Full text of Economic Perspectives (Federal Reserve Bank of Chicago) : September/October 1986, Volume X, Issue 5

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

fo O

a

j; economic

PERSPECTIVES




A m&ynmy ffessm

W hat is the n atu ral rate
o f unem p lo ym en t?
A n n u al co n fe re n c e a sse sse s
b a n kin g risk
U n insu red d e p o sits as a so u rce o f
m arket d isc ip lin e : So m e new
e vid e n ce

ECONOMIC PERSPECTIVES
September/October 1986

Volume X, Issue 5

Karl A. Scheld, senior vice president and
director of research
Edward G. Nash, editor
Anne Weaver, administrative coordinator
Gloria Hull, editorial assistant
Roger Thryselius, graphics
Nancy Ahlstrom, typesetting
Rita Molloy, typesetter
Economic Perspectives is
published by the Research Depart­
ment of the Federal Reserve Bank
of Chicago. The views expressed are
the authors’ and do not necessarily
reflect the views of the management
of the Federal Reserve Bank.
Single-copy subscriptions are
available free of charge. Please send
requests for single- and multiplecopy subscriptions, back issues, and
address changes to Public Informa­
tion Center, Federal Reserve Bank
of Chicago, P.O. Box 834, Chicago,
Illinois 60690, or telephone (312)
322-5111.
Articles may be reprinted pro­
vided source is credited and The
Public Information Center is pro­
vided with a copy of the published
material.

ISSN 0164-0682




Contents
What is the natural rate
of unemployment?

Ellen R. Rissman
A policy response to high unemployment
may be wrong, if the policymakers
do not know what the natural rate
of unemployment is.

Annual conference assesses
banking risk

Richard D. Simmons
Bankers, academics, regulators, and
others gather at the 22nd Conference
on Bank Structure and Competition.

Uninsured deposits as a source of
market discipline: Some new
evidence

Herbert Baer and Elijah Brewer
The perception of risk leads to bank runs
by uninsured depositors, but it also
generates market pressure on
banks to take less risk.

W hat is the natural rate o f unem ploym ent?
Ellen R. Rissman
The unemployment rate is the composite
of three distinct types of unemployment: fric­
tional, cyclical, and structural. This fact poses
a potentially serious problem for government
policymakers because high unemployment rates
are not necessarily indicative of a slack econ­
omy. Structural change as well as cyclical fac­
tors affect the unemployment rate. If
policymakers are not able to distinguish higher
unemployment rates due to a change in the
structure of employment from higher unem­
ployment rates due to a weak economy, then
they run the risk of implementing expansionary
policies at the wrong time, thereby creating or
adding to inflationary pressures. Hence, to
adequately gauge the state of the economy, it
is necessary to know what portion of the cur­
rent unemployment rate is due to purely cy­
clical phenomena as opposed to structural and
frictional.
The natural rate of unemployment is de­
fined simply as the rate of unemployment that
is compatible with a steady inflation rate. The
natural rate can therefore be thought of as the
rate of unemployment that would occur in the
absence of cyclical fluctuations. In other words
the natural rate is essentially the sum of struc­
tural and frictional unemployment. Because
structural and institutional factors change over
time, the natural rate of unemployment will
also vary. However, the need to understand
the determinants of the natural rate and its re­
lation to the actual rate of unemployment is
quite real as the cost of error may be acceler­
ating inflation or deflation.
The purpose of this article is to answer the
question: What is the natural rate of unem­
ployment? The answer relies heavily on the pi­
oneering work of Lilien (1982) and is in two
parts. First, a working definition of the natural
rate of unemployment is developed. Second,
with this definition, estimates of the natural
rate of unemployment are calculated.
The analysis indicates that the natural
rate of unemployment has been quite variable
over the last 27 years, reaching a high of 7.01
percent in the third quarter of 1981 and at­
taining a low of 3.48 percent in the first quarter
of 1966. But to understand the performance
Federal Reserve Bank of Chicago



of the economy, it is the difference between the
natural rate and actual rate of unemployment
that is significant. This difference has varied
widely over time. From 1958 through 1966 the
natural rate was well below the actual; the re­
verse held from 1967 to 1973. From 1974
through 1976 the actual rate again exceeded
the natural rate although in more recent years
the reverse appears once more to be the case.
Because the difference between the nat­
ural and actual rates of unemployment is
thought to be indicative of the degree of
tightness in the labor market, this measure
should be positively correlated with the in­
flation rate. Indeed, the correlation coefficient
between the difference and the inflation rate
as measured by the Consumer Price Index is
0.46. This compares with an almost zero cor­
relation of inflation with the actual unemploy­
ment rate.
Categories of unemployment

In general it is useful to distinguish con­
ceptually among three distinct types of unem­
ployment in analyzing the historical pattern of
the unemployment rate.1 First, there is fric­
tional unemployment. Frictional unemploy­
ment arises as a result of the normal labor
turnover that occurs in a healthy dynamic
economy. At any given time employed workers
change jobs, lose jobs, or leave the labor force.
Similarly, unemployed workers may find em­
ployment or may decide to stop seeking em­
ployment, while still others may enter or
reenter the labor force. Even in the best of
times there is some unemployment that arises
from this dynamic friction in the economy.
The type of unemployment that is per­
haps perceived and felt most acutely is cyclical
unemployment. As its name suggests, it is the
type of unemployment that is associated with
business cycles. Decreases in aggregate de­
mand such as occur during recessions cause a
general overall decline in labor demand. The
real wage rate is relatively unresponsive to
Ellen R. Rissman is an economist at the Federal Reserve
Bank of Chicago.
3

these changing conditions, that is, real wages
do not decline as labor demand declines.2 As a
result, unemployment occurs. If real wages
were free to adjust to these changed conditions
in the labor market, then recessions would not
produce any noticeable increase in the unem­
ployment rate. Cyclical unemployment is
temporary and when demand conditions return
to their previous level, the excess labor supply
disappears. Even permanent declines in ag­
gregate demand result in only temporary un­
employment because sooner or later wages in
a competitive economy must adjust so as to
equate labor supply and labor demand, though
now at a lower equilibrium wage rate.
The third type of unemployment is prob­
ably the least understood and also the most
traumatic to endure.
Unlike cyclical
unemployment, structural unemployment is the
result of shifts in the relative demand for differ­
ent types of labor. Whether these relative shifts
in labor demand are caused by changes in rel­
ative factor prices (e.g., an oil price shock),
technological innovations, changes in tastes
and preferences, or perhaps changes in institu­
tional or other characteristics of the economy,
is not important. The essential point is that as
labor demand for one type of labor falls relative
to another, a temporary mismatch occurs be­
tween the skills that employers desire and those
that the work force actually possesses. This
produces only temporary unemployment be­
cause in time those who are structurally unem­
ployed will either retrain to find employment
in the now higher labor demand industries, re­
locate to find jobs requiring the types of skills
they already have, or perhaps leave the labor
force altogether, in which case they are not
counted as unemployed. How long this process
takes depends upon the costs of education, the
costs associated with relocating or finding em­
ployment further from one’s original location,
and the costs of job search, and, of course, ad­
ditional opportunity and psychological costs.
In terms of these three components, the
natural rate of unemployment is simply the rate
that would occur in the absence of cyclical
fluctuations. It is the sum of frictional and
structural unemployment.

with the periods between business cycle peaks
and troughs shaded for reference. There ap­
pear to be three distinct phases. The decade
of the 1950s is characterized by three re­
cessions, with unemployment peaking at each
economic downturn. Between these periods,
the unemployment rate hovered somewhere
between four and five percent. Even when the
unemployment rate reached its highest value
of 7.37 percent, it was substantially below the
two-digit unemployment rates of recent years.
The decade of the 1960s was one of eco­
nomic growth with no major recessions re­
corded after 1961. And as a result, the
unemployment rate drifted downwards from a
high at the depth of the recession of 7.00 per­
cent to a low in 1969 of 3.40 percent.
Unsurprisingly, structural unemployment
was not an issue at this time. Indeed, the pat­
tern of unemployment is very well explained
by two components: cyclical and frictional.
The business cycles of the 1950s and early
1960s attest to the significance of the cyclical
element, while the relatively economically calm
remainder of the 1960s underscores the impor­
tance of frictional unemployment.
Economists and policymakers of the time
alike recommended a seemingly reasonable
unemployment rate target for policy of around
three percent. This three percent level was
called, with perhaps unconscious irony, the full
employment rate of unemployment. While the
nomenclature is unfortunate, the term was
meant to indicate the level of unemployment
that would occur in the absence of cyclical
factors. From the perspective of the 1950s and
1960s, then, the full employment level of unFigure 1
C iv ilia n u n em p loym en t rate
percent

Historical perspective

Figure 1 presents the civilian unemploy­
ment rate quarterly from 1948 through 1985
4




Economic Perspectives

employment was essentially the frictional level
of unemployment.
The 1970s and 1980s to date exhibit a
much different unemployment rate pattern.
Over this time the unemployment rate rose
from a low of 4.17 percent in the first quarter
of 1970 to a high of 10.60 percent in the fourth
quarter of 1982. As in previous years, the un­
employment rate responded to cyclical factors,
peaking in the trough of each of the four major
recessions. But, the unemployment rate ap­
pears to be trending upwards during the period
so that the average unemployment rate from
1970 through 1985 was 6.94 percent as com­
pared to 4.51 percent and 4.78 percent respec­
tively for the 1950s and 1960s. In addition, the
unemployment rate appears to be much more
volatile in these later years: The calculated
standard deviation is 1.50, compared to stan­
dard deviations of 1.28 and 1.08 in the two
earlier decades.
Demographic change

This abrupt change in the pattern exhib­
ited by the unemployment rate suggests that
there were factors involved other than merely
frictional and cyclical unemployment. One
possible explanation is that the underlying la­
bor force demographics changed, thereby ad­
versely affecting the unemployment rate.
Specifically, the labor force composition
changed over the 1970s relative to what it was
in the 1960s in such a way that the labor force
now contains a significantly higher proportion
of individuals subject to higher unemployment
rates, such as nonwhites, females, and youths.
A simple way of testing the effects of the
changing demographic composition of the la­
bor force on the unemployment rate is to com­
pare the actual civilian unemployment rate
(UR) with a fixed-weight unemployment rate
(WUR). Specifically, the unemployment rate
is calculated as:
/
UR, = Y , y“ UR'‘
[l]
1=1
where URt is the unemployment rate at time t,
i indexes the I demographic groups, yit is the
fraction of the total labor force in group i at
time t, and the sum of the y1(’s equals one.
The fixed-weight unemployment rate at
time t is calculated as:
Federal Resen/e Bank of Chicago




/

WUR, = Y , y« URi, i

I

[2]

l'= 1

where r is some pre-assigned base period.
Thus, the fixed-weight unemployment rate
computes what the civilian unemployment rate
would have been if the demographic composi­
tion of the labor force had remained as it was
in base period t.
Figure 2 plots the differences between the
actual quarterly unemployment rate and vari­
ous fixed-weight measures where the base pe­
riod t is selected to be the first quarter of
I960.3 Positive values indicate that the demo­
graphic changes that have occurred relative to
the first quarter of 1960 unfavorably affect the
unemployment rate while negative values indi­
cate that the unemployment rate would have
been higher if the demographic composition of
the labor force had been the same as in the base
period. The calculations were done for race,
sex and age categories.4
As is obvious from Figure 2, the increase
in the proportion of females, nonwhites, and
young people in the labor force resulted in a
small increase in the unemployment rate. The
most important effect occurred as a result of
changes in the age distribution. At its peak in
1975, the changing age distribution contributed
around three quarters of a percentage point to
the overall unemployment rate. However, this
effect has been decreasing as the labor force has
aged.
In contrast, the changing racial composi­
tion of the labor force tended to increase the
Figure 2
E ffe c t o f c h a n g in g d e m o g ra p h ics on
the u n em p lo ym en t rate
percent

