Full text of Review (Federal Reserve Bank of Dallas) : July 1981

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

I

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

llR. ALURDIC&-

July 1981
1

Uncertainty Costs of High Inflation

11

"Fed Quotes"

12

Regulatory Briefs and Announcements

15

Now Available from the Federal Reserve

This publication was digitized and made available by the Federal Reserve Bank of Dallas' Historical Library (FedHistory@dal.frb.org)

Uncertainty Costs
of High Inflation
By Deborah A. Frohman, Leroy O. Laney, and Thomas D. Willett*

Is reducing inflation really worth what it costs?
There is little disagreement among economists
and policymakers that the rate of increase in the
general price level, which began to rise markedly
about 15 years ago, is a very serious economic
problem. Reducing it has been accorded top priority by those formulating U.S. policy, including the
Federal Reserve. But policy pronouncements are
almost always accompanied by warnings that
bringing inflation down to a more acceptable level
can be costly and painful in the intermediate run.
During a transition, while expectations adjust,
there will be substantial lost output and high unemployment. Moreover, the climb in interest rates
to the levels experienced recently in the United
States tests even the resolve of some of those most
convinced of the crucial role of tighter monetary
policy in the process of reducing inflation.
Even if there is widespread agreement that inflation is too high, the high interest rates and unemployment accompanying the adoption of restrictive macroeconomic policy are obviously a source
of considerable dissatisfaction. This, in turn, can
foster an opinion that the cure is worse than the
disease.
Estimates from some large macroeconometric
July 1981/Voice

models suggest it can take many years of high unemployment and reduced output to cure inflation.
Such estimates have led some to argue that learning to live with the current level of inJlation would
be preferred to eliminating it. However, recent developments in the analysis of expectations have
suggested that these results may substantially overstate the length of time and consequent unemployment costs required to cure inflation.!

*Frohman and Laney are research associate and senior
economist, respectively, at the Federal Reserve Bank of
Dallas. Willett is Horton Professor of Economics at
Claremont Graduate School and Claremont Men's College,
California. His contribution was supported in part by
grants from Texaco and the Lincoln Foundation.
1. For a discussion and comparison of the results for
various models. see Laurence H. Meyer and Robert H.
Rasche, "On the Costs and Benefits of Anti-InIlatioD
Policies," Review, Federal Reserve Bank of St. Louis,
February 1980. A number of useful contributions on the
role of expectations in the inflation reduction process
are found in the Journal of Money, Credit, and Banking,
ovember 1980, pI. 2, a special issue containing the
papers and proceedings of a seminar on rational expectations that was sponsored by the American Enterprise
Institute for Public Policy Research.
1

Furthermore, even if the time frame for eliminating inflation is uncomfortably long, many analyses
have not fully recognized inflation's costs. In particular, continued high rates of inflation are
unlikely to be steady or predictable. Living with
inflation thus involves substantial uncertainty that
itself depresses employment and worsens overall
economic performance. Recognition of these uncertainty costs of inflation strengthens the case for
adoption of serious anti-inflation policies despite
the considerable transitional pain that is likely to
be involved. The following article addresses this
issue.

Obsolescence of the Phillips Curve
as a guide to policy
The menu of policy choices in earlier years presumed a short-run trade-off between inflation and
unemployment, embodied in the traditional Phillips
Curve. However, it has been recognized increasingly that thinking in terms of a negative relationship between these two variables can be a
misleading guide to policy. Only in the short run,
as the economy adjusts to unanticipated changes
in economic variables, will a negative relationship
hold. Over the long run, output and employment
are likely to be lower because of inflation than
they would be otherwise.
Many analyses have overlooked this positive
long-run relationship and, consequently, underestimated the costs of high inflation. It is true that if
inflation were held at a perfectly steady and predictable level, even one that was high by historical
standards, lost economic efficiency and income redistribution costs might be relatively low. Costs
of changing prices frequently would remain, but
inflation could be accurately incorporated into contracts and loans. Distortions would still exist because of difficulties in indexing everything, but
these might not be great. A major uncorrectable
cost would be an "inflation tax" on non-interestbearing money balances, but even this could be
relatively low. 2 The real problem is that inflation
is actually not steady or predictable.

The costs of uncertainty about inflation
In recent years greater recognition of the variability of inflation has developed, along with an
awareness that accompanying uncertainty can be
quite damaging to the economy.3 As inflation be2

comes more variable and future price level movements become more uncertain, more resources must
be devoted to attempts at forecasting general inflation. Even if these efforts are successful, some
opportunity cost is involved, but it is quite likely
that forecast errors will increase with more variable inflation. This means additional inefficient
allocation of resources as wrong decisions are
made more often. The role played by relative prices
in the efficient allocation of resources is impaired,
and the usefulness to economic planning of the
signals from the price system is reduced. The information content of these signals is less because
of uncertainty about whether a given price movement reflects a change in relative prices or general
inflation.
Furthermore, the very prospect of increased uncertainty about price level developments is likely
to depress economic activity directly. Because it is
feared that a wrong decision will be made, participants in the economy may simply opt for doing
nothing in some cases. For any given level of expected return on an undertaking, economic activity
is riskier. The planning horizon is also shorter.
Hence, the level of activity by normally risk2. It is even possible that for some levels of steady
inflation, the rate of increase in prices would be below
an "optimum" inflation tax. Instead of striving for a
negative rate of inflation that would compensate for the
opportunity cost of holding non-interest-bearing money
balances-as suggested by Milton Friedman, The
Optimum Quantity of Money and Other Essays (Chicago:
Aldine Publishing Company, 1969), chap. 1-the optimum
inflation tax would equate the marginal excess burden
from the inflation tax with that on other forms of taxation. The optimum rate of inflation on these grounds
will be positive. For discussion, see Edward Tower,
"More on the Welfare Cost of Inflationary Finance,"
Journal of Money, Credit, and Banking, November 1971,
and Robert J. Gordon, "The Demand for and Supply of
Inflation," Journal of Law and Economics, December 1975.
3. More variable inflation would not necessarily imply
greater uncertainty if it followed a highly predictable
pattern, but this is unlikely to be the case. It is practical
to view variability of inflation and uncertainty about
inflation as having a strong positive relationship. This
relationship will be considered here in more detail later.
4. See, for example, Donald J. Mullineaux, "Unemployment, Industrial Production, and Inflation Uncertainty in
the United States," Review of Economics and Statistics,
May 1980, and Maurice D. Levi and John H. Makin,
"Inflation Uncertainty and the Phillips Curve: Some
Empirical Evidence," American Economic Review,
December 1980.

