
Consequences, Economic Policy Conference Series, co
sponsored by the Center for the Study of American Business
at Washington University and the Federal Reserve Bank of
St. Louis. Kluwer-Nijhoff, 1983, pp. 3-41.
______ . “Monetary Policy, Money Supply, and the Federal
Reserve’s Operating Procedures,” Federal Resen/e Bulletin
(January 1982), pp. 13-24.
______ . “Recent Monetary Developments and Controversies,”
Brookings Papers on Economic Activity (January 1982),
pp. 245-68.
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
55
______ , and others. “Short-Run Monetary Control: Evidence
Under a Non-Borrowed Reserve Operating Procedure,”
Journal o f Monetary Economics (January 1984), pp. 87-111.
McCallum, Bennett. T. “On Consequences and Criticisms of
Monetary Targeting," Journal o f Money, Credit and Banking
(November 1985, part 2), pp. 570-97.
Meek, Paul. U.S. Monetary Policy and Financial Markets,
Federal Reserve Bank of New York, 1982.
Pearce, Douglas K. “Discount Window Borrowing and Federal
Reserve Operating Regimes,” Economic Inquiry (October
1993), pp. 564-79.
Pierce, David A. ‘Trend and Noise in the Monetary Aggregates,”
New Monetary Control Procedures, Federal Reserve Staff
Study, vol. II (February 1981).
Pierce, James L. “Did Financial Innovation Hurt the Great
Monetarist Experiment?” The American Economic Review
(May 1984), pp. 392-6.
Poole, William. “Federal Reserve Operating Procedures:
A Survey and Evaluation of the Historical Record Since
October 1979,” Journal of Money, Credit and Banking
(November 1982, part 2), pp. 575-96.
Spindt, Paul A., and Vefa Tarhan. 'The Federal Reserve’s New
Operating Procedures: A Post Mortem,” Journal o f Monetary
Economics (January 1987), pp. 107-23.
Stevens, E. J. ‘The New Procedure,” Federal Reserve Bank of
Cleveland Economic Review (summer 1981), pp. 1-17.
Thornton, Daniel L. ‘The Borrowed-Reserves Operating
Procedure: Theory and Evidence,” this Review
(January/February 1988), pp. 30-54.
______ . “The FOMC in 1982: De-emphasizing M1,” this
Review (June/July 1983), pp. 26-35.
______ . ‘T he FOMC in 1981: Monetary Control in a Changing
Financial Environment,” this Review (April 1982), pp. 3-22.
SEPTEMBER/OCTOBER 1994
56
Appendix 1
Illustration of Staff Projections and Estimates
of Total Reserves
This appendix describes the steps involved in
staff estimates of the TR path and projections of
TR for the intermeeting period after the FOMC
meeting on February 4-5, 1980. The staff divided
the intermeeting period into two subperiods of
three weeks each, ending on February 27 and
March 18. They made such divisions when the
periods between meetings were longer than five
weeks to avoid using projections of variables
several weeks into the future in determining the
supply of NBR early in an intermeeting period.
To aid in clarifying the timing of relationships
between deposits and reserves, Table A l presents
a calendar of January and February 1980. At its
meeting on February 4-5, the FOMC specified
its short-run objectives as growth of M l-B at a
5 percent rate and M2 at a 6.5 percent rate over
the first quarter of 1980. To estimate the TR path
for the three weeks ending February 27, the staff
would do the following calculations:
1. Project the weekly levels of M l and M2 growing
at the desired rates from mid-December through
the three weeks ending February 13. Deposits
over the three weeks ending February 13 deter
mine required reserves over the three weeks
ending February 27. These weekly levels
are projected from the seasonally adjusted
data for December and then converted into
n o n se a so n a lly a d ju sted le v e ls u sin g th e sea
so n a l fa cto rs for th o se w eeks.
2. Estimate currency in the hands of the public,
not seasonally adjusted, "for the three weeks
ending February 13.
3. Subtract the estimate of currency in the hands
of the public from the projection of M l to
derive the level of checkable deposits, not
seasonally adjusted, if M l grew at the rate
desired by the FOMC.
4. Multiply the average level of checkable deposits
as derived in step 3 by an estimate of the average
reserve requirement on checkable deposits.