5

unemployment rate at an accelerating rate over
the 1970s and early 1980s, reaching its maxi­
mum effect in 1983. But race never contributes
more than one quarter of a percentage point to
the aggregate unemployment rate.
Finally, the increased labor force partic­
ipation of women relative to 1960 has for the
most part adversely affected the unemployment
rate, contributing approximately an additional
two tenths of a percentage point in 1978.
However, since 1979 the relation between the
sex composition of the labor force and the un­
employment rate has become less marked due
to a decline in the unemployment rate of fe­
males relative to males.
This change is not necessarily attributable
to lower levels of sex discrimination. An alter­
native explanation may be that women are
clustered in jobs that are relatively more pro­
tected from market forces. For example, blue
collar jobs are more frequently filled by men
than women. Those blue collar jobs that are
located in declining industries would contribute
to a higher unemployment rate for males than
for females, all other things equal.
Thus, it seems that the changing demo­
graphic composition of the labor force has re­
sulted in an increase in the civilian
unemployment rate since 1960, but the magni­
tude of the effect is quite modest—adding less
than one percentage point to the total unem­
ployment rate. Even after controlling for
changes in the demographic composition of the
labor force, the unemployment rate of the
1970s and early 1980s is still significantly
higher and more volatile than in the previous
two decades.
Changing industrial composition

Just as the demographic distribution (and
possibly the geographic distribution) of the la­
bor force provides clues to analyzing the more
recent behavior of the unemployment rate, the
distribution of employment across industries
also plays a role. It is the changes in the dis­
tribution of employment across industries that
is most closely related to the concept of struc­
tural unemployment. As noted previously,
structural unemployment arises due to relative
shifts in the demand for different types of labor
causing a period of economic adjustment dur­
ing which time some displaced labor will be
temporarily unemployed. Changes in the rela­
6



tive demands for labor will be accompanied by
changes in the distribution of employment
across industries.
Perhaps the most prominent movement in
the employment profile in recent history is the
change of the private economy from one based
upon manufacturing and other traditional in­
dustries to one based upon services and
service-related industries.3 Figure 3 presents
this trend for selected industries, and prompts
important observations. First, the decline in
manufacturing and concurrent rise in the share
of employment in services are not recent phe­
nomena. The graph shows that these adjust­
ments have been occurring almost continuously
throughout the post-World War II period.
Secondly, even within manufacturing
there are notable differences between the be­
havior of employment shares in durable and
nondurable goods. The decline in nondurable
goods has proceeded much more smoothly than
the decline in employment share in durable
manufacturing.
This steady decline in the relative impor­
tance of nondurable manufacturing is not nec­
essarily an indication of structural change in
the sense that it documents the ebb and flow
of the fortunes of the industry in question. The
historical pattern is also consistent with a steady
stream of technological innovation which enables pro­
duction to remain unchanged while employment levels
decline. While the steady decline in employ­
ment share is almost certain to contribute to
the flow of unemployment, it may well be that
the unemployment generated is much less in
volume and of shorter duration than that which
would occur in industries experiencing a more
sporadic, volatile decline such as durable man­
ufacturing. The reason is that rational workers
are more likely to be able to predict and
therefore cushion or even avoid the blow of
unemployment altogether by preparing for the
event sufficiently in advance.
The third observation concerning the
patterns seen in Figure 3 pertains to the effect
of business cycles on the distribution of em­
ployment across industries. Recessions clearly
and consistently are associated with declines in
employment share in durable goods manufac­
turing. It is well known that business cycles
have a differential impact across industries, af­
fecting some more adversely than others.6 Just
why this occurs depends upon the nature of the
demand for the good as well as the costs of inEconomic Perspectives

Figure 3
S h a re o f to ta l e m p lo ym en t in
se le cte d in d u strie s
percent

ventorying. If the good is viewed as a luxury
item or requires a relatively large expenditure,
then purchases are more likely to be postponed
during periods of low aggregate demand, when
discretionary income falls. For example, hous­
ing starts and new construction are particularly
susceptible to changes in the economic outlook.
In addition, those industries with high inven­
tory costs are less able to smooth production
and are therefore more susceptible to the
vagaries of the market.
The post-World War II era has seen
considerable change in the distribution of em­
ployment across industries. Such shifts in em­
ployment are likely to generate unemployment
temporarily as displaced workers search for
employment. Large movements in employ­
ment across industrial sectors are likely to be
associated with temporary increases in the un­
employment rate because these movements
signify a change in the underlying structure of
the economy. However, change in and of itself
does not cause unemployment. The unem­
ployment arises because of friction or inertia in
the economy which make it difficult for indi­
viduals to adapt instantaneously. Given these
frictions, the larger is the flow of workers into
and out of the various industries, the more
likely it is that a larger volume of unemploy­
ment will be generated.
One way of measuring these flows is to
define a variable of where:
/
^
[3]
i=l
Federal Reserve Bank of Chicago




sit is the share of total employment in industry
i at time t, git is the growth rate of employment
in the ith industry between period t and period
t — 1, and / is the number of industries. Thus,
of is the weighted sum of squared deviations of
industry growth from average aggregate
growth where the weights are given by the
employment share of the itk industry.
This measure captures those employment
flows that are associated with changes in the
distribution of employment across industries
and not those changes in employment that oc­
cur as a result of economic growth. Further­
more, those industries experiencing a large
deviation in employment growth relative to the
average growth rate of employment are given
more weight in the calculation due to the
squaring of the term in parentheses. Such a
weighting scheme is appropriate if, for exam­
ple, large deviations in employment growth
from the average are associated with dispro­
portionately large increases in unemployment.
For further details on the interpretation of of
see Box, Measuring employment flows.
Figure 4 displays the measure of employ­
ment adjustment of from the first quarter of
1947 through 1985.7 There appear to be many
periods of rapid employment adjustment across
industries during the post-war period. Fre­
quently, these adjustments are coincidental
with business cycles as noted in the preceding
discussion of industry employment shares. The
period from 1947 to 1960 is marked by three
episodes of employment adjustment corre­
sponding roughly with the recessions in 1950,
1954, and 1958. The more stable 1960s exhibit
very little change in the distribution of em­
ployment across industries. The 1970s and
early 1980s in contrast indicate a pronounced
change in employment shares occurred in late
1970 and again in 1975 and 1978. The 1980s
are surprisingly stable in comparison to the ex­
perience of the 1970s, providing preliminary
evidence that structural change was perhaps
not a major contributing factor to the histor­
ically high unemployment observed in the 1982
recession.
The effect of employment adjustment
across industries on the civilian age-weighted
unemployment rate is analyzed over the period
from 1954 through the third quarter of 1985.8
The results of the analysis are found in Table
1, which presents the estimates and associated
7

Figure 4
M easure o f e m p lo ym e n t a d ju stm e n t
a c ro s s in d u strie s

Table 1
The e ffe c t o f e m p lo ym e n t ad ju stm en t
on th e un em p lo ym ent rate

(D

(2)

(3)

G NPt

-0 .0 1 8
(0.002)

-0 .0 1 9
(0.002)

-0 .0 1 6
(0.002)

G NPt._ !

-0.011
(0.003)

-0 .0 1 2
(0.003)

-0 .0 0 8
(0.003)

G NPt_ 2

-0 .0 0 5
(0.002)

-0 .0 0 5
(0.003)

-0 .0 0 4
(0.002)

Mt

8

-2.551
(3.312)

—

2.444
(3.700)

1.054
(3.417)

Mt - 2




-3 .2 1 5
(3.556)

Mt- 1

standard errors of the parameters of interest as
well as some additional descriptive statistics.
Other variables included in the analysis
are measures of unanticipated changes in real
Gross National Production (GNP) and unan­
ticipated money growth (M). Unanticipated
real GNP is calculated as the residuals from the
estimated ARIMA process generating real
GNP where the estimates are obtained by the
maximum likelihood method. Unanticipated
money growth is computed as discussed in
Barro (1978).
Columns (1) through (3) of Table 1 pre­
sent estimates of ordinary least squares re­
gressions on various sets of variables including
two lagged dependent variables.9
All three models reported in Table 1 in­
dicate that unanticipated movements in real
Gross National Product are negatively associ­
ated with the age-weighted unemployment
rate. Thus, realizations of real GNP above
trend tend to decrease the unemployment rate
while realizations below trend tend to increase
the unemployment rate.
Intuition suggests that unanticipated
money growth should also be negatively asso­
ciated with the unemployment rate if unantic­
ipated positive changes in monetary growth
signal expansionary monetary policy. As seen
in columns (2) and (3) of Table 1, the coeffi­
cient on unanticipated money growth is nega­
tive only for current realizations and positive
for lagged values. However, the magnitude of
the effect is imprecisely determined as seen by
the large associated standard errors.
Finally, the inclusion of current and
lagged values of the measure of employment

—

—

3.313
(3.946)

2.721
(3.619)

AO
°t

-

-

790.443
(212.599)

AO
°t-1

-

-

393.051
(234.715)

AO
at - 2

-

-

-3 8 7 .7 0 9
(234.671)

AO
at - 3

-

-

106.700
(215.610)

AO
°t-4

-

-

-6 0 3 .3 4 3
(199.532)

U * t-1

1.387
(0.074)

1.385
(0.076)

1.308
(0.086)

U Pt- 2

-0 .4 0 6
(0.076)

-0 .4 0 2
(0.079)

-0 .3 1 9
(0.086)

0.093
(0.087)

0.078
(0.091)

0.030
(0.083)

R2

0.976

0.977

0.982

Q

5.74

5.43

3.80

C

adjustment has a clear and significant effect on
the unemployment rate. Increases in the
amount of interindustry employment adjust­
ment have an initial adverse affect upon the
unemployment rate, as expected. Thus, the
larger are the changes in the distribution of
employment across industries, the higher is the
unemployment rate. The long term effects of
such shifts in employment are not immediately
obvious, however, due to the inclusion of the
two lagged dependent variables in the re­
gression model. The difficulty arises because
current changes in cr2 affect not only the cur­
rent unemployment rate but also future unemEconomic Perspectives