Federal Reserve Bank of Dallsl

averse economic agents is reduced. Recent studies
have found that greater uncertainty about price
level developments in the United States has had
significant effects in raising unemployment and
lowering industrial production. 4 Such uncertainty can also raise interest rates and depress
investment. ~
Association of higher inflation levels
with greater uncertainty about inflation

Of course, these costs of more uncertainty about
inflation can be used to support a case for bearing
the pain of reducing inflation only if lower inflation
rates are likely to be associated with more certain
price level developments. If uncertainty is independent of the level of inflation, restoration of
lower average inflation rates will do nothing to
reduce this uncertainty over the longer run.
It seems likely, however, that at higher rates of
inflation, price level developments will become
more uncertain. The effect of this uncertainty will
be more important as estimates of future inflation
become more crucial in determining profitability
of activities relative to real economic factors. 6 In

5. For the effect of inflation uncertainty on interest rates,
see Amir Barnea, Amihud Dotan, and Josef Lakonishok,
"The Effect of Price Level Uncertainty on the Determination of Nominal Interest Rates: Some Empirical
Evidence," Southern Economic fournal, October 1979.
For effects on investment, see Stephen 1. Able, "Inflation Uncertainty, Investment Spending, and Fiscal
Policy," Economic Review, Federal Reserve Bank of
Kansas City, February 1980, and Gary D. Praetzel, "How
Inflation Has Affected Investment Decisions," Voice of
the Federal Reserve Bank of Dallas, June 1981.
6. For example, a real rate of return of 3 percent on an
undertaking would account for 60 percent of a 5-percent
nominal return in a world in which expected inflation is
2 percent. But if expected inflation is 9 percent and the
nominal return is 12 percent, the real return of 3 percent
only constitutes 25 percent of the total.
7. In Mohsin S. Khan, "The Variability of Expectations
in Hyperinflations," Journal of Political Economy, August
1977, it is demonstrated that in hyperinflationary situations at least, the higher the level and the greater the
change in inflation, the faster expectations are revised.
8. For discussions on these points, see William Fellner,
Towards a Reconstruction of Macroeconomics: Problems
of Theory and Policy (Washington, D.C.: American Enterprise Institute for Public Policy Research, 1976), and
Arthur M. Okun, "The Mirage of Steady Inflation,"
Brookings Papers on Economic Activity, 1971, no. 2.

July 1981/Voice

a higher-inflation world, psychological imponderables may become increasingly important relative to real economic factors, and significant shifts
in expectations may become larger and more frequent. 1 For any given set of underlying circumstances, there would likely be greater variability in
the behavior of the private sector.
Moreover, it would seem much more difficult to
establish the credibility of steady macroeconomic
policies at high inflation levels than at low ones.
Persistent short-run accommodation of economic
shocks is likely to lead to intermittent upward
ratchets of inflation rates over time. Inconsistent
macroeconomic policies driven by short-run political pressures reinforce tendencies for private sector expectations and behavior to become more
variable at higher inflation rates. 8
While there is a need for more theoretical work
in this area, these arguments present a strong case
for expecting a significant positive relationship between the rate and the variability of inflation, and
such a relationship has been used to explain the
currently observed correspondence between high
inflation and high unemployment. 9 Based on previously published studies. however. it is not entirely clear that such a relationship does indeed
hold for the United States, Empirical work has
indicated that, on average, countries with higher
inflation also have more variable inflation. This
has not held for all time periods, though, nor for
some groups of countries when the inflation rate
falls in a certain range, Most of the previous empirical work has examined cross-sectional data.
comparing average relationships across countries. 10
9. See Milton Friedman, "Nobel Lecture: Inflation and
Unemployment," Journal of Political Economy, June 1977.
(Okun, "Mirage of Steady Inflation," suggested an opposite view that more variable inflation will result from
a negatively sloped, static Phillips Curve trade-off
between inflation and unemployment if the economy is
operating in the steeper, higher-inflation segment of the
curve. Deborah Frohman, "The Relationship Between the
Rate and Variability of Inflation," senior thesis, Claremont
Men's College, August 1980, does not find empirical
support for this explanation, presumably because of the
shiftability of inflation-unemployment relationships.)
10. For empirical contributions analyzing the crosssectional relationship between the rate and variability
of inflation, see: Okun, "Mirage of Steady Inflation";
Robert J. Gordon, "Steady Anticipated Inflation: Mirage
or Oasis?" Brookings Papers on Economic Activity, 1971,
no. 2; Dennis E. Logue and Thomas D. Willett, "A Note on

