View original document

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



Number 93-44, December 31,1993

Inflation and Growth
In recent years, the Federal Reserve has stressed
its long-run commitment to maintaining low in­
flation because it believes that persistent inflation
imposes burdens that reduce economic welfare.
A variety of such burdens have been identified in
the economics literature. This Weekly Letter ex­
amines the argument that a lower rate of inflation
increases the potential growth rate of the economy,
and discusses the possible size of this effect.
Since a policy to reduce inflation generally has
some short-run costs because it requires a tem­
porary slowing of economic activity, it is useful
to estimate its benefits in the form of higher longrun growth.
How inflation might lower growth

Inflation might affect potential output in a num­
ber of ways. First, inflation may interfere with the
efficiency of the price system and make it more
difficult for households and firms to make correct
decisions in response to market signals. It is often
argued that when most prices are rising, economic
agents find it harder to distinguish between
changes in relative prices that require them to
reallocate resources and changes in the overall
price level that require no such microeconomic
response. A widget-producing firm that observes
that its customers are bidding higher prices for its
widgets may interpret this as indicating a rise in
the demand for its products, when it actually re­
presents the effects of generalized inflation in
which the prices of competing products (wadgets,
wodgets, etc.) also are rising at the same pace.
Second, inflation imposes various costs on the
economy that would disappear if prices were sta­
ble. The costs of changing prices and wage rates
frequently, the search costs imposed on buyers
and sellers when prices change often, and the
costs of economizing on holdings of non-interest­
bearing money (“shoe-leather" costs) are familiar
Inflation also has differing effects on individuals.
For example, the incomes of wage and salaried
workers generally are adjusted for inflation only
annually, whereas self-employed workers can
alter the prices of their services more frequently.

Similarly, inflation, especially when it is unex­
pected, tends to benefit borrowers at the expense
of lenders. Finally, because some parts of the tax
code are indexed for inflation whereas others are
not, generalized increases in prices have differing
effects on individual tax-payers. As a result of
these considerations, inflation often is perceived
as causing unfairness, since some households
and firms benefit and others are harmed.
However, whether or not these differential effects
of inflation are unfair, they do impose real costs
on society at large. They frequently add to the
uncertainties that households and firms face,
which may be undesirable even for those that
turn out to benefit. And many activities that seek
to reduce the impact of inflation on individuals
may hurt the overall economy but yield no cor­
responding overall benefits. In an inflationary
economy, for example, talented persons may de­
vote their energies to developing strategies to
avoid the deleterious consequences of inflation
for themselves rather than to inventing new prod­
ucts and processes that would raise overall living
standards. Unfortunately, many of these activi­
ties that aim to mitigate the effects of inflation
are counted as additions to measured GDP, even
though they may not add to welfare in any mean­
ingful sense.
Finally, inflation may affect investors' saving and
investment decisions, reducing the proportion
of GDP devoted to investment and causing the
economy to accumulate less productive capital.
For example, when inflation is high, it usually
tends to be more variable and so harder to fore­
cast. Uncertain inflation makes it more difficult
to deduce the real returns on investments from
available market information. As a result, savers
and investors often are less willing to enter into
long-term nominal contracts or to invest in long­
term projects. The reduced stock of productive
capital that results from decreased investment
will, in turn, imply lower levels of future GDP.
These considerations suggest that there are sound
a priori reasons why persistent inflation might re­
duce the growth rate of GDP in the long run. A

number of studies have investigated the sources
of long-term growth using data from a cross­
section of countries, and several have examined
whether differences in growth among countries
are related to differences in average rates of in­
flation. The results have been mixed, perhaps
because it is difficult to isolate the impact of
inflation on long-run growth.
Why the effects of inflation are hard to discern

The most obvious difficulty is that inflation is
only one of many factors that may affect a coun­
try's long-run growth. For example, a country that
saves a large share of its output and devotes it to
investment in productive capital or to educating
its workers is likely to enjoy a higher and more
rapidly growing GDP than one that devotes most
of its output to current consumption. Conversely,
a country with a rapidly increasing population
is likely to have lower GDP per worker because
more of its saving is needed just to provide the
existing levels of education and capital to the new
entrants to its work-force and so less is available
to increase the stocks of human and tangible
capital per worker. Finally, many empirical stud­
ies find that countries with initially low levels of
output tend to grow more rapidly than advanced
countries, perhaps because they find it easier to
adopt technologies that already are in use in
more advanced economies or because their out­
put levels are below the long-run equilibrium
associated with their rates of saving and popula­
tion growth.
To isolate the impact of inflation on growth, we
also must estimate the effects of these (and other)
systematic influences, in order to judge what, if
any, additional explanatory power inflation pro­
vides. Unfortunately, there is no fully accepted
model of economic growth that can serve as a
baseline for estimating the additional effects of
inflation. Although most economists agree that
factors such as the saving rate and the population
growth rate are important determinants of over­
all growth, they are far from unanimous on the
precise mechanisms through which their effects
are felt.
In addition to variables that received economic
theory suggests should be related to long-term
growth, there is a vast array of factors that might
plausibly have an influence. Political stability, the
size of the financial sector, openness to world
trade, and the size of the government sector all
have been suggested as potential influences on