Measuring employment flows

To further motivate the use of of ,
defined in equation [3], let eit be employ­
ment in industry i at time t and let
et= H eit be the total level of employment
in the economy at time t. The change in
the number of people employed in indus­
try i between periods / and / — 1 is simply
eu ~ eit-1* However, employment changes
can occur for two reasons: economic
growth and shifts in the underlying indus­
trial composition of employment. For
purposes of measuring structural change
and relating structural change to the un­
employment rate, adjustments in employ­
ment due to shifts in the employment
distribution across industries alone are of
interest. Thus, the expression elt — sit_xet is
simply the difference between employment
in industry i at time t and the amount of
employment in industry i that would have
occurred at time t if the ilh industry had
grown at the same rate as the aggregate
economy, i.e. the employment share of in­
dustry i had remained unchanged. Obvi­
ously, if no change in employment share
had occurred, then the expression eit —site,
equals zero. Similarly, if eit —sit_xet is posi­
;= i

ployment rates. Simulations show that the
maximum effect of an increase in the volume
of interindustry employment changes is felt af­
ter a one-quarter lag, damping thereafter.
It should be noted that while the coeffi­
cient estimates on current and lagged values of
o2 are quite large in magnitude, the actual
values of o2 are relatively small, with an aver­
age value over the entire time period of
1.3xl0-4. If, for example, a one standard de­
viation increase in o2 occurred at time 0, the
unemployment rate would rise by only 0.20
percent in the first quarter, 0.36 percent in the
second quarter, and 0.16 percent after one
year.
While the evidence reported in Table 1
suggests that the volume of interindustry
movement of employees is positively related to
the unemployment rate, the interpretation that
Federal Reserve Bank o/ Chicago



tive (negative), then the ith industry’s em­
ployment share is rising (falling).
The change in employment attribut­
able solely to changes in employment
share and not economic growth can be re­
written in terms of growth rates as
S t ) • Since the unemployment rate
is assumed to respond to the magnitude
and not the direction of employment
changes, the total volume of employment
flows attributable to shifts in the distrib­
ution of employment is simply calculated
as i=l i I Sit ~ St I which is proportional to
Heit1
I
.=1 *.7-i I Sit ~St I • Finally, by squaring the
amount within the absolute value signs,
the original expression for of results.
As discussed briefly in the text, by
squaring | gu —gt | those industries expe­
riencing relatively large deviations of
employment growth from the aggregate
are given more weight in the calculation.
Since structural shifts of large magni­
tudes are thought to have a dispropor­
tionately large impact upon the
unemployment rate, such a weighting
scheme is appropriate.
the associated movements in the unemploy­
ment rate are due to structural change is not
that easily justified. Thus, drawing inferences
about structural unemployment or the natural
rate of unemployment from the results found in
Table 1 is inappropriate. The difficulty arises
because of the simultaneous effect of cyclical
and structural factors on employment flows.
As illustrated in Figure 3, the ebb and flow of
employment shares is dependent upon the stage
of the business cycle. In durable manufactur­
ing recessions are invariably associated with
declining employment shares and therefore a
greater amount of employment adjustment.
The problem therefore is to develop a measure
that distinguishes employment flows attribut­
able only to structural factors from employ­
ment flows attributable to purely cyclical
factors.
9

Structural change

There are a variety of ways to extract the
purely cyclical effect on the distribution of em­
ployment across industries from the purely
structural. These techniques all rely upon an
assumption that cyclical changes in employ­
ment are temporary while structural changes
are more or less permanent by definition.
In attempting to eliminate the effect of
cyclical factors on the distribution of employ­
ment across industries, calculations can proceed
along one of two lines. A measure of the vari­
ability of employment shares (or possibly em­
ployment growth) across industries can be
calculated first and then decomposed into a
permanent (structural) component and a tem­
porary (cyclical) one. Alternatively, the em­
ployment share or level in each industry is
decomposed into its permanent and cyclical el­
ements and then, using only the permanent
portion, a single measure of permanent change
in employment distribution is devised. The
first approach, while computationally easier,
may obscure much of the underlying dynamics
which by hypothesis are what give rise to
structural unemployment. For this reason the
second approach is preferred.10
As noted above, certain industries experi­
ence relatively smooth changes in employment
shares over time while others experience much
more volatile changes. While both of these
types of changes can be permanent, intuition
suggests that abrupt permanent changes in
employment share add more to the volume of
unemployment than do smoothly occurring
changes. Thus, the permanent portion of
changes in employment shares that is not
explainable by past experience is the appropri­
ate measure of structural change.
Calculating the difference between the
actual employment share in industry i at time
t and that which would be expected based upon
past behavior is relatively straightforward.
However, separating this measure into its per­
manent and temporary components is a more
complicated endeavor. See Box, Measuring
structural change.
Assuming that deviations of employment
shares from trend in industry i at time t can be
accurately decomposed into permanent
changes (A£) and temporary changes (A/f) ,
then the measure of permanent structural
10



change for the aggregate economy at time t
(Ap) is simply defined as:

Similarly, the measure of temporary change in
employment shares (AJ) , is defined as:
i=l
Because the expression in parentheses is
squared, effectively those industries experienc­
ing relatively large permanent changes in em­
ployment shares are weighted more heavily in
the calculation.
The behavior of Ap and Ar from the first
quarter of 1952 through the third quarter of
1981 is examined in Figure 5.11 As can be seen,
permanent changes in the distribution of em­
ployment across industries correspond closely
to business cycles, exhibiting quite noticeable
peaks in 1958, 1961, 1970, and 1975, and pos­
sibly in 1980. In contrast, temporary changes
in the employment distribution do not appear
to be significantly correlated with the business
cycle.
A comparison of the measurement of em­
ployment adjustments, a2, with the constructed
measure of permanent structural change, Ap,
yields some interesting insight. The crude
measure of employment adjustment records its
largest value in 1975, leading to the premature
conclusion that structural change was most
Figure 5

Perm an en t and te m p o ra ry c h a n g e s
in th e d istrib u tio n o f e m p lo ym en t
a c ro ss in d u strie s

Economic Perspectives

Measuring structural change

Let $ be the lx 1 vector of employ­
ment shares at time t. Thus s, is simply
defined as (^1/5 s2„..., sIt) for t = 0,. . T
where (') indicates the transpose and sit is
the employment share of industry i at time
t. The vector of employment shares is as­
sumed to be related to its past and future
values. Specifically, assume that
3
k = 0 - «/)][] Pjk-j

j= 1

3
X!

+atj= 1

+

[i]

£t

where a, is some time varying parameter,
the ft’s are geometrically declining
3
weights, S ft = 1 and st is an additive in7=1
dependent and identically distributed
random error term. Thus, the current
vector of employment shares is assumed
to be a two-sided moving average of its
past and future values. Subtracting
3
£ Pfi-j from both sides of equation [i],
7=1
the following results:

j=l

t- j =

[“ ]

pronounced at that time. The more refined
measure of permanent structural change, on
the other hand, clearly indicates that structural
change was far less important a factor in the
1975 recession than it was in the 1970 recession.
Interestingly enough, even the recession that
occurred in 1958 appears to have been associ­
ated with a more pronounced permanent
change in the structure of employment than
was the 1975 recession.
Federal Reserve Bank of Chicago




j= 1

+

j= 1

The left hand side of equation [ii] can be
interpreted as the deviation in current
employment shares from its expected value
based upon past experience. This devi­
ation is seen to be the sum of two compo­
nents: a temporary component, st, and a
permanent component,
3
3
7-1
7=1
Equation [ii] can be estimated by
ordinary least squares assuming a fixed J
and specific values for the ft’s. The per­
manent component for the ith industry is
simply defined as:
Al =
[iii]
3
3
az C ^ fysit+j ~ ^ fysit-j]
j= 1
j=1

where indicates the estimated value of
the parameter, while the temporary com­
ponent is calculated as the regression resi-

3

3
h -'Y j

3
3
* /[ £ ¥ t + j ~ Y j

j= 1

j= 1

3

j= 1

Structural change and the
unemployment rate

The calculation of permanent and
transitory changes in the distribution of em­
ployment across industries is a refinement of the
measure of interindustry employment flows of
employed previously. It is constructed so as to
give meaning to the concept of structural
change. If structural change is rapid and ac­
companied by large employment shifts, then
11

unemployment is thought to be the by-product
as workers struggle to adapt to the changing
situation.
Analysis of the relation between the com­
puted permanent and transitory variation of
the employment distribution and the unem­
ployment rate may proceed along lines similar
to that presented in Table 1. However, it is
implicitly assumed there that once demo­
graphic changes have been controlled for, all
other unemployment results from cyclical,
structural, or frictional factors where frictional
unemployment is assumed to be some constant
amount. In essence, this assumption denies the
existence of other factors, particularly institu­
tional arrangements, that have an effect upon
the unemployment rate.
Changing institutional conditions are not
cyclical by nature. Nor should they be thought
of as contributing to structural unemployment
because structural unemployment as defined
here is the result of the changing relative de­
mand for different types of labor. These
changing institutional characteristics are most
properly associated with frictional unemploy­
ment. As discussed elsewhere, frictional unem­
ployment arises due to the functioning of a
dynamic labor market where workers are con­
tinuously making decisions as to the proper al­
location of their labor. These decisions are
based upon the parameters of the underlying
institutional framework. Thus, when this
framework changes, it will also have an effect
upon the decisions of the workers to seek work
or quit work, and therefore it will have an ef­
fect upon the frictional rate of unemployment.
Much research has been devoted to ana­
lyzing the effect of unemployment insurance on
job search. Critics argue that the existence of
such unemployment insurance schemes lowers
the costs of job search and therefore encourages
unemployed workers to remain unemployed for
a longer duration than they would have in the
absence of such benefits. Thus, more lenient
benefits tend to increase the unemployment
rate. While this may in fact occur initially, it
is also quite possible that by encouraging peo­
ple to search longer for employment, better job
matches between employees and employers will
result, thereby having a negative long-run ef­
fect on the unemployment rate.
The regression models presented in Table
2 analyze the effect of permanent and
transitory changes in the distribution of em­
72



ployment on the age-weighted unemployment
rate. As presented previously, other explana­
tory variables include deviations in real Gross
National Product from trend and unantic­
ipated money growth. In light of the preceding
comments on institutional arrangements, an
additional variable is included (SI) which is
social insurance expenditures as a percentage
of Gross National Product.12 This variable is
assumed to proxy for the costs associated with
unemployment.
Table 2 presents ordinary least square es­
timates from regressions on the civilian ageweighted unemployment rate over the period
from 1954 through the third quarter of 1981.
Parameter estimates and their associated
standard errors in parentheses are reported
along with some descriptive statistics. As in the
regressions reported in Table 1, the models of
Table 2 include two lagged dependent vari­
ables. Therefore, the OLS estimates are
asymptotically equivalent to maximum likeli­
hood estimates only if the errors are not
heteroskedastic. The adjusted Box-Pierce sta­
tistic (Q) is reported testing for autocorrelation
of the estimated residuals for a lag length of six
quarters. Judging from the small magnitude
of this statistic, the residuals appear to be
“white noise”.
The results indicate that structural
change adversely affects the unemployment
rate while transitory changes in the distribution
of employment across industries have no dis­
cernible effect. As in Table 1, the inclusion of
lagged dependent variables in the regression
model complicates the interpretation of the co­
efficients. This occurs because current struc­
tural change not only affects the current
unemployment rate but also influences the fu­
ture time path of the unemployment rate di­
rectly through a one-quarter lag and indirectly
through the two lagged dependent variables.
Figure 6 reports the results of a simulation
based upon the parameter estimates found in
column (1) of Table 2. The effect of a one
standard deviation temporary increase in Ap at
time 1 on the time path of the unemployment
rate is analyzed. By temporary, it is meant that
the disturbance occurs at time 1 after which
Ap returns to its previous level. As seen in the
graph, although current structural change ad­
versely affects the unemployment rate both
currently and into the future, the effects damp
quite quickly. A one standard deviation rise in
Economic Perspectives

Table 2
S tru c tu ra l change and th e
u n em p lo ym en t rate

(D

(2)

G N Pt

-0 .0 0 9
(0.003)

-0 .0 0 9
(0.003)

G " P t- 1

-0 .0 0 8
(0.003)

-0 .0 0 8
(0.003)

G N P t_ 2

-0 .0 0 2
(0.003)

-0 .0 0 2
(0.003)

Mt

-7 .7 4 0
(3.330)

-7 .4 1 3
(3.444)