s

A time series versus a cross-sectional approach
and proxies for uncertainty

Ideally, we would like to know not so much the
relationships across countries as the relationship
for individual countries over time as their rates
of inflation vary. This kind of analysis was unwieldy in earlier work studying the relationships
between the average level of inflation and its
standard deviation, because the number of observations needed to calculate standard deviations
was relatively large. l1 The use of other proxies for
uncertainty does allow the study of these relationships for individual countries over time, and that
is the approach taken here.
In using this approach, the first step is to
choose a time series proxy for uncertainty about
inflation. At the outset, this raises the question of
whether higher observed variability in the inflation
the Relation Between the Rate and Variability of Inflation," Economica, May 1976; Dwight Jaffee and
Ephraim Kleiman, "The Welfare Implications of Uneven
Inflation," in Inflation Theory and Anti-inflation Policy,
proceedings of a conference held by the International
Economic Association at Saltsjobaden, Sweden, edt
Erik Lundberg (Boulder, Colo.: Westview Press, 1977);
and Edward Foster, "The Variability of Inflation,"
Review of Economics and Statistics. August 1978.
Among these, Okun and Gordon simply examined crosssectional scatter diagrams of the data. Okun found a
strong positive correspondence between high average
inflation and the standard deviation of inflation over
1951-68 for 17 industrial countries, but Gordon argued
that the relationship existed only for the first part (195159) of the interval Okun examined. Jaffee and Kleiman
tested the relationship with regression techniques; they
found a positive relationship for both subperiods but a
significant relationship statistically only for the earlier
period. They also examined Latin American countries
and found a positive relationship cross-sectionally.
Logue and Willett examined 41 countries crosssectionally over the 1949-70 period. They analyzed different subperiods and groups of countries and used both
standard deviations and naive forecast errors to gauge
uncertainty. Results were generally strong and positive;
however, countries with relatively low rates of inflation
showed no significant relationship between the inflation
rate and its standard deviation, and highly industrialized
countries demonstrated a weak negative relationship.
Foster, who suggests using the average absolute yearto-year change in inflation (rather than standard deviation) as the measure of cross-sectional variability,
examined 23 advanced economies and 17 Latin American
economies over 1954-75 and subperiods thereof. Foster's
results confirm a strong positive relationship between the
rate and variability of inflation.
4

rate necessarily indicates greater uncertainty. Although they are generally assumed to be closely
associated, highly variable inflation need not lead
to greater uncertainty. If economic agents base
their expectations on variables other than past
inflation and these variables are predictable, high
inflation variability can be accompanied by low
uncertainty. This seems highly unlikely to occur
frequently in practice, however, and recent empirical work supports the case that forecast uncertainty does tend to increase with the variability of
inflation. 12 Thus, the assumption that higher variability of inflation is associated with greater uncertainty seems quite plausible.
The most commonly used proxy for inflation
uncertainty is either the standard deviation of
the inflation rate in cross-sectional work or some
form of moving standard deviation in an individualcountry time series approach. However, since a
systematically changing, and therefore predictable,
inflation rate can have as high a standard deviation as one that fluctuates randomly, a standard

11. In one of the few published studies to date attempting to measure the possibility of a relationship between
the rate and variability of inflation over time, "Our
New Monetary Standard: The Measurement and Effects
of Price Uncertainty, 1880-1973," Economic Inquiry.
December 1975, Benjamin Klein contends (using a six-term
moving average of annual differences in the natural log
of the GNP deflator as the measure of expected inflation
and a six-term moving standard deviation of the same
series as the measure of variability) that there has not
been the positive relationship between the rate and
variability of inflation postulated by other analysts. Klein
argues that his mean measure was high by U.S. historical
standards over the last decade of his analysis while the
variability measure was low over the same period. A
relationship between the rate and variability of inflation
may exist across countries, Klein concludes, but does not
seem to have existed recently for the United States.
12. For example, Alex Cukierman and Paul Wachtel,
"Differential Inflationary Expectations and the Variability
of the Rate of Inflation: Theory and Evidence," American
Economic Review, September 1979, using a moving standard deviation in order to measure inflation variability
and using the dispersion among respondents' inflation
forecasts in surveys by the University of Michigan Survey Research Center and by Livingston (published in the
Philadelphia Inquirer) in order to measure uncertainty,
find a positive correlation for the 1948-75 period. Douglas
W. Mitchell, "Determinants of Inflation Uncertainty,"
Eastern Economic Journal, April 1981, provides evidence
that uncertainty about inflation is related to both its
variability and the level of inflation expectations.
Federal Reserve Bank of Dallas

Annual Rates of Change in U.S. Consumer Price Index and GNP Price Deflator
12 PERCENT CHANGE IN INDEX L E V E L - - - - - - - - - - - - - - - - - - - - - - - CONSUMER PRICE INDEX

10 -

8-

6-

4-

2-

O-~rd---------------------------------

- 2 --'jr--T""""""'"T"""--r--.-rj--.-r-.,.....-r---rj---,r--T"""""""T'"""--,---,jr--r-..."""'T'......,-rj......,-r--r-. .--rj-

1954

1959

1964

1969

1974

1979

SOURCES: U.S. Department of Commerce.
U.S. Department of Labor.