5. Subtract the estimate of average currency
holdings as described in step 2 and checkable
deposits as described in step 3 from the pro
jection of M2, as described in step 1.
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
Table A1
Calendar of January and
February 1980
January
S
M
T
W
Th
F
S
6
7
14
21
28
1
8
15
22
29
2
9
16
23
30
3
10
17
24
31
4
11
18
25
5
12
19
26
13
20
27
February
S
M
T
W
Th
F
S
3
10
17
24
4
11
18
25
5
12
19
26
6
13
20
27
7
14
21
28
1
8
15
22
29
2
9
16
23
Multiply by an estimate of the average reserve
requirement on deposits in M2 but not in M l.
6. Sum estimates of required reserves as described
in steps 4 and 5 and an estimate of required
reserves on deposits not in M2 to derive an
estimate of what required reserves would be
in the three weeks ending February 27 if M l
and M2 grew at the rates specified by the
FOMC at its meeting on February 4-5. Add
an estimate of the average level of excess
reserves for the three weeks ending February
27 to get an estimate of the TR path over the
three weeks ending February 27.
The steps involved in projecting TR are similar
to the steps in estimating the TR path:
1. Estimate liabilities subject to reserve require
ments for the three weeks ending February 13,
not seasonally adjusted. The Federal Reserve
57
staff generally had data on reservable liabilities
eight days after the end of a reserve mainte
nance week. By February 7, the date of the first
projection, the staff would have had information
on reservable liabilities for the week ending
January 30. They would have to estimate lia
bilities for the weeks ending February 6 and 13.
2. Estimate average reserve requirements on
various categories of liabilities.
3. Sum the projections for required reserves for
the three weeks ending February 27, based on
calculations described in steps 1 and 2, and
add an estimate of average excess reserves.
Appendix 2
A Tool for Describing the Conduct of Monetary Policy:
Supply and Demand for Reserves
This paper describes the conduct of monetary
policy under the NBR operating procedure using
diagrams of the supply and demand for bank
reserves.' This appendix describes the determi
nants of the supply and demand curves, and the
following section uses this analytical tool to
describe the mechanics of the NBR operating
procedure.
Reserves available to meet reserve require
ments include currency that banks hold in their
vaults and their reserve balances at Federal
Reserve Banks. The Federal Reserve supplies
reserves. Banks demand reserves to facilitate
their customers’ transactions and to meet
reserve requirements imposed by the Federal
Reserve, which are based on the amount and
composition of their liabilities.
Banks earn no interest on reserves. This
article identifies the opportunity cost to banks
of holding reserves as the federal funds rate,
which is the interest rate that banks charge each
other for lending reserves.2 A bank changes its
reserves by borrowing or lending at the federal
funds rate.
Demand for reserves by banks is drawn as a
function of the federal funds rate in Figures 4-6.
Reserve requirements on deposits included in
the money stock create a close relationship
1 For convenience of exposition, the term “bank" refers to all
depository institutions.
2 Federal funds brokers facilitate the operation of the federal
funds market. These brokers receive orders from depository
institutions located throughout the nation to lend or borrow
reserves, and the brokers match lenders and borrowers at
mutually agreeable interest rates. Most of the transactions
through the federal funds market involve borrowing and
between the demand for money by the public
and the demand for reserves by banks. Demand
for reserves, therefore, depends on reserve
requirements and the demand for money.
Demand for money is assumed to be a function
of total spending in the economy and interest
rates. Various influences can cause shifts in the
demand curve for reserves. A change in total
spending in the economy, which influences the
demand for money, would cause the demand
curve for reserves to shift. Shifts in the demand
for reserves could reflect other influences: changes
in the random component of money demand; the
average reserve requirement on deposit liabilities
included in the money stock; reserve require
ments on other liabilities; or the demand for
excess reserves.
Elasticity of the demand for reserves depends
on the relevant time period over which average
reserves are measured. The demand curves for
reserves in Figures 4-6 are steeply sloped because
it is for a period between FOMC meetings. Over
these periods, there is little time for a change in
interest rates to change the quantity of money
demanded, feeding back to a change in the
quantity of reserves demanded.