Mt- 1

3.083
(3.492)

3.293
(3.564)

Mt - 2

-0 .5 6 3
(3.780)

-0 .3 5 8
(3.848)

Af

94.448
(14.734)

96.450
(16.439)

-8 6 .2 1 3
(14.683)

-8 8 .5 1 4
(16.545)

A f-,
AI

—

0.537
(35.092)

A [-1

—

9.103
(36.516)

3.332
(1.038)

3.158
d -1 1 4 )

1.185
(0.078)
-0 .2 4 7
(0.078)

1.177
(0.082)
-0 .2 3 9
(0.081)

0.018
(0.101)

-0 .0 7 2
(0.224)

P2

0.978

0.978

Q

3.30

3.49

s /t

U P t- 1
U P f—2
C

A/J at time 1 causes the unemployment rate to
rise by approximately one half of a percentage
point that quarter. The following quarter
when Ap returns to its previous level, the un­
employment rate is still larger than it would
have been by approximately one tenth of a
percentage point. Within four quarters of the
structural change the effect on the unemploy­
ment rate is small, being only one hundredth
of a percentage point and continuing to decline
thereafter. Thus, the long-term effects, i.e.
greater than one year, of structural change on
the unemployment rate are negligible.
The evidence provided in Table 2 and
Figure 6 suggests that the unemployment rate
adjusts quite rapidly to changes in the under­
Federal Reserve Bank oi Chicago



lying structure of employment. This is surpris­
ing because it is widely held that structural
change is responsible for creating a large pool
of chronically unemployed workers. However,
the numbers indicate that most of the effect
occurs within two quarters of the disturbance
and long-term effects are minimal. This evi­
dence is at least partially corroborated by sta­
tistics on the distribution of unemployment by
duration.
Table 3 reports for the period 1960 to
1985 the percentage of unemployed workers in
a given year who have been unemployed for
various specified lengths of time. As can be
seen, the vast majority of the unemployed be­
come reemployed (or perhaps leave the labor
force) within six months of losing or leaving a
job. Even in the worst year from unemploy­
ment duration standards, less than a quarter
of the unemployed were unemployed for longer
than twenty-six weeks. In fact, much of the
change in the distribution of unemployment by
duration that has occurred over this time ap­
pears to be related to cyclical factors associated
with a general weakness in the labor market.
Thus, the perception that structural
change leads to a more or less permanent pool
of chronically unemployed workers is not en­
tirely justified. However, this evidence should
not be taken as confirmation that structural
factors have an impact of only limited duration
on the overall performance of the labor market.
It may well be the case that structural change
results in an increased frequency of unemploy­
ment rather than an increased duradon so that
Figure 6
E ffe c t o f a one sta n d a rd d e v ia tio n
in cre a se in s tr u c tu r a l c h a n g e on
th e u n e m p lo ym e n t rate
change

13

Tab le 3
D is trib u tio n o f unem ployed
by d u ratio n o f u n em p lo ym en t,
1960-1985

Year
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985

Less
than
5 weeks
45
38
43
43
45
48
55
55
57
58
52
45
46
51
51
37
38
42
46
48
43
42
36
33
39
42

5-14
weeks

15-26
weeks

27
weeks
and
over

13
15
14
13
13
12
10
9
9
9
10
13
12
11
11
16
14
13
12
12
14
14
16
15
13
12

12
17
15
14
13
10
8
6
6
5
6
10
12
8
7
15
18
15
10
9
11
14
17
24
19
15

31
29
29
30
30
29
27
30
29
29
32
32
30
30
31
31
30
30
31
32
32
31
31
27
29
30

SOURCE: Economic Report of the President. February \
Table B-33.

a worker who has been displaced by structural
events may become unemployed more often in
the future than those workers who have not
been so affected.13
It also appears from Table 3 that social
insurance expenditures are significantly posi­
tively associated with the unemployment rate.
Thus, the evidence supports the hypothesis that
increases in the amount and availability of so­
cial insurance that tend to reduce the opportu­
nity cost of unemployment cause an increase in
the unemployment rate.
Measuring the natural rate
of unemployment

If the natural rate of unemployment is,
as stated earlier, the sum of frictional and
structural unemployment, it is now relatively
straightforward to calculate the natural rate
from the regression results presented previously.
14




The natural rate is simply calculated as the rate
of unemployment that would result if all cy­
clical variables, namely GNP and M, were set
identically equal to zero over the entire time
period. To implement these computations, it
is necessary to specify initial values for the nat­
ural rate. However, the effect of these initial
values on the calculations decreases rapidly.
As a result, within two years the natural rate
is virtually independent of the assumed initial
values.
Figure 7 presents the actual age-weighted
unemployment rate and the estimate of the
natural rate of unemployment based upon the
parameter estimates found in column (1) of
Table 2. Initial values of the natural rate were
taken to be equal to the actual values of the
unemployment rate for the first and second
quarters of 1954. The figure shows the esti­
mates over the period from 1958 through the
third quarter of 1981 so as to minimize the in­
fluence of this assumption about initial values
on the natural rate.
As seen from the graph, the natural rate
of unemployment has at times been below the
actual unemployment rate and at other times
has been above it. Until late 1966 the natural
rate was consistently below the actual by as
much as two percentage points. From late
1966 through 1973 the reverse occurred al­
though the natural rate never exceeded the ac­
tual by more than one percentage point. The
rise in the natural rate over this time is due
predominantly to the relatively large amount
of structural change that occurred and to a
lesser extent the increase in social insurance
expenditures as a percentage of GNP. From
1974 through 1977 the actual rate again ex­
ceeded the natural rate while for the brief pe­
riod from 1978 to 1981 the opposite was true.
Not only has the relation between the ac­
tual and natural rates of unemployment
changed over time, but the estimate of the
natural rate has varied widely from a high of
7.01 percent in the third quarter of 1981 to a
low of 3.48 percent in the first quarter of 1966.
This variability of the natural rate makes ap­
propriate policy-making difficult.
As suggested in the introduction, the dif­
ference between the natural rate of unemploy­
ment and the actual rate of unemployment is
directly related to movements in the inflation
rate. If the natural rate exceeds the actual
rate, then labor market conditions are tight
Economic Perspectives

Figure 7
Th e natu ral and fix e d -w e ig h t
u n em p lo ym en t rates

Figure 8
In fla tio n and th e n atu ral rate
o f em p lo ym en t

percent

percent

and inflation occurs. Conversely, if the actual
rate exceeds the natural rate, then labor mar­
ket conditions are slack and lower inflation or
even possibly deflation results. Thus, the in­
flation rate should be positively correlated with
the calculated difference. This indeed seems to
be the case. The estimated correlation coeffi­
cient between the inflation rate and the differ­
ence between the natural and actual rates of
unemployment is computed to be 0.46, indi­
cating that the two do vary directly. In addi­
tion, there is no apparent linear relation
between the inflation rate and the actual un­
employment rate as the calculated correlation
coefficient is a mere -0.01. Although these cal­
culations are somewhat crude, they indicate
that inflation does not depend upon the actual
level of unemployment but rather the actual
rate relative to the natural rate.
Figure 8 displays both the difference be­
tween the natural and actual rates and the an­
nual inflation rate based upon quarterly data.
The inflation rate from 1958 through 1966
fluctuates around two percent per year with no
noticeable upward trend. During this time the
actual unemployment rate was above the nat­
ural unemployment rate, implying that labor
market conditions were somewhat slack. From
1967 to 1973 labor market conditions appear
to be tighter as the natural rate rose above the
actual. Inflation appears to be trending up­
wards during the same time period. Finally,
the two drops in the difference between the
natural and actual rates of unemployment oc­
curring in 1975 and again in 1980 appear to
Federal Reserve Bank of Chicago



coincide with rapid declines in the inflation
rate.
While the two series are clearly positively
related, a great deal of the variation in the in­
flation rate is unexplainable by changes in this
measure of labor market tightness. If the esti­
mates of the natural rate of unemployment are
indeed correct, then a more adequate under­
standing of inflation requires incorporating
other elements of the economy, such as mone­
tary policy, into the analysis.
Conclusions

The historically high unemployment rates
of recent decades are attributable in large part
to a combination of two factors: rapid and
pronounced structural change and low aggre­
gate demand. Although demographic changes
in the composition of the labor force have
tended to adversely affect the unemployment
rate, the actual impact has been quite modest.
The unemployment of the 1970s is at­
tributable in large part to shifts in the distri­
bution of employment across industries brought
on by some sort of structural change. Unfor­
tunately, the measure of structural change de­
veloped can be computed only with a four-year
lag, thereby making policy decisions based
upon such dated calculations inadvisable.
Nevertheless, the need for policymakers -to have
some knowledge of the current magnitude of
structural change is quite real.
The evidence on interindustry employ­
ment flows suggests that structural change has
15

not been as large a determinant of unemploy­
ment in the 1980s as it was in the 1970s. Thus,
the double-digit unemployment of recent years
is more closely associated with cyclical rather
than structural or frictional factors.
Extrapolation of data used in the compu­
tation of Figure 7 suggests that the current
natural rate of unemployment is approximately
6 percent. Given the actual unemployment
rate of 7.07 for the first quarter of 1986, it ap­
pears that policymakers need not be unduly
concerned with inflation at this time.
1 The discussion here is based to a large extent on
Ronald G. Ehrenberg’s and Robert S. Smith’s book
entitled Modem Labor Economics: Theory and Public
Policy, 2nd edition, published by Scott, Foresman
and Company, 1985.
2 Real wages may not readily respond to decreases
in aggregate demand because of long-term labor
contracts which specify nominal wages, minimum
wage legislation, and risk aversion on the part of
workers who prefer fixed real wages and more var­
iable employment.
3 The unemployment rates have been constructed
as in equation [1] so as to guarantee that the sum
of the y,/s equals one.
4 The age categories investigated were
16-to- 19-year-olds, 20-to-24-year-olds, and those
25-years-of-age or older.
5 Other traditional industries include construction,
mining, transportation and public utilities.
Service-related industries refer to wholesale and
retail trades, finance, insurance, and real estate.
6 This phenomenon was originally documented by
Wesley C. Mitchell in Business Cycles and Their
Causes, (Berkeley, CA: University of California
Press, 1941).
7 The industry categories examined include gov­
ernment, construction, mining, durable manufac­
turing, nondurable manufacturing, transportation
and public utilities, services, wholesale trades, retail
trades, and finance, insurance, and real estate.

76




8 The dependent variable in the analysis is the
fixed-weight unemployment rate adjusting for the
effects of the changing age composition of the labor
force. Similar calculations were also performed on
the unadjusted unemployment rate but provide lit­
tle additional insight.
9 In the absence of serially correlated errors, ordi­
nary least squares is equivalent to maximum likeli­
hood estimation for large sample sizes. The
adjusted Box-Pierce statistic (Q) is reported testing
for serial correlation of the residuals through a lag
length of six quarters. In all three regressions the
hypothesis that the estimated residual is not serially
correlated can be accepted at the five percent sig­
nificance level.
10 The measurement of permanent change in the
distribution of employment across industries dis­
cussed below is found in George R. Neumann and
Robert H. Topel, “Employment Risk, Sectoral
Shifts, and the Geographical Distribution of Un­
employment,” forthcoming Quarterly Journal of Eco­
nomics.
11 Calculations are based upon the same ten indus­
tries as those used in computing a2. The measures
have been compiled assuming J=16, creating a
four-year lag in the estimate. As a result, values
of Ap and Ar can be estimated only through the
third quarter of 1981. The s are assumed to be
geometrically declining weights that sum to unity
over 16 quarters. Therefore, = Cq> where
C = (1 —q)!\_q(1 —017)]. The results reported here
are based upon the assumption that q = 0.9.
However, in practice the actual weighting scheme
used makes little difference in the final results. [See
box for a discussion of the estimation.]
12 Social insurance expenditures is available annu­
ally from the Social Security Administration’s Social
Security Bulletin: Annual Statistical Supplement.
Quarterly data were calculated by linear interpo­
lation.
13 Some limited evidence to this effect is found in
Robert E. Hall’s article, “Why Is the Unemploy­
ment Rate So High at Full Employment?” in
Brookings Papers on Economic Activity, vol. 3: 1970, pp.
369-396.