July 1981/Voice

5

Table 1
CORRESPONDENCE BETWEEN FORECAST ERROR
AND U.S. INFLATION RATE, 1954-79

(IPt - Pt-11 =

80

+

8 1 Pt-1)
Pt-l

R'

SEE

OW

.26

1.03

1.89

.20
(3.84)* •

.32

.87

2.01

.51
(2.01)

.16
(2.32)*

.19

.75

2.05

.11
(.59)

.27
(5.08)··

.56

.63

2.46

Price level measure

Constant

Equation 1 ...

Consumer price index

.55
(1.70)

(3.10)* *

Equation 2 .••

GNP price deflator

.27
(1.16)

Equation 3 ..•

Consumer price Index

Equation 4 .••

GNP price deflator

coefflclant

.22

Rho

-.45

-.45

NOTE: Figures In parentheses are t statistics; • Indicates significance of the Independent variable at the 95-perceAt level, using a
slngle·tall test that the variable is signed as hypothesized, and" Indicates significance at the 99-percent level.
R' Is the correlation coefficient adjusted for degrees of treedom. SEE is the standard error of the equation. OW Is the
Durbin·Watson autocorrelation test statistic. Rho Is the flrst·order autocorrelation coefficient when a Cochrane·Orcutt
procedure was used to correct serial correlation ot residuals.

Table 2
RESULTS USING ALTERNATIVE MEASURES OF UNCERTAINTY
AND EXPECTATIONS ABOUT INFLATION, 1954·76
(Dependent variable

= Carlson's standard deviation of Livingston forecasts)
Carlson's
mean
expected
Inflation

Constant

Equation 5 . . .

.60
(3.05)

Equation 6 . . .

.56
(2.39)

.10
(2.49)·
.15
(2.23)·

SEE

OW

Rho

.41

.34

1.70

.49

.38

.35

1.65

.51

NOTE: Figures In parentheses are t statistics; • Indicates slgnltlcance ot the Independent variable at the 95percent level, using a single-tail tesl that the variable Is signed as hypothesized.
R' Is the correlalion coeltlcient adjusted for degrees of freedom. SEE Is the standard error of the
equation. OW is the Durbin·Watson autocorrelation tesl statistic. Rho is the firsl·order autocorrelation
coefficient when a Cochrane-Orcull procedure was used to correcl serial correlation of residuals.

6

Federal Reserve Bank of DaDas

deviation can be a poor measure of variability and
uncertainty. The standard deviation computed
from a series of observations is independent of
the serial pattern of those observations. Furthermore, a rising rate of inflation over time will generate a spurious positive association between the
average rate of inflation and its standard deviation,
and a declining trend will cause a downward bias.
These considerations argue against use of this measure. Empirical work to follow uses a simple forecast error method to gauge uncertainty, which overcomes problems in using a standard deviation. 13
Empirical tests and results
In the first test here, the difference between the
actual inflation rate for a period and the value that
was predicted for the period is taken as the proxy
for uncertainty. The assumption is that the higher
this forecast error, the greater is the economic
agent's subjective distribution of possible future
inflation rates. Expectations about inflation are
assumed here to be generated by a simple adaptive
expectations process, in which all inflation forecasts project no change in the inflation rate from
the previous year. Such expectations can be
proxied in a number of ways, of course. For example, prior inflation rates over time can be analyzed for any systematic movement. u
The absolute value of the difference in inflation
rates is used, since the only concern here is the
magnitude of the difference. This error measure is
regressed against the earlier year's inflation rate,
representing the level of inflation in the year with
which the error was associated historically. In
equation form:
Ipt - pt-ll = 00
01 pt-l,
where pt-l is the rate of inflation in year t -1 and
Ipt - pt-ll is the error between the inflation forecast for period t, Pt-l, and the actual value in period
t, pt. 15
This model was estimated using annual rates of
change in both the consumer price index and the
GNP (gross national product) price deflator for
1954 through 1979, an interval beginning just after
the Korean War inflation and ending in the latest
year for which data were available when the tests
were conducted. Results are reported in Table 1.
Equations 1 and 2 were run over the entire interval
for inflation rates of the consumer price index and
the GNP deflator, respectively. Coefficients on the

+

July 190t/Voice

pt-l variable are both significant statistically, and
they are close in value. 16
In equations 3 and 4, the same regressions were
run with certain years excluded-those in which,
it might be argued, the variability of inflation was
due to special factors rather than the general
operation of the inflationary process. The years
1971 and 1972 were exceptional with respect to
inflationary expectations because of the constraining influence of wage and price controls in the
United States; then, when controls were removed
and pent-up pressure was released, expectations
may have been greater that prices would surge.
Certainly, 1974 and 1979 were exceptional. The
oil-related shocks then were unusual compared
with previous experience and were unanticipated.
Omitting these observations (which also requires
exclusion of the following year in each case because of the way the equation is specified) yields
one comparison with coefficients in equations 1 and
2. The Pt-l coefficients in the second two equations

13. For illustration of problems in using the standard
deviation method, see Foster, "Variability of Inflation."
14. Several types of forecasting models, including autoregressive integrated moving average (ARIMA) models,
have been used with this method of measuring
uncertainty. See, for example, I. B. Ibrahim and Raburn
Williams, "Price Unpredictability and Monetary Standards: A Comment on Klein's Measure of Price Uncertainty," and Benjamin Klein, "The Measurement of
Long- and Short-Term Price Uncertainty: A Moving
Regression Time Series Analysis," both in Economic
Inquiry, July 1978, and Frohman, "Relationship Between
the Rate and Variability of Inflation."
15. One might argue that it is just as appropriate to
associate the absolute error term with p. as the independent variable rather than p.-,. This would compare
the previous year's forecast error with the inflation level
in the current year, but near the end of the period at
least, both variables would be known to economic agents.
Equations were estimated using both specifications.
Results using Pt as the independent variable do not contradict findings using p.-,. In general, the outcomes were
quite similar.
16. Equations using Pt as the independent variable rather
than Pt-" corresponding to equations 1 and 2 in Table 1,
are the following (t statistics in parentheses):
(la) Ip. - pt -,I
.27 + .27 p.
(1.03) (5.12)
R" .50; SEE .84; DW 2.09.
and
(2a) Ip.- p.-,I
.29 .18 p.
(1.18) (3.63)
R" .28; SEE .89; DW 1.96; rho
.46.