Factors that influence the supply of reserves
can be analyzed by considering separately the
lending reserves for one day. The transfers of reserves to
borrowers are made the same day through wire transfer sys
tems, including the Fed Wire of the Federal Reserve
System.
SEPTEMBER/OCTOBER 1994
58
determinants of borrowed reserves and NBR.
The Federal Reserve determines the amount of
NBR directly through the open market opera
tions. Banks decide the amount of reserves they
borrow from the Federal Reserve, but their deci
sions are shaped by lending terms set by the
Federal Reserve, including the discount rate and
limits on the size and frequency of borrowings
by individual banks. Banks try to avoid exceeding
these borrowing limits to ensure that they main
tain access to credit from the Fed to cover their
short-term liquidity requirements. If a bank
borrows now, it will be subjected to greater
administrative pressure to limit its borrowings
in the future, when the attractiveness of borrowing
from the discount window might be greater.
3 Goodfriend (1983) derives the relationship between borrow
ings and the rate spread from a theoretical framework that is
based on profit-maximizing bank behavior.
Digitized forFEDERAL
FRASER RESERVE BANK OF ST. LOUIS
The supply curve for reserves in Figure 4 is
drawn as a vertical line from the level of NBR
(labeled N) up to the level on the vertical axis at
which the federal funds rate equals the discount
rate (rd). If the discount rate is above the federal
funds rate, the amount of reserves borrowed from
Federal Reserve Banks tends to be relatively low
and insensitive to small changes in the federal
funds rate. The supply curve of reserves is
upward sloping in the range with the federal
funds rate above the discount rate. Given the
terms for lending set by the Federal Reserve, it
takes an increase in the spread between the fed
eral funds rate and the discount rate to induce
banks to increase their borrowings from the
discount window.3
59
FEDERAL RESERVE BANK OF ST. LOUIS
WORKING PAPERS SERIES
Working papers from the Federal Reserve Bank of St. Louis reflect preliminary results of staff research and are made
available to encourage comment and discussion, as well as invite suggestions from other researchers for revision.
The views expressed in the working papers are those of the individual authors and do not necessarily reflect official
positions of the Federal Reserve Bank of St. Louis, the Federal Reserve System, or the Board of Governors.
Once a working paper appears in publication, it is removed from the Working Papers Series and is no longer available
for distribution through the Bank. For more information on working papers not available for distribution through the
Bank (denoted by an asterisk), please refer to the source or author indicated. Titles listed as forthcoming will continue
to be available until publication.
Single copies of working papers (those without an asterisk) are available by writing to:
Federal Reserve Bank of St. Louis
Research Department
P.O. Box 442
St. Louis, MO 63166-0442
1994 WORKING PAPERS
94-001A - Peter S. Yoo, “The Baby Boom and Economic Growth.”
94-002A - Peter S. Yoo, “Age Distributions and Returns of Financial Assets.”
94-003A - Peter S. Yoo, “Age-Dependent Portfolio Selection.”
94-004A - Joseph A. Ritter, “The Transition From Barter to Fiat Money.” FORTHCOMING: American
Economic Review.
94-005A - Sangkyun Park, “The Bank Capital Requirement and Information Asymmetry.”
94-006A*- Richard G. Anderson and Kenneth A. Kavajecz, “A Historical Perspective on the Federal Reserves
Monetary Aggregates: A Timeline.” PUBLISHED: this Review (March/April 1994).
Kenneth A. Kavajecz, “The Evolution of the Federal Reserves Monetary Aggregates: A Timeline."
94-007A - Richard G. Anderson and William G. Dewald, “Replication and Scientific Standards in Economics a
Decade Later: The Impact of the JMCB Project.” FORTHCOMING: this Review.
94-008A - Christopher J. Neely, ‘Target Zones and Conditional Volatility: An ARCH Application to the EMS.”
94-009A - Christopher J. Neely, Dean Corbae, and Paul Weller, “The Distribution of Target Zone Exchange Rates
Linder Alternative Realignment Rules.”
94-010A - Christopher J. Neely, “A Reconsideration of the Properties of the Generalized Method of Moments in
Asset Pricing Models.”
94-011A - James B. Bullard and John Keating, “Superneutrality in Postwar Economies.”
94-012A - James B. Bullard and Steven H. Russell, “Monetary Steady States in a Low Real Interest Rate Economy.”