Economic Perspectives

Bibliography

Barro, Robert J., “Unanticipated Money, Output,
and the Price Level in the United States,”
Journal of Political Economy, August 1978,
86(4), pp. 549-80.
Hall, Robert R., “Why Is The Unemployment
Rate So High at Full Employment?,”
Brookings Papers on Economic Activity, 1970(3),
pp. 369-410.
Lilien, David. “Sectoral Shifts and Sectoral Unem­
ployment,” Journal of Political Economy, Au­
gust 1982, 90(4), pp. 777-93.

Federal Reserve Bank of Chicago



Mitchell, Wesley C., Business Cycles and. Their
Causes, (Berkeley, CA: University of
California Press, 1941).
Neumann, George R. and Robert H. Topel, “Em­
ployment Risk, Sectoral Shifts, and the Ge­
ographical Distribution of Unemployment,”
forthcoming Quarterly Journal of Economics.

17

NOW AVAILABLE

Toward Nationwide Banking

One of the major issues facing the financial industry today is that of interstate
banking. Toward Nationwide Banking, recently published by the Research
Department of the Federal Reserve Bank of Chicago, examines this timely
topic from a variety of perspectives.
• Is there a need for interstate
banking?

• What is the driving force
behind interstate banking?
• What are the implications
of various provisions of interstate
banking legislation?
• How will banking law liberalization
affect local market structure?
• Where have nonbanks chosen
to locate and why have they
selected these locations?
• Who will be the acquirers and
who will be acquired when
banking laws are liberalized?
• How will the new interstate
banking laws affect the viability
and independence of small banks?
The research contained in Toward Nationwide Banking should be of valuable
assistance to bankers, legislators, academics, and consumers who are con­
cerned about this emerging development. Toward Nationwide Banking is
available from the Federal Reserve Bank of Chicago, P.O. Box 834, Chicago,
IL 60690 at $10 a copy. Make checks payable to Federal Reserve Bank of
Chicago.

Digitized for 18
FRASER


Economic F
’erspectives

Annual conference assesses banking risk
Richard D. Simmons
Banking risks—and how to deal with
them—were major topics at the Federal Reserve
Bank of Chicago’s 22nd annual Conference on
Bank Structure and Competition held in
Chicago on May 14-16, 1986. Since last year’s
conference, the financial industry has been
rocked by several major events. Privately in­
sured thrifts in Maryland and Ohio were tem­
porarily closed; a record number of agricultural
banks failed; and the Supreme Court issued
decisions upholding nonbank banks and re­
gional interstate banking compacts. In addi­
tion, 31 states have now passed interstate
legislation, and many large nonbank holding
companies compete with banks.
Given these events, this year’s conference
addressed risk-related issues in the context of a
deregulated environment. Some 375 bankers,
regulators, and academicians had an opportu­
nity to hear many different perspectives on risk
in the banking system. Among the speakers
were William M. Isaac, former Chairman of
the FDIC, Walter B. Wriston, former Chair­
man of Citicorp, and George J. Vojta, Execu­
tive Vice President at Bankers Trust Company.
Risk in historical perspective

George G. Kaufman, professor of eco­
nomics and finance at Loyola University of
Chicago, stated that between 1875 and 1919,
before either the FDIC or the Fed existed, rel­
atively few bank failures occurred, due to high
capitalization and significant market discipline.
In addition, illiquid banks were closed imme­
diately, which halted depositor losses. If a
closed bank was still solvent, it reopened soon
afterwards.
Kaufman continued that many depositors
now rely on federal deposit insurance rather
than bank capital for the safe return of their
funds. Accordingly, capital and loan loss re­
serves have decreased, and the risk of
insolvency has increased. Further, regulators
are often lenient regarding loss recognition and
slow to close insolvent institutions. In this en­
vironment, insolvent institutions with nothing
to lose have a strong incentive to take impru­
Digitized Federal Reserve Bank of Chicago
for FRASER


dent risks in an attempt to regain solvency.
Moreover, due to the discount window,
illiquidity does not necessarily limit losses or
force immediate closures. Therefore, while in­
solvent institutions are left open, costs to tax­
payers will increase as loan losses escalate.
Kaufman drew the following conclusions
from this analysis. First, to minimize economic
costs, regulators must close a financial institu­
tion promptly when the market value of the
institution’s net worth reaches zero. However,
large institutions should be sold instead of liq­
uidated. Second, financial institutions should
be required to rebuild capital and loan loss re­
serves quickly, in preparation for any future
losses. On this basis, Kaufman disagreed with
the capital forbearance program for agricul­
tural banks. Third, since only the deposit in­
surance agencies have a monetary incentive to
minimize the costs of failures, authority to de­
clare financial institutions legally insolvent
should be transferred from the chartering
agencies to the FDIC or FSLIC. Finally, the
FDIC/FSLIC should insist on higher capital
ratios, just as depositors did before deposit in­
surance existed.
Risk from a banker’s viewpoint

George J. Vojta, executive vice president
at Bankers Trust Company provided a second
perspective on risk in the banking system.
Vojta described several problems in today’s
banking system. First, banks are too insulated
from market discipline. Currently, nearly
8,000 banks are not audited, too little disclo­
sure exists, and bank examinations are too
confidential. Second, uniform capital ratios
and insurance premiums contribute to poor risk
pricing and encourage excessive risk taking.
Third, unnecessary legal and regulatory barri­
ers preclude banks from diversifying their
product lines and hinder banks’ competitive
abilities. These problems increase failures,
weaken the banking system, and threaten the
system’s long term viability.
Richard D. Simmons is an associate economist at the
Federal Reserve Bank of Chicago.
19

To solve these problems, Vojta argued for
stronger examinations, increased disclosure,
risk-based insurance premiums, and risk-based
capital ratios. He also stressed that barriers to
product diversification must be removed and
that commercial banks must be allowed to sat­
isfy the equity underwriting needs of their best
clients.
Far from seeing a conflict between com­
petition and safety, Vojta agreed with
Kaufman that fostering competition and mar­
ket discipline would provide the best path to a
stronger banking system. Though some banks
would fail, most would adjust successfully, re­
sulting in a stronger global financial system.
Banking risk and the investor

Providing yet another perspective, Harry
V. Keefe, Jr., Chairman and CEO of Keefe,
Bruyette & Woods, Inc., an investment bank­
ing firm specializing in bank securities, said the
problems are with individual banks, not with
the banking system. Although the media have
dramatized the 120 bank failures that occurred
this year, Keefe stressed that this number is
minuscule given that 14,400 banks exist in the
country.
However, Keefe emphasized his belief
that banks’ capital ratios are too low. Many
banks have lower price-to-earnings ratios than
industrial companies with comparable earnings
and growth because the market perceives these
banks as undercapitalized. Keefe asserted that,
by issuing additional equity capital, these
banks could increase their stock prices and de­
crease their funding costs.
In addition, Keefe agreed that more dis­
closure and market discipline are needed.
However, he disagreed with the FDIC’s at­
tempt to promote market discipline by requir­
ing banks to maintain a capital-to-assets ratio
of nine percent, of which up to three percent
could be subordinated debt, because commu­
nity banks would have to pay overly large pre­
miums on subordinated debt due to the small
size of their issues. He also stated that com­
mercial banks should not be allowed to under­
write equity securities because it is
inappropriate for commercial banks to own
stock in their clients.
20




A regulatory perspective

The first luncheon featured guest speaker
William M. Isaac, President of the Secura
Group and former Chairman of the FDIC. A
strong proponent of competition, Mr. Isaac
emphasized that many of the problems in
banking result from competitive inequities. He
recommended reducing these inequities by
equalizing capital requirements for banks and
S&Ls, by including foreign deposits in the cal­
culation of FDIC deposit insurance premiums,
by developing a procedure to ensure that large
and small bank failures will be handled simi­
larly, and by allowing commercial banks to
engage in insurance, real estate, and under­
writing activities. In addition, he argued that
risk-related insurance premiums, a stronger
bank examination force, and increased disclo­
sure would help bring about much needed
market discipline.
Elaborating on the market discipline
topic, Isaac stated that FDIC insurance could
provide less than 100 percent coverage in order
to promote discipline through uninsured
depositors. However, he stated that the
FDIC’s capital proposal is a better approach.
This capital proposal would gradually increase
capital requirements from six to nine percent
of total assets, and subordinated debt could be
used to satisfy up to one third of the nine per­
cent requirement. This debt should increase
discipline by forcing each bank to pay a rate
based on the market’s perception of that bank’s
risk.
Contrary to Keefe’s view, Isaac stated
that banks under SI00 million would not be
significantly burdened by the increased capital
requirement because their primary capital-tototal assets ratios currently average 9.1 percent
and deficiencies at banks with lower ratios are
small. Further, Isaac asserted that these small
banks could place subordinated debt at rea­
sonable costs through correspondent banks, in­
surance companies, and pension funds.
Instead, the heaviest burden of this capital
proposal would be on thrifts and large banks.
According to Isaac, the proposal would equal­
ize capital requirements for large and small
banks, reduce the failure rate, and minimize
FDIC losses.
Economic Perspectives

A banker’ s perspective:
banks

Nonbanks vs.