=

=

=

=

=

=

+

=

=

=-

7

are fairly close to those in the first two. The coefficient falls somewhat using the consumer price
index and rises using the GNP deflator. Presentation of equations 3 and 4 provides a check on the
stability and significance of coefficients over the
interval investigated. 17
In addition to the naive forecast error model, a
model using direct survey data was estimated.
Data from the original source, the Livingston survey of consumer price inflation forecasts of professional economists that has been published
periodically in the Philadelphia Inquirer, were recently revised (through 1976) by Carlson. 18 The
standard deviation of these revised data provides
a time series of a directly observed measure of
uncertainty about inflation, on the assumption that
a wider dispersion of individual forecasts is positively associated with greater uncertainty in each
forecaster's mind.
This standard deviation was first regressed
against the inflation rate for the year in which the
forecast was made. (For example, the standard
deviation of the forecasts for inflation in 1970 that
were made in 1969 was regressed against the actual
inflation rate in 1969.) Results for the same interval
as before, shown as equation 5 in Table 2, indicate
the inflation rate is statistically significant.
Furthermore, the standard deviation of these expectations was regressed against the mean of the
expectations to test whether higher expectations
are associated with higher uncertainty. This mean
of expectations also demonstrates significant explanatory power with respect to the standard

17. Among other modifications of the basic interval
analyzed were rolling regressions-simply adding and! or
subtracting data points from the 1954-79 sample to test
for stability and significance of coefficients over time.
The existence of threshold effects was also investigated.
(For example, below or above some threshold level the
relationship between the forecast error and the level of
inflation might change.) These modifications also produced some variation in results, but perhaps understandably so since degrees of freedom were often substantially
diminished. Usually the coefficient on the independent
variable did indicate a positive association with the
forecast error term, and in many cases results reinforced
those presented in Table 1. For threshold results in
investigating differential effects above 3-percent inflation,
for example, the p. -1 coefficient for the consumer price
index was .23, quite close to that in equation 1.
16. See John A. Carlson, "A Study of Price Forecasts,"
Annals of Economic and Social Measurement. Winter 1977.
6

deviation in equation 6. 19
Although these investigations focus on forecast
errors and expectations formation over the relatively short time frame of up to one year, it is also
interesting to look at a somewhat longer period
and assume a more extended process of expectations formation. Such a process was not the explicit focal point of the present analysis but can
well be the subject of future experimentation. One
investigation worth mentioning, however, substituted for Pt-l in the first naive forecast error equation an average of inflation over the previous four
years to proxy actual and expected inflation. While
the outcomes were similar to equations 1 and 2
in Table 1, the results were not as significant. In
some equations, a term was added to measure
inflation's deviation from trend, but this term was
usually insignificant statistically. 20
19. Dividing the interval examined in equations 5 and 6
into two subperiods-breaking the interval at 1965-does
indicate a shift in the relationship, however. The early
subperiod shows only very weak statistical significance
for the actual inflation rate and none for the mean of
the inflationary expectations variable. But in the latter
subperiod, t statistics for both variables rise markedly
(to 4.14 and 5.00, respectively), even though the coefficients themselves change only slightly from those
computed for the entire interval in equations in the
table (to .11 from .10 and to .19 from .15).
20. Another avenue in the area of a longer time frame
was explored using the analysis of Klein, "Our New
Monetary Standard," as a springboard. Regressing a
six-term moving standard deviation of the annual inflation
rate, as the dependent variable proxying uncertainty, on
a six-term moving average of the rate in several equations
over the post-World War II interval (Klein compares
graphically over the 1860-1973 period the variables
regressed in our study over modern times only), a highly
significant positive correspondence was usually found for
both measures of the price level. For the 1958-79
interval-several observations are lost in computing
moving averages and standard deviations-results for
the inflation rates of the price index and the deflator,
respectively, were:
I1ma.
.24
.30 E ..a.
(1.03) (7.21)
ji" .94; SEE .17; DW 1.95; rho .72.
and
I1ma.
.22
.27 Ema.,
(2.37) (13.40)
ji" .96; SEE .12; DW 1.66; rho .41.
where 11 mao is a six-year moving standard deviation and
Ema. is a six-year moving average of inflation. For conclusions reached by Klein for the postwar period, see
footnote 11.