94-013A - James B. Bullard and John Duffy, “Learning in a Large Square Economy.”
94-014B - James B. Bullard anti John Duffy, “A Model of Learning and Emulation with Artificial Adaptive Agents.”
94-015A - Michael J. Dueker, “Mean Reversion in Stock Market Volatility.”
94-016A - Michael J. Dueker, “Compound Volatility Processes in EMS Exchange Rates.”
94-017A - Michael J. Dueker and Daniel J. Thornton, “Asymmetry in the Prime Rate and Firms’ Preference
for Internal Finance.”
SEPTEMBER/OCTOBER 1994
60
FEDERAL RESERVE BANK OF ST. LOUIS
WORKING PAPERS SERIES (continued)
94-018A - John A. Tatom and Dieter Proske, “Are There Adverse Real Effects from Monetary Policy Coordination?
Some Evidence from Austria, Belgium and the Netherlands.”
94-019A - Michael R. Pakko, “Reconciling International Risk Sharing with Low Cross-Country Consumption
Correlations.”
94-020B - Christopher J. Neely, “Realignments of Target Zone Exchange Rate Systems:
What Do We Know?”
94-021A - David C. Wheelock and Paul W. Wilson, “Productivity Changes in U.S. Banking: 1984-93.”
94-022A - Alison Butler and Michael Dueker, “Product Cycles, Innovation and Relative Wages in European Countries.”
94-023A - Sangkyun Park, “Market Discipline By Depositors: Evidence from Reduced Form Equations.”
94-024A - Byung Chan Ahn, “Monetary Policy and the Determination of the Interest Rate and Exchange Rate in a
Small Open Economy With Increasing Capital Mobility.”
94-025A - Sangkyun Park, “Banking and Deposit Insurance As a Risk-Transfer Mechanism.”
94-026A - Michael R. Pakko, “Characterizing Cross-Country Consumption Correlations.”
94-027A - Michael Dueker and Richard Startz, “Maximum-Likelihood Estimation of Fractional Cointegration With An
Application to the Short End of the Yield Curve.”
94-028A - James Bullard and John Duffy, “Using Genetic Algorithms to Model the Evolution of Hetrogeneous Beliefs.”
1993 WORKING PAPERS
93-001A - Patricia S. Pollard, “Macroeconomic Policy Effects in a Monetary Union.”
93-002A - David C. Wheelock and Paul W. Wilson, “Explaining Bank Failures: Deposit Insurance Regulation, and
Efficiency.” FORTHCOMING: The Review of Economics and Statistics.
1992 WORKING PAPERS
92-001 A*- John A. Tatom, “The P-Star Model and Austrian Prices.” PUBLISHED: Empirica, 1992, vol. 19, no. 1.
92-002A - David C. Wheelock and Subal C. Kumbhaker, ‘The Slack Banker Dances: Insurance and Risk-Taking
in the Banking Collapse of the 1920s.” PUBLISHED: Explorations in Economic History (July 1994),
vol. 31, no. 3.
92-003A - Daniel L. Thornton, “The Market’s Reaction to Discount Changes: What’s behind the
Announcement Effect?”
92-004A - Daniel L. Thornton, “Why Do T-Bill Rates React to Discount Rate Changes? FORTHCOMING:
Journal of Money, Credit and Banking.
92-005A*- John A. Tatom, Heinz Gluck, and Dieter Proske, “Monetary and Exchange Rate Policy in Austria: An Early
Example of Policy Coordination. PUBLISHED: Economic Policy Coordination in an Integrating Europe,
Bank of Finland, 1992.
92-006A - John A. Tatom, “Currency Appreciation and ‘Deindustrialization’: A European Perspective.”
92-007A*- David C. Wheelock, “Government Policy and Banking Instability: ‘Overbanking’ in the 1920s.” PUBLISHED:
Journal of Economic History (December 1993), vol. 53, no. 4. (Published as “Government Policy and
Banking Market Structure in the 1920s”).
92-008A - Michael T. Belongia and Dallas S. Batten, “Selecting an Intermediate Target Variable for Monetary Policy
When the Goal is Price Stability.”
FEDERAL RESERVE BANK OF ST. LOUIS
@FedFRASER