The second luncheon featured guest
speaker Walter B. Wriston, retired Chairman
and CEO of Citicorp. Like previous speakers,
Wriston emphasized the need to remove re­
strictions on banks so that banks could compete
in the market on an equitable basis.
Throughout his talk, Wriston emphasized
that banks are losing an increasingly large
share of the market to large nonbank compet­
itors such as GMAC, GE, Ford, Chrysleiy
American Express, and Sears. In the mean­
time, bankers and regulators quibble about
how many yards from the head office a branch
may be located. According to Wriston, if this
mentality of looking at the trees instead of the
forest continues, trivial issues such as allowable
distance to a branch will be irrelevant because
banks will have been supplanted by large non­
bank competitors.
Wriston acknowledged that some banks
will fail when a recession occurs. However, he
argued that the purpose of bank regulation is
to ensure a sound banking system, not to keep
poorly managed banks afloat. Regulators must
not try to restrict banks to “safe” activities in
an attempt to limit failures. Banks must be at
liberty to offer new products and to expand
geographically.
Wriston continued that these procompetitive actions will not cause another de­
pression because the Fed will not allow the
money supply to decrease sharply; the discount
window provides emergency liquidity to banks;
and the FDIC guarantees deposits. Moreover,
allowing banks more flexibility to compete will
strengthen the banking system. Accordingly,
Wriston argued that regulatory restrictions
must be removed so that banks can survive
among and freely compete with other financial
organizations.
Encouragement of market discipline

In addition to being a common thread for
the preceding speakers, market discipline was
the topic for many of the research papers pre­
sented. Robert B. Avery and Terrence M.
Belton of the Federal Reserve Board and
Michael A. Goldberg of the Federal National
Mortgage Association found that the interest
rate spread between the subordinated debt of
t ederal Reserve Bank of Chicago




large U.S. bank holding companies and of
comparable Treasury securities was not signif­
icantly related to bank size, capitalization,
earnings, liquidity, or loan quality. These
findings argue that subordinated creditors have
not imposed market discipline on banks.
Looking at depositors instead of creditors,
Herbert Baer and Elijah Brewer, economists at
the Federal Reserve Bank of Chicago, pre­
sented evidence that uninsured depositors re­
quire higher risk premiums on certificates of
deposit when a bank’s market value of equityto-total assets ratio is low or when the variance
of returns on a bank’s stock is high. These re­
sults, presented elsewhere in this issue of Eco­
nomic Perspectives, indicate that uninsured
depositors have been exercising market disci­
pline and that more disclosure would increase
this discipline. Risk premiums on CDs were
also found to be much greater than the differ­
ences in assessments proposed by the FDIC for
risk-based deposit insurance. In addition, any
proposals to extend insurance to these
uninsured depositors would increase bank risk
taking and reduce existing market discipline.
John M. Harris, Jr. of Clemson Univer­
sity, James R. Scott of the University of
Arkansas, and Joseph F. Sinkey, Jr. of the
University of Georgia analyzed market disci­
pline from a different perspective. They argued
that the bailout of Continental Illinois Corpo­
ration discouraged market discipline and
caused a cumulative excess return of forty per­
cent to stockholders of the nation’s largest
banks, because the market perceived that reg­
ulators would not let these large banks fail.
O ff balance sheet activities

Another risk-related topic receiving much
attention at the conference was bank off bal­
ance sheet activities. These activities include
standby and commercial letters of credit, fi­
nancial futures, interest rate swaps, and loan
commitments. Because these activities have
grown rapidly in recent years, with potentially
adverse effects on bank safety, regulators are
considering including them in an adjusted
capital ratio.
Lawrence M. Benveniste and Allen N.
Berger of the Federal Reserve Board argued
that standby letters of credit and other off bal­
ance sheet items improve the social allocation
21

of investment funds because investors can make
direct loans to a bank’s customers by renting
the bank’s credit information on those custom­
ers. Elijah Brewer, Gary D. Koppenhaver, and
Donald H. Wilson of the Federal Reserve Bank
of Chicago argued that off balance sheet guar­
antees are priced by the market, and only the
strongest and most creditworthy banks can ef­
fectively offer these guarantees. Finally,
Marcelle V. Arak, Laurie S. Goodman, and
Arthur Rones of Citicorp Investment Bank
presented an approach for establishing credit
lines for off balance sheet items. They consid­
ered both default and interest-rate risks in de­
veloping their approach.
Although the measurement and manage­
ment of risks in banking were the dominant
topics of this year’s conference, some sessions
were devoted to other issues of importance to
financial institutions and markets. Among
these were alternative banking strategies, mar­
ket value accounting, interstate mergers and
acquisitions, the use of economic models in
banking, and the impacts of deregulation on
banking performance.
Conference consensus

A surprising consensus seemed to emerge
at the conference that banks are not special,

22



that no bank should be considered too large to
fail, that more disclosure is needed, that bank­
ing is in most respects like any other industry,
and that more deregulation is needed. In such
an environment, banks could freely compete
with other financial service providers; well
managed banks would thrive; poorly managed
banks would fail; and a stronger and healthier
banking system would result.
However, it is clear that regulators do
believe banks are special. Regulators subsidize
banks by providing federal deposit insurance
and discount window access at below-market
rates. Further, regulators are proposing tighter
capital adequacy guidelines to decrease the
number of failures and increase the safety and
soundness of the banking system. Regulators
also continue to judge banks’ financial condi­
tions and require improvements in various
areas, again in the name of safety and
soundness.
These conflicting views raise two unre­
solved questions: First, what advantages and
disadvantages do banks have which make them
special in comparison to other financial organ­
izations? And second, how far should deregu­
lation go in removing these differences to level
the playing field between banks and other fi­
nancial service providers?

Economic Perspectives

Uninsured deposits as a source o f
m arket discipline: Some new evidence
Herbert Baer and Elijah Brewer
Money center banks typically place a
heavy reliance on purchased funds, not explic­
itly insured by the FDIC. Suppliers of these
funds will withdraw them from a bank if they
believe that losses are imminent. Since the
creation of the FDIC such deposit runs have
been rare. But in the 1980s Continental Illinois
National Bank experienced two deposit runs.
The first occurred after the failure of Penn
Square National Bank in July 1982 and the
subsequent discovery that Continental had
purchased more than a billion dollars of Penn
Square energy loans. The second run occurred
in spring 1984 and eventually forced the FDIC
to guarantee all of Continental’s creditors.
The experience with Continental has led
many regulators to question the wisdom of a
heavy bank reliance on purchased funds in
general and uninsured deposits in particular.
Others have argued that uninsured deposits
are a source of market discipline, which means
that when they are an important funding
source, banks are likely to take less risk. This
article examines the proposition that CD mar­
kets charge riskier banks higher rates. It begins
by discussing recent trends in reliance on
uninsured deposits, then summarizes previous
evidence on their risk sensitivity, and ends by
presenting the results of some of our own re­
cently completed research.
Previous studies found little evidence that
the market charges riskier banks more for de­
posits outside crisis situations. However, many
of these studies employed inappropriate mea­
sures of bank risk. When we employ bank risk
measures derived from stock price data, we
find, among other things, that even when banks
are solvent, the deposit market does charge
riskier banks more for funds. The new evidence
summarized here suggests that proposals to re­
strict bank reliance on uninsured, purchased
deposits are not costless. While such proposals
might reduce the likelihood of bank runs, they
would at the same time reduce banks’ incen­
tives to control risk.
Federal Reserve Bank of Chicago



Trends in reliance on purchased funds

Purchased funds are generally defined as
all uninsured liabilities with maturities of one
year or less. Uninsured deposits make up the
bulk of most banks’ purchased funds. These
deposits have come to make up a decreasing
portion of deposits at domestic branches of U.S.
banks (see Figure 1). However, from the point
of view of bank safety and soundness, a more
relevant figure is the ratio of uninsured deposits
to total deposits, foreign and domestic. As
Figure 1 illustrates, uninsured deposits’ share
of total deposits fell from 1964 to 1970, rose
from 1970 to 1979 and fell again from 1979 to
1984. By 1984, uninsured deposits had re­
turned to their 1970 share levels.
The data presented in Table 1 suggest
that the recent decline in the relative impor­
tance of uninsured deposits is a result of two
factors. First, there was a modest drop in reli­
ance on uninsured deposits by banks in the
largest size class. Second, and more impor­
tantly, the share of total deposits held by the
largest banks fell from 31 percent in 1974 to 26
percent in 1984. These movements in the im­
portance of uninsured deposits seem to have
more to do with the elimination of Regulation
Q than with any profound change in deposit
insurance or bank supervision.
While there have been no long-term
trends in the overall importance of uninsured
deposits, Figure 1 shows that U.S. banks have
experienced a steady shift from domestic
uninsured deposits to foreign uninsured depos­
its. Unlike domestic uninsured deposits, foreign
uninsured deposits are subject neither to reserve
requirements nor to deposit insurance premi­
ums. This suggests that the shift in uninsured
deposits from domestic to foreign branches
represents in part an attempt to avoid the reHerbert Baer is a senior economist and Elijah Brewer an
economist at the Federal Reserve Bank of Chicago. The
authors thank George Kaufman, Gary Koppenhaver, C. F.
Lee, and Steven Strongin for useful comments.
23

decline in bank reliance on uninsured deposits
also weakens market discipline.

Figure 1
Insured d ep o sit share o f to ta l d e p o sits
ratio

Market discipline and purchased funds

serve requirement tax as well as deposit insur­
ance assessments.
There have also been clear trends within
particular size classes. Table 1 shows how re­
liance on uninsured deposits has varied be­
tween 1974 and 1984 for banks in four size
classes (as of 1984). As one would generally
expect, banks in the largest size class placed
significantly greater reliance on uninsured de­
posits than did banks in other size classes.
Outside the largest size class, bank reliance on
uninsured deposits has steadily increased. This
increase has been greatest for banks in the
smallest size class where the share of uninsured
deposits increased by roughly 67 percent be­
tween 1974 and 1984.
The implications of these changes in the
composition of total deposits for bank risk are
complex. The recent decline in uninsured de­
posits relative to insured deposits has reduced
bank vulnerability to funding risk. However,
to the extent that market discipline exists, a

In the aftermath of the Continental crisis,
the importance of market discipline has been
subject to sometimes heated debate. On the
one hand it has been argued that, because the
funds are not explicitly insured, purchasers of
large CDs will demand higher rates from banks
that are taking more risks. The risk-return
trade-off set by the market will create incen­
tives for bank managers to avoid unwarranted
risk. On the other hand, de facto extension of
deposit insurance to all depositors reduces the
incentive of uninsured depositors to accurately
evaluate bank risk. While the presence of
uninsured depositors creates the potential for
greater market discipline, particularly for
money center and regional banks, realizing this
potential depends on how these depositors per­
mit analysis of available data to affect their
decisions. This, in turn, depends on the ca­
pacity and willingness of these depositors to
evaluate publicly available information on in­
dividual bank performance.
Do CD markets evaluate bank risk?

Since the Franklin National Bank failed
in 1974, the FDIC has conducted various sur­
veys of large depositors to determine how they
evaluate their banking relationship, their sensi­
tivity to their uninsured deposit status, and
their reaction to adverse publicity (Eisenbeis
and Gilbert, 1985). The results of these surveys
suggest that if market discipline exists, it arises
primarily from the actions of large institutional
investors dealing with a few large banks. Past
studies of the links between bank risks and rates

Table 1
Trends in reliance on uninsured deposits by size class o f bank
(to ta l deposits o f size class as p ercent o f to ta l banking system
deposits in parentheses)
< 0 .1 billion

0.1 billion to 1 billion

1 billion to 10 billion

> 1 0 billion

1974

7.9%

(16.0)

13.4%

(24.1)

21.7%

(27.6)

61.9%

1979

9.4%

(17.2)

14.9%

(24.1)

25.1%

(27.3)

63.3%

(31.2)

1984

13.2%

(18.6)

19.9%

(24.9)

33.3%

(30.3)

61.0%

(25.9)

24



(32.1)