=
=

+

=

+

=

=

=

=

=

=

=

Federal Reserve Bank of Dallas

Measuring the gains from lowering inflation
If uncertainty about inflation is positively associated with its level, then reducing inflation should
also reduce uncertainty, stimulate employment and
investment, and raise output. Is it possible to
quantify this? To do so with any great confidence
is difficult. Recently published studies, however,
do allow some rough estimates. Neither the linkage between inflation and uncertainty nor that
between uncertainty and economic performance is
likely to be precisely estimated or highly stable,
but the following calculations are of interest.
Mullineaux, who also uses the standard deviation from the Carlson data in measuring the impact
of inflation variability on unemployment, presents
results suggesting that a unitary increase in this
standard deviation produces a 1- to 2-percentagepoint increase in unemployment over a two-year
period. 21 If the coefficient estimated in our equation 5 is any guide to the relationship between the
existing inflation rate and the standard deviation
of inflation forecasts, then bringing inflation down
about 10 percent could result in a unitary reduction
in the standard deviation, lowering unemployment
by 1 to 2 percentage points.
Levi and Makin have estimated the impact of
inflation uncertainty, measured in this fashion, on
the percentage change in total nonagricultural employment. They find that over the 1965-75 period,
a unitary increase in the standard deviation of inflation forecasts lowers employment by 2.25 percent. 22 Likewise, if a 10-percent reduction in infla-

July 1981/Voice

tion caused a unitary reduction in the standard
deviation, employment could rise by 2.25 percent.
Therefore there is considerable scope for substantial g~ins from measurable progress against
inflation.
Conclusion

This article has provided empirical support for
the existence of a positive relationship between
inflation and economic uncertainty in the United
States. Since this uncertainty can reduce economic
activity and misallocate resources in the economy,
the incentive for lowering inflation is clear.
We have not been able to isolate a tight enough
correspondence between inflation and uncertainty
to be able to say with great accuracy what the
reduction in uncertainty would be for a marginal
decline in inflation of, say, 2 or 3 percentage points.
And for the ultimate impact on the economy, we
also must await more conclusive evidence on how
much effect the reduced uncertainty itself has in
increasing real economic variables. If the long-.run
gains from reducing inflation are balanced agamst
the transitional costs, however, a strong case can
be made for bearing the costs in order to restore
a noninflationary environment.
21. See "Unemployment, Industrial Production. and
Inflation Uncertainty," p. 166.
22. See "Inflation Uncertainty and the Phillips Curve,"
p.1025.

9

New Member Banks

Northway National Bank, Addison, Texa<;, a newly organized institution
located in the territory served by the Head Office of the Federal Reserve
Bank of Dallas, opened for business June 1, 1981, as a member of the Federal
Reserve System. The new member bank opened with capital of $1,500,000
and surplus of $1,500,000. The officers are: Charles A. Richardson, Chairman
of the Board; David M. Bernardin, President; Joe Key, Senior Vice President;
and Vickie Edgar, Cashier.
Citizens National Bank-West, Houston, Texas, a newly organized institution
located in the territory served by the Houston Branch of the Federal Reserve
Bank of Dallas, opened for business June 1, 1981, as a member of the
Federal Reserve System. The new member bank opened with capital of
$1,000,000 and surplus of $1,000,000. The officers are: W. Phillip Johnson,
Jr., Chairman of the Board; Jimmy G. Cox, President; and Michael L. Burnett,
Vice President and Cashier.
Frontier National Bank, Round Rock, Texas, a newly organized institution
located in the territory served by the San Antonio Branch of the Federal
Reserve Bank of Dallas, opened for business June 1, 1981, as a member of
the Federal Reserve System. The new member bank opened with capital of
$750,000 and surplus of $750,000. The officers are: Mark J. Silverstone,
Chairman of the Board; Bob E. Lively, President; and Gary P. Bowles,
Cashier.
RepublicBank Post Oak, N.A., Houston, Texas, a newly organized institution located in the territory served by the Houston Branch of the Federal
Reserve Bank of Dallas, opened for business June 8, 1981, as a member of
the Federal Reserve System. The new member bank opened with capital of
$1,250,000 and surplus of $1,250,000. The officers are: Joe M. Bridges,
Chairman of the Board; Bob Pizzitola, President and Chief Executive Officer;
Bob Shaw, Vice President; Bonnie Basham, Cashier; and Karin Andrews,
Administrative Officer.
Metropolitan National Bank, Farmers Branch, Texas, a newly organized institution located in the territory served by the Head Office of the Federal
Reserve Bank of Dallas, opened for business June 15, 1981, as a member of
the Federal Reserve System. The new member bank opened with capital of
$1,500,000 and surplus of $1,500,000. The officers are: James A. Moran,
Chairman of the Board; Roland Walden, President; Wendell C. Howie, Vice
President and Cashier; Loretta S. Phillips, Assistant Vice President; and
Sallie Lundy, Assistant Cashier.

10

Federal Reserve Bank of Dallas

Ncp'ed Quotes~
Brief Excerpts from Recent Federal Reserve Speeches, Statements, Publications, Etc.

"The habit of taking a long-run view of things, I believe, has much diminished in
recent years. This seems apparent from the way in which we have allowed the growth
of our economy to lag for the sake of immediate satisfactions. VVe have increased
consumption and reduced savings and therewith investment in future productivity.
We have been hesitant to come to grips with the problem of inflation which beclouds
our future because of the fear of temporary pain from the actions needed. We have
treated ourselves to 'entitlements' of all sorts without providing the means of paying
for them. And in our business functions, we have laid ourselves open to the charge
that our habit of focusing on the profits of the next quarter makes us vulnerable to
being overtaken by more long-run oriented competitors like the Japanese."
"We already see many of the consequences of past neglect-in the form of our
inflation, declining productivity, and uncertainty about future supplies of oil and other
resources. We see them in the way we are being overtaken economically by other
countries and in the way in which our influence in the world has diminished. We are
now experiencing the long-run consequences of the many short-run expediencies
resorted to in the past. Having too often chosen the easy way, all future options are
becoming harder and less inviting.
"I do not believe that this must be a lasting condition. We can learn from past
mistakes. American society, I believe, is particularly capable of that kind of
learning.... It is based on the ability to analyze, to see current problems in the light
of general principles, the ability to take a long view....
"If we are willing to take that kind of a look at where we have been, where we
are, and where we should be going, we shall be able to make progress in many
directions. We can end our inflation, raise the output of our economy, and make our
future more secure. We can improve our society and the quality of our life at home and
strengthen our role abroad and contribute to the peace and progress of the world.
These gains will not always come cheaply. They will require sacrifices in the short
run. But, over the years, this investment will payoff. Indeed, if we do not make the
sacrifices and invest in our future, then the ultimate cost in terms of living standards
and quality of life will be far higher."
Henry C. Wallich, Member, Board of Governors of the
Federal Reserve System (At the Commencement
Exercises. Washington College, Chestertown,
Maryland, May 17, 1981)