growth. Each of these factors may be measured
in a variety of ways. Econometricians may be
tempted to search the data to find the set of var­
iables that were most closely related to growth in
the past. The problem is that statistical relation­
ships that are uncovered by such "data-mining"
but that have no genuine causal basis are unlikely
to hold up in the future. Moreover, some of these
"plausible" variables may be statistically related
to inflation so that their inclusion may obscure
the link between inflation and growth.
Another serious problem is that even if inflation
does reduce long-run economic growth, this ef­
fect may be difficult to detect in the data because
there also may be short-run links that obscure the
long-run relation. In the short run, real growth
may be positively related to inflation because
fluctuations in GDP reflect variations in aggre­
gate demand. In a cyclical upswing, increasing
demand may raise real output while it also bids
up prices by putting greater pressure on re­
sources. As a result, we may observe a positive
relation between real GDP growth and inflation
in the short run, even though there is a negative
relation over the long haul.
Any long-run relation between inflation and
growth also may be obscured if the world is af­
fected by supply shocks that influence prices
and/or real output by differing amounts in dif­
ferent countries. Indeed, the observed negative
across-country correlation between long-run
inflation and growth might be due to a small
number of major supply shocks that affected the
levels of prices and of output in a significant
group of countries. For example, if a worldwide
increase in the price of oil were to permanently
raise the level of prices and lower the level of
output by more in some countries than in others,
a cross-country comparison over a longer period
that included the oil shock years might detect a
negative correlation between the average growth
rates of prices and output, even if there were no
causal relation between these two growth rates.
Dealing with the statistical problems

Recent research at this Bank (Motley 1993) has
attempted to deal with these problems in a vari­
ety of ways. This research used a model of eco­
nomic growth that economists previously have
found helpful in understanding differences in ex­
perience among countries (Solow 1956, Mankiw,
Romer, and Weil 1992). This model emphasizes
the key role of the rates of saving and population

growth as the fundamental determinants of longrun growth, since these factors ultimately deter­
mine the supplies of capital and labor.
This model was expanded to include the effects
of differences in rates of inflation. Inflation was
assumed to influence long-run growth by affect­
ing the pace of "technological change," a port­
manteau term that includes the effects of all vari­
ables apart from the supplies of the factors of
production. This choice of theoretical framework
implies that although technological change is af­
fected by inflation, it is otherwise exogenous and
so is independent of the saving and population
growth rates.
No other variables apart from those suggested by
this theoretical model were included in the em­
pirical analysis. This limitation avoids the temp­
tations of data mining, but it implies the risky
assumption that economic growth is driven by
the same fundamental forces in countries with
widely diverse environments.
The statistical problem of separating the longerrun effects of inflation on potential GDP from
any short-run business cycle relation between
prices and output was dealt with by examining
growth and inflation over a long period that
spanned several business cycles. This makes it
more likely that the results represent a long-run
relationship. At the same time, the possibility that
any long-run correlation between inflation and
growth actually represents the influence of a few
supply shocks was examined by estimating cross­
section equations over a series of shorter time
periods in addition to the cross-section covering
a longer span. Finding a negative cross-section
relation between inflation and real growth in sev­
eral time periods of varying length might con­
stitute evidence that it represents a true causal
relation rather than the effect of supply shocks
that affected countries differently. However, this
procedure raises the possibility that the equations
estimated over the shorter time periods may be
contaminated by the business cycle problem
mentioned earlier.
The results of this research suggest that coun­
tries with lower inflation rates do tend to exhibit
higher rates of long-run growth. The cross­
country comparisons indicate that a 5 percent­
age point decrease in the average rate of inflation

is associated with an increase in annual growth
per capita of about 0.2 percentage point. Since
inflation in the U.S. between 1982 and 1992 was
about 5 percentage points lower than during
the previous ten years, this result suggests that the
lowering of inflation will add about 0.2 percent­
age point to long-run growth. This is sufficient to
increase the discounted lifetime income of a typ­
ical worker by an amount equal to about one
year's income, assuming a 40-year working life
and a 3 percent real discount rate. This estimated
benefit exceeds the costs of lowering inflation
which typically are estimated as amounting to at
most one-fourth of one year's GDP for a 5 per­
centage point reduction in inflation (Ball 1993,
Mankiw 1992, p. 309). It must be recognized,
however, that although the benefits exceed the
costs in present