Economic Perspectives

on CDs suggest that the resulting market disci­
pline is weak or nonexistent. There is some
evidence that CD markets respond to crises af­
ter the fact, but little evidence that CD markets
distinguish among banks on the basis of infor­
mation regarding the relative soundness of
banks.
Developments in the large CD market in
the aftermath of the Franklin National Bank
(1974) and Penn Square (1982) failures shed
some light on the market’s efficiency in re­
sponding to greater perceived banking risks.
Evidence collected by Gary Gilbert (1983)
subsequent to the Franklin failure indicated
market “tiering,” suggesting that size served as
a proxy for lower risk. This tiering could be
interpreted as evidence of the market’s inability
to isolate individual banking risks on the basis
of differing performance characteristics. After
Franklin National, tiering became somewhat
more selective and the basis point spread be­
tween banks widened. Gilbert found that CD
purchasers required a return from a regional
bank that was 25 basis points higher than the
return required from a large money center in­
stitution. This was double the normal spread
prior to that period. It is not clear whether the
tiering was a rational response to a situation in
which regulators pursued a “too big to fail”
policy, or simply reflected poor use of available
data.
In contrast to these earlier findings, a
preliminary FDIC analysis subsequent to the
1982 Penn Square failure did not reveal a
short-term or a long-term effect on the general
market for large bank CDs, or any tiering by
size. However, for several months after Penn
Square, the CD market penalized the Conti­
nental Illinois National Bank, which was linked
most closely with Penn Square (Gilbert, 1983).
A more recent study by Robert Cramer and
Robert Rogowski (1985) indicates that Penn
Square’s failure did have an effect on the mar­
ket for CDs. They found that CD risk premi­
ums rose approximately 63 basis points after
the announcement of problems at Penn Square
and Continental.
There are several statistical studies of the
factors influencing bank CD rates. A 1974
study by Dwight Crane of the largest 30 banks
revealed a high inverse relationship between
CD rates and bank size. The study found no
consistent relationship between CD rates and
measures of financial condition, such as the re­
Federal Reserve Bank of Chicago




turn on equity or assets, or capital ratios among
banks of comparable size. Crane did find,
however, an apparent relationship between the
profitability of a bank in a given quarter and
its CD rate. It is uncertain whether lower
profitability induced higher CD rates or vice
versa. A 1979 study by Chayim Herzig-Marx
and Anne Weaver found that risk premiums
decreased with increases in total assets and de­
creases in bank liquidity. A recent study by
Robert Cramer and Robert Rogowski (1985)
failed to find any relationship between their
measure of bank-specific default risk and CD
risk premium.
In a recent article, Michael Goldberg and
Peter Lloyd-Davies (1985) perform a timeseries analysis in which dealer quotes on large
CD rates and other variables are aggregated
across the ten prime, top-tier banks included
on the Federal Reserve System’s so-called
“No-name” list. Goldberg and Lloyd-Davies
find that the risk premiums the financial mar­
kets assign to large bank CDs increase as the
amount of risky assets increases relative to bank
capital.
If these studies are to be taken at face
value then we would be forced to conclude that
there is only a tenuous link between bank risk
and CD rates. There are two plausible expla­
nations for such a conclusion. First, holders of
uninsured CDs may believe that regulators will
probably protect them from losses, either by
disposing of the failed banks through purchase
and assumption transactions or by funding de­
posit runs through the discount window. Sec­
ond, regulators may do a fairly good job of
detecting and closing troubled institutions be­
fore uninsured depositors have suffered serious
losses.
Acceptance of either of these conclusions
may not be warranted. Because these studies
were conducted without much attention to
possible sources of CD risk, there is no assur­
ance that risk was properly measured. To
properly measure risk, we must understand the
exact nature of the risks borne by holders of
uninsured CDs.
Sources of CD market risk

Bank debt, including uninsured deposits,
can be viewed as an option contract (Merton,
1974). As long as the book value of the bank
remains above a critical point, the bank is
25

considered solvent and shareholders maintain
control of the firm. However, when the book
value of the firm falls below that point, the
creditors’ option to acquire the bank’s assets is
exercised by having regulators close the bank.
The bank’s debtors receive the value of the
underlying assets. The value of the debt con­
tract increases and the interest rate demanded
decreases when the market value of the firm’s
assets increases, because any such increase in­
creases the cushion available to absorb future
losses. The greater the cushion, the smaller the
chance that depositors will suffer a loss.
The value of the debt contract also in­
creases when the standard deviation of returns
on the bank’s assets declines. A decrease in the
standard deviation of the return on assets
means that there is less chance that the value
of the bank’s assets will fall below the level
needed to fully pay back all depositors.
The impact of a change in book value is
unclear. If book value is perfectly correlated
with market value, then changes in book value
would have no effect on debt values that was
not already captured by changes in market
value. However, book value may diverge from
market value for long periods of time. This
makes it legally possible for a bank to continue
operating after the economic value of its assets
is less than the present value of its liabilities.
This can create incentives for the managers of
the firm to take more risk, leading to a further
decline in debt values. On the other hand such
a policy lowers the probability that the bank
will be closed in the near future. Whether
higher book values result in higher or lower CD
rates depends on whether the prospect of rising
losses in the bank portfolio is offset by the re­
duced probability of default before the CD
matures.
Risk premiums and the probability of
runs can both be reduced if the regulator closes
the bank as soon as its expected market value
hits zero. But even if the regulator tries to use
market value closure rules, the values of many
assets are difficult to monitor. More accurate
estimates of assets values require a greater ex­
penditure of resources. Thus CD holders will
charge a risk premium to cover both the cost
of monitoring asset values and the possibility
that their assessments will be incorrect. Bal­
ance sheet data may be useful in estimating this
type of risk. In particular, publicly traded se­
curities are easily valued using market data,
26




while loans, for which secondary markets are
often thin or nonexistent, are not. As a conse­
quence, risk premiums will be lower, the lower
a bank’s holdings of loans.
The maturity of the CD will also affect
the risk premium demanded by depositors.
How the risk premium changes with maturity
depends on whether the bank is economically
solvent—whether the market value of its assets
exceeds that of its liabilities. If the bank is
economically solvent and its deposits all mature
on the same date, then the risk premium will
decline with maturity. If the bank is econom­
ically insolvent and all deposits mature on the
same date, then the risk premium will initially
increase as the maturity of deposits increases
(Merton, 1974). This suggests that for solvent
institutions, average CD rates should decline
as average maturity increases.
Two other factors may play an important
role in determining CD risk premiums. There
is a strong belief that the larger the bank, the
more likely that any problems will be resolved
in a way that does not penalize CD holders.
This belief was given greater support in 1984
Congressional testimony by former Comptroller
Todd Conover who stated that the nation’s 12
largest bank holding companies were too im­
portant to be permitted to fail. Second, banks
in unit banking states may have less funding
flexibility due to their limited access to retail
deposits. This lack of flexibility may also lead
to an increase in the risks borne by the
uninsured depositors.
Summarizing the preceding discussion,
we would expect that the average rate on
uninsured CDs would increase with increases
in the riskless rate, the standard deviation of
asset returns, and the size of the loan portfolio.
Other things held equal, banks in unit banking
states should pay more for uninsured CDs than
banks in states which permit branching. On
the other hand, increases in total assets and the
ratio of market value of equity to total assets
should cause rates on uninsured CDs to decline.
The effect of changes in the average maturity
of a bank’s CDs or in the ratio of book value
to assets cannot be predicted ex ante.
Data and estimation

We chose to test the preceding prop­
ositions by identifying those factors which affect
the average rate paid on uninsured CDs. This
Economic Perspectives

variable was estimated by dividing total inter­
est paid on large domestic CDs over a quarter
by the average value of large domestic CDs
during the quarter. The average value of CDs
was calculated by averaging weekly data. This
measure of CD rates is less than perfect. In
particular, it fails to account for differences in
maturity. Nevertheless it does reflect the aver­
age cost of uninsured deposits and should ad­
just to changes in bank risk, albeit with a lag.
Because our measure of CD rate is an av­
erage across a number of maturities and origi­
nation dates, it was necessary to control for

differences in CD rates which have nothing to
do with differences in bank risk. We attempted
to address this problem by developing a riskless
rate which controls for the maturity date and
age of each bank’s portfolio.
At least one other macroeconomic factor
is likely to affect the level of CD rates. Many
researchers have found that the rate on a secu­
rity is influenced by its supply relative to the
supply of government securities (Cramer and
Rogowski, 1985, for instance). An increase in
the relative supply of CDs should cause their
rate to rise relative to Treasury securities.

Table 2
Lead banks included in th e study

Holding company name

1979 uninsured deposits as
percentage of total deposits

1979 total assets
(billions of dollars)

American Fletcher Corporation, Indianapolis
American Security Corporation, Washington, D C.
Bank of New York Company, New York
Bankers' Trust New York Corporation, New York
C B T Corporation, Hartford
Central National Chicago Corporation, Chicago
Chase Manhattan Corporation, New York
Chemical New York Corporation, New York
Connecticut National Bank Corporation, Bridgeport
Continental Illinois Corporation, Chicago
Crocker National Corporation, San Francisco
Fidelcor Inc., Philadelphia
First and Merchants, Richmond
First Chicago Corporation, Chicago
First Empire State, Buffalo
First Pennsylvania Corp. Philadelphia
Girard Company, Philadelphia
Harris Bankcorp, Chicago
Hartford National Corp, Hartford
Indiana National Corp., Indianapolis
Lincoln First Banks, Rochester
Manufacturers Hanover Corporation, New York
Marine Midlands, Buffalo
Maryland National Corporation, Baltimore
Mellon National Corporation, Pittsburgh
J.P . Morgan and Company, New York
Northern Trust, Chicago
Pittsburgh National Corporation, Pittsburgh
Provident National Corporation, Philadelphia
R iggs National Bank, Washington D.C.
Security Pacific Corporation, Los Angeles
State Street Boston Corporation, Boston
U.S. Bancorp., Portland
U.S. Trust Company, New York
Union Commerce, Cleveland
Union Planters Corporation, Georgia
Union Trust Bancorp., Baltimore
United Virginia Bancshares, Richmond
Virginia National Bancshares, Norfolk
Wells Fargo and Company, San Francisco

Federal Reserve Bank of Chicago




26
55
46
60
19
43
65
58
12
73
39
30
19
75
18
60
37
53
22
14
12
58
50
27
54
67
49
38
36
33
41
33
18
42
48
08
10
11
14
39

$2,620
2.303
8.989
29.647
2.592
.669
64.129
38.777
.753
34.294
16.087
2.728
2.235
28.984
1.697
8.406
4.305
7.104
2.555
2.080
3.122
45.019
15.690
3.580
13.291
42.435
5.326
5.310
2.361
2.686
23.537
2.220
4.147
1.976
1.173
1.127
1.144
3.052
2.470
19.342

27

Data on daily stock prices and returns
were obtained from Chase Econometrics and
the Center for Research in Security Prices
(CRSP) data base. Thirty-seven bank holding
companies were included in the study. Each
holding company had an identifiable lead bank
and in every case the lead bank accounted for
at least 80 percent of total holding company
assets. On average the lead bank accounted for
94 percent of holding company assets. Table
2 shows total assets and reliance on uninsured
deposits for each lead bank as of December
1979. Balance sheet data and interest paid on
large domestic CDs were obtained from the
Quarterly Reports of Income and Condition. Total
holding company assets and shares outstanding

were obtained from Moody’s. Average
holdings of uninsured CDs were calculated us­
ing the Federal Reserve Board’s Weekly Re­
porting Bank series.
The market value of each bank’s asset
portfolio and the variance in returns on that
portfolio were proxied by the market value of
equity and the standard deviation of the return
on equity. For each month, estimates of the
standard deviation of returns on a bank’s stock
were made using daily data. These monthly
estimates were then averaged together to gen­
erate quarterly estimates of bank stock price
volatility.
Twelve quarters of data beginning in the
fourth quarter of 1979 and ending in the third

Table 3
D ete rm in a n ts o f average CD rates 1979:1V to 1982:111
(t values in parentheses)
Expected
impact on
CD rates

Ordinary least squares
(D

(2)

maturity weighted
T-bill rate

+

relative supply
of CD s

+

average maturity
of CD s

?

.00005
(1.44)

.
, book value ,
109 ( assets 1

?