July 1981/Voice

11

GRegulatoryGJ3riefs
andcfinnouncements
Regulations T and U:
Board Adopts Amendment,
Publishes Proposed Amendments
The Board of Governors of the Federal Reserve
System has recently amended Regulation T (Credit
by Brokers and Dealers) and has published for
comment proposed amendments to that regulation
and Regulation U (Credit by Banks for the Purpose
of Purchasing or Carrying Margin Stocks).
The adopted amendment deleted the paragraph
permitting the use of foreign currency in a margin
account. Deletion of this paragraph in Regulation
T is intended to make it clear that speculative
holding of foreign currency in a margin account
is not permissible and that any transactions in foreign currency should be effected in accounts
insulated from securities credit transactions.
The first two proposed amendments to Regulation T concern margin requirements for trading of
options on government and government agency
debt issues. The Securities and Exchange Commission recently approved trading on the Chicago
Board Options Exchange in option contracts on
Government National Mortgage Association
(GNMA) securities. The New York Stock Exchange,
the American Stock Exchange, and the Chicago
Board Options Exchange have proposed trading in
option contracts on Treasury bills, notes, and
bonds. The proposed amendments would affect
margin requirements in option contracts on such
securities.
Under one proposed amendment, brokers and
dealers would be allowed to give "good faith"
loan value to an option that has been purchased,
and a "good faith" margin would be permitted
when an option contract is written. The maintenance margin requirements that the individual
12

securities exchanges set for their members, subject
to review by the Securities and Exchange Commission, would be expected to apply.
The alternative proposed amendment would set
a uniform margin requirement of 130 percent of
the option premium plus $1,000 for the initial
writing of all uncovered option contracts on
exempt debt securities. Under this amendment, no
option contract would be permitted to have loan
value. Exchanges would continue to be free to set
maintenance margin requirements. Under either
proposal, no margin would be required where the
option is covered as specified by the Board.
In the absence of an amendment, the Board's
current margin rules relating to options on corporate equity securities would apply to options on
government debt issues. The current margin
requirement-30 percent of the value of the underlying security, with additional adjustment for
unrealized losses and gains-would impair the
usefulness of these contract markets, since the
underlying securities can be purchased, under current industry practice, on a much lower margin.
The Board has asked specifically for comments
regarding the likelihood that with the "good faith"
approach, the exchanges would set different margin requirements on the same or similar option
contracts; the denial of any loan value to a long
contract; Board specification of an initial margin
requirement; and the appropriateness of the
Board's proposed margin for options on all exempt
debt securities.
Subsequent proposed amendments concern both
Regulation T and Regulation U. These proposed
Federal Reserve Bank of Dallas

amendments are the first steps in a review by the
Board intended to simplify and modernize all Federal Reserve margin regulations, reducing the burden of regulatory compliance wherever possible.
For Regulation T the Board proposed to eliminate "equity building" devices; consolidate the
bond accounts with the General Account; and
require, in certain circumstances, an offsetting adjustment to any highly leveraged General Account
by transfers from the customer's Special Miscellaneous Account. The Board also proposed to relax
the restriction on the arranging of credit by investment bankers to permit investment banking
services that may otherwise be prohibited.
For Regulation U the Board proposed to change
the collateral test so as to exempt from quantitative limitation all bank credit not secured by
margin equity securities.
Comments on the first two proposed amendments must be received by August 3, 1981, and
should refer to Docket No. R-0082. Comments on
the other proposed amendments must be received
by September 15,1981, and should refer to Docket
No. R-0362. All should be submitted to the Secretary, Board of Governors of the Federal Reserve
System, 20th Street and Constitution Avenue,
N.W., Washington, D.C. 20551.

July 1981/Voice

Banking Institutions Allowed
to Establish International
Banking Facilities
The Federal Reserve Board has amended its regulations regarding reserve requirements and payment
of interest on deposits to permit the establishment
of international banking facilities (IBFs) in the
United States. It is believed that the establishment of IBFs at U.S. banking offices will enhance
the international competitive position of banking
institutions in this country.
Subject to conditions specified by the Board,
IBFs may be established by U.S. depository institutions, Edge Act and'Agreement corporations, and
U.S. branches and agencies of foreign banks. In
general, an IBF may accept deposits from and extend credit to foreign residents or other IBFs.
Regulation D reserve requirements and Regulation
Q interest rate limitations are not applicable to
such funds.
The Board made its action effective December 3,
1981, in order to give all interested banking institutions time to make necessary arrangements for
implementation of IBFs.
The Board has made a general statement of
policy regarding the use of IBF deposits and IBF
loans. The policy states, in part, that with respect
to nonbank customers located outside the United
States, the Board expects IBFs to accept only
deposits that support a customer's operations outside the United States and to extend credit only
to finance the customer's non-U.S. operations.
13

Moreover, deposits should not be used as a means
of circumventing interest rate restrictions or reserve
requirements. The Board specifies that this policy
be communicated in writing to IBF nonbank customers when a credit or deposit relationship is
initially established. The IBFs are also required
to obtain acknowledgement of receipt of the
notice from nonbank customers that are foreign
affiliates of U.S. residents at the time the credit
or deposit relationship is established with an IBF.
Advances by an IBF to U.S. offices of its parent
institution will be subject to the reserve requirement on Eurocurrency liabilities of the U.S. office
in the same manner as advances from a foreign
office to its U.S. office. An IBF will be subject to
the same examination and supervisory procedures
as apply to other operations of its parent
institution.