.0065
(1 7 0 )

,
, market value ,
109 (
assets
>

-

-.0011
(5 0 )

-.0047*
(1.90)

+

.1657
(1 9 7 )

.1751*
(2.14)

standard deviation
of daily stock returns

.8 5 3 8 "
(16.28)

+

log (loans)

+

degree of freedom

.3 7 2 8 "
(4.46)

.00004
(1.12)
.0313tt
(4.28)

-.00 006T
(1 8 8 )

.0044
(6 6 )

.0068
(.71)

- .0 0 8 6 "
(3.26)

.0089**
(3.25)

.1267*
(2 3 1 )

.1252*
(2.29)
-.0 0 1 6
(1 2 2 )
.0005
(1.16)

-.0 0 6 4
(1.22)
-.0 1 5 0

438

434

.3879

.0023
(-38)
.0615*
(2 8 4 )

.4391

438

The branching dummy equals 1 in unit banking states and zero otherwise.
'Significant at the 5% level, one tailed test.
"Significant at the 1% level, one tailed test.
tSignificant at the 5% level, two tailed test.
ttSignificant at the 1% level, two tailed test.

28




.3154**
(3.67)

-.000061(1.94)

.0005**
(3.73)

.0290
(2.40)

(2')

1.3739**
(2.69)

.0108
(1 7 8 )

log (total assets)
x branching dummy

R2

O ')

.6 0 5 1 "
(2.95)

log (total assets)

intercept

.7 7 2 1 "
(13.68)

Fuller-Battese

Economic Perspectives

-.0 3 0 4
(7 0 )
433

quarter of 1982 were pooled, yielding 444 ob­
servations. Using this pooled data, the
equations were estimated using both ordinary
least squares regression and the Fuller-Battese
technique for estimating regression coefficients
when dealing with cross-section time series
data.
Results

The results of this exercise are shown in
Table 3. Each variable’s expected impact on
CD rates is shown in the first column. A re­
gression coefficient of .0001 indicates that a one
unit increase in the variable causes the average
rate paid on uninsured CDs to rise by one basis
point. Changes in the maturity-weighted
Treasury bill rate explain 37 percent of the
variation in CD rates using ordinary least
squares. Including all other risk measures
raises the proportion explained by another 5
percent. The first set of equations, (1) and (T),
includes the weighted T-Bill rate, the relative
supply of CDs, the average maturity of the
bank’s CDs, the book-to-asset ratio, the
market-to-asset ratio, and the standard devi­
ation of stock price returns. Using ordinary
least squares, both the market-to-assei ratio
and the standard deviation of returns have the
hypothesized sign. However, only the standard
deviation of returns is significantly different
from zero. Equation (T) presents alternative
estimates of equation (1) using an estimation
technique designed for cross-section time series
data. In this regression, the market-to-asset
ratio and the standard deviation of stock re­
turns both have the expected sign and are sta­
tistically significant.
Equations (2) and (2') present coefficient
estimates of taking other possible factors into
account. In both equations the market-to-asset
ratio and the standard deviation of stock re­
turns have the expected sign and are statis­
tically significant. The effect of changes in the
relative supply of bank CDs is as expected and
is significant; however, in equation (2) neither
total assets or total loans have the expected ef­
fect. In equation (2') total assets and total
loans have the expected sign but are not sig­
nificantly different from zero. The branching
variable has the expected sign in both cases but
is only significantly different from zero in
equation (2).
Federal Reserve Bank ol Chicago




These results suggest that CD holders are
sensitive to differences in bank risk. They de­
mand higher rates when a bank’s market-toasset ratio is low or when the volatility of bank
stock returns is high. The next question is
whether or not the implied differences in CD
rates are large. To answer this question we
need to know what changes in variables are
plausible. One way this can be established is
by looking at the impact of a one-standarddeviation change in a variable. There is a 68
percent chance a variable will be within one
standard deviation of its mean. Table 4 shows
how a one-standard-deviation change in the
market-to-asset ratio and the standard devi­
ation of bank stock returns translate into
changes in CD rates. Based on the results of
equation (2), a one-standard-deviation increase
in the market-to-asset variable causes CD rates
to fall by 17 basis points. A one-standarddeviation increase in the standard deviation of
stock returns causes CD rates to rise by 16 basis
points. Equation (2') yields even stronger re­
sults in these cases.
This sensitivity of CD rates to change in
these risk variables suggests that the FDIC’s
recent proposal for risk-related insurance pre­
miums ranging from 1 to 8 basis points is sig­
nificantly less sensitive to risk than are the
money markets. It also suggests that a
strengthening of implicit guarantees for
uninsured deposits could eliminate an impor­
tant source of market discipline.
There is, however, one potential problem
with the preceding results. Many researchers
have found a negative relationship between
bank size and CD rates. Our regression results
do not indicate such a relationship.
Nonetheless, our results are consistent
with the earlier findings. While equations (2)
and (2') fail to display a significant negative
relationship between asset size and CD rates,
the market-to-asset ratio and total assets are
positively correlated. This suggests that large
banks will be observed paying lower interest
rates because they have a higher market-toasset ratio.
Postscript

About the same time we completed our
work, we obtained another newly completed
study whose conclusions support our own
(Gerald Hanweck and Timothy Hannon,
29

Equations (1) and (1')

CD rate =

a\ 4-

maturity weighted T-bill rate

+ C relative
j*
+

supply of CDs

rfj* average maturity of CDs

book value of capital
+ •>* '0g <
total assets
>
market value of capital
+ /,* log (--------------------------------- )
total assets
+ gl* standard deviation of stock returns
+

error

Equation (1) assumes that any errors are independently
distributed. Equation (T) assumes that there are three
components to the error term: a bank-specific compo­
nent, a time-specific component, and an observationspecific component.

1985). This study, which employed survey
data on large CD rates for each of five different
maturities, found that the CD risk premiums
increase with both the ratio of risky assets to
capital and uncertainty regarding bank returns
on assets. These effects, in turn, tend to be
more important in the case of the longer CD
maturities, where insolvency risk is presumably
more of an issue. As with our study, the im­
plication is that the market for large CDs helps
to discipline bank risk-taking. The study also
suggests that bank CD rates are strongly af­
fected by accounting-based measures of bank
risk-taking. This latter point is in contrast to
the findings of previous research regarding the
effects of accounting-based measures of risk.
Summary and policy recommendations

The Continental experience indicates that
uninsured depositors will run when they per­
ceive that losses are possible. Many observers
view these runs as potentially dangerous.
However the same factor that generates runs
would also be expected to generate market in­
centives for banks to take less risk. While ear­
lier work using accounting measures of risk
suggests little market discipline, our research
suggests that holders of uninsured CDs set risk
premiums as if they are at least partially at risk.
30




Equations (2) and (2')

CD rate = a% + b2* maturity weighted T-bill rate
+ c2* relative supply of CDs
+ df average maturity of CDs
book value of capital
+ ,5' IOg(
.o,aI assets
1
market value capital
+ J f log (--------— ----- --------)
total assets
+ g2* standard deviation of stock returns
+ h f log (total assets)
+ i f log (total assets) x branching dummy
+ j 2* log (loans) + error
Equation (2) assumes that any errors are independently
distributed. Equation (2') assumes that there are three
components to the error term: a bank-specific compo­
nent, a time-specific component, and an observationspecific component.

This leads to the imposition of market disci­
pline, in a nondisruptive fashion, on large in­
stitutions that are most dependent on the
money market for funding.
Policies that cause banks to reduce reli­
ance on purchased funds by increasing their
reliance on insured deposits will reduce the
likelihood of runs. However, our results suggest
that an important source of discipline will be
lost. This loss will certainly create further in­
centives for banks to take risks and would re­
duce funding flexibility. Purchased funding
became popular precisely because it provides
flexibility.
However, our findings are not yet com­
plete enough to pass judgment on supervisory
policies designed to link capital requirements
to dependence on purchased funds. It is not
enough to show that the purchased funds mar­
ket provides market discipline. We also need
to evaluate the cost and likelihood of runs on
banks which rely on purchased funds. In par­
ticular, we need to show that the costs of bank
runs are or can be made small (George
Kaufman, 1985).
While we cannot presently recommend
acceptance or rejection of proposals to limit
reliance on purchased funds, our findings do
suggest several actions that would improve
market discipline. Our results suggest that CD
Economic Perspectives

Table 4
The im p act o f bank characteristics on th e average cost
o f uninsured CDs

Sample
average
Standard deviation
of daily stock returns
,
, market value .
109 <
assets
>

Sample
standard
deviation

Change in CD rate due to a one
standard deviation increase in variable
based on (2)
based on (2')

.0168

.009

16 basis points

11 basis points

-3 .5 3

.367

17 basis points

32 basis points

markets are trying to evaluate risk. Proposals
that improve the quality of information will
improve the quality of the market discipline.
First, shortcomings of the marketplace in
restraining bank risk-taking could be corrected
to some degree by broadening disclosure. In
particular, the disclosure of bank examination
data could help bank-funding markets to iden­
tify an institution’s weakness while remedial
action is still possible. The impact of such dis­
closure on stock price and deposit flows may
not be as disruptive as some expect. The re­
cently required bank disclosure of past-due and

other nonperforming loans should greatly help
the market assess bank risk-taking.
Second, as demonstrated by the Conti­
nental experience, it is important to accurately
value and close troubled banks of all sizes.
Better monitoring of asset values by regulators
would reduce the likelihood of runs.
Third, our results point out the need for
risk-based premiums. If our results are correct,
the FDIC is dramatically underpricing many
of its deposit insurance policies. If the FDIC
were to adopt the CD market’s attitudes to­
wards risk, then market discipline and the
FDIC’s revenues would both be increased.

References

Crane, Dwight. “A Study of Interest Rate
Spreads in the 1974 CD Market,” Journal
of Bank Research, vol. 7 (Autumn 1976),
pp. 213-224.
Cramer, Robert H. and Robert J. Rogowski.
“Risk Premia on Negotiable Certificates
of Deposit and the Continental Illinois
Bank Crisis,” paper presented to the Fi­
nancial Management Association, Octo­
ber, 1985.
Eisenbeis, Robert A. and Gary Gilbert.
“Market Discipline and the Prevention of
Bank Problems and Failures,” Issues in
Bank Regulation, vol. 8 (Winter 1985), pp.
16-23.
Goldberg, Michael A. and Peter R. LloydDavies. “Standby Letters of Credit: Are
Banks Overextending Themselves?” Jour­
nal of Bank Research, vol. 16 (Spring 1985)
29-39.
Federal Reserve Bank of Chicago




Hanweck, Gerald and Timothy Hannan.
“Bank Insolvency Risk and the Market
for Large Certificates of Deposits,” Board
of Governors of the Federal Reserve Sys­
tem, mimeo 1985.
Herzig-Marx, Chayim and Anne S. Weaver.
“Bank Soundness and the Market for
Large Negotiable Certificates of
Deposit,” Federal Reserve Bank of
Chicago Research Paper 79-1.
Kaufman, George G. “Implications of Large
Bank Problems and Insolvencies for the
Banking Industry and Economic Policy,”
Issues in Bank Regulation, vol. 8 (Winter
1985), pp. 16-23.
Merton, Robert. “On the Pricing of Corporate
Debt: The Risk Structure of Interest
Rates,” Journal of Finance, vol. 29 (May
1974), pp. 449-470.
31

BU LK RATE

E (G3HKQMBC) economic
PERSPECTIVES
Public Information Center
Federal Reserve Bank
of Chicago
P.O. Box 834
Chicago, Illinois 60690
Do Not Forward
Address Correction Requested
Return Postage Guaranteed




U.S. P O ST A G E
PAID
C H IC A G O , ILLIN O IS
PERMIT NO. 1942