Regulation Y:
Board Confirms Rule
on Real Estate Appraisals
by Bank Holding Companies
The Board of Governors of the Federal Reserve
System has confirmed its decision to amend Regulation Y (Bank Holding Companies and Change in
Bank Control) by adding the performance of real
estate appraisals, including single-family residences, to the list of permissible nonbanking
activities in which bank holding companies may
engage.
Although the amendment became effective December 31,1980, comments were solicited until
January 15,1981, because the Board had modified
the original proposal by including appraisals of
single-family residences in the final rule. The final
rule was confirmed June 10,1981.

14

Federal Reserve Bank of Dallas

(JVowcfivailable
Recently issued Federal Reserve circulars, speeches, statements to Congress, publications. etc .• may
be obtained by contacting the Department of Communications. Financial and Community Affairs,
Federal Reserve Bank of Dallas. Station K, Dallas. Texas 75222, unless indicated otherwise. Requests
for circulars should specify the circular numbers.

Circulars

Speeches and Statements

Iranian Assets Control Regulations: Amendments. 30 pp.
Circular No. 81-110 (June 3, 1981).
Regulations D [Reserve Requirements of Depository Institutions] and Q [Interest on Deposits]: Amendment. 5
pp. Circular No. 81-115 (June 8, 1981).
Policy Statement: Sale of Third Party Commercial Paper.
7 pp. Circular No. 81-117 (June 11, 1981).
Regulation Z-Truth in Lending: Revised Regulation Z;
Technical Amendments; Proposed Official Stall Interpretation for Regulation Z Commentary. 5 pp. Circular
No. 81-118 (June 12, 1981).
Rules of Organization. Rules of Procedure: Revised Pamphlet. 13 pp. Circular No. 81-119 (June 12, 1981).
Regulations Z [Truth in Lending] and M [Consumer Leasing] Pamphlets. 143 pp. Circular No. 81-121 (June 16,
1981).
Regulation T [Credit by Brokers and Dealers]: Amendment
to Delete Provision Permitting Use of Foreign Currency
in a Margin Account. 3 pp. Circular No. 81-122 (June 18,
1981).
Iranian Assets Control Regulations: Amendments. 3 pp.
Circular No. 81-124 (June 22, 1981).
A Reminder Concerning Attempts to Perpetrate Fraudulent Transfers of Funds. 2 pp. Circular No. 81-125 (June
23, 1981).
Reclassification of Member Banks for Electoral Purposes.
1 p. Circular No. 81-126 (June 25, 1981).
Iranian Assets Control Regulations: Amendments. 7 pp.
Circular No. 81-127 (June 26, 1981).
Regulation Y-Bank Holding Companies and Change in
Bank Control: Affirmation of Final Rule Relating to
Real Estate Appraisals by Bank Holding Companies.
1 p. Circular No. 81-128 (June 26, 1981).

Remarks by Henry C. Wallich ("LDC Debt-To Worry or
Not to Worry") at the 59th Annual Meeting of the
Bankers' Association for Foreign Trade, Boca Raton,
Florida. 16 pp., including summary. June 2, 1981.
Statement by J. Charles Partee before the Subcommittee on
General Oversight and Renegotiation of the Committee
on Banking, Finance and Urban Affairs, U.S. House
of Representatives. 8 pp. June 4, 1981.
Remarks by Lyle E. Gramley at the 47th Annual Session of
The Stonier Graduate School of Banking, New Brunswick, New Jersey. 12 pp. June 11,1981.
Statement by Frederick H. Schultz before the Subcommittee on Conservation, Credit, and Rural Development
of the Committee on Agriculture. U.S. House of Representatives. 11 pp., including charts. June 23. 1981.
Statement by Paul A. Volcker before the Subcommittee on
Domestic Monetary Policy of the Committee on Banking, Finance and Urban Affairs, U.S. House of Representatives. 20 pp. June 25, 1981.

July 1981/Voice

Pamphlets, Brochures, and Reports
A Citizen's Guide to CRA. Published by the Federal Financial Institutions Examination Council. (For use by
those who wish to comment on applications subject
to the Community Reinvestment Act of 1977 filed with
the Board of Governors of the Federal Reserve System,
the Federal Deposit Insurance Corporation, the Federal
Home Loan Bank Board, or the Office of the Comptroller of the Currency) 39 pp. January 1981.
Through the Discount Window at the Federal Reserve. Published by the Federal Reserve Bank of Dallas. (Describes the programs under which Federal Reserve
credit is extended and explains the process used in
extending such credit to depository institutions) 8 pp.
May 1981.
Pamphlet announcing Consumer and Community Allairs
Handbook. Issued by the Board of Governors of the
Federal Reserve System. (Describes a new looseleaf
service of the Federal Reserve Board and includes an
order form) 4 pp. June 1981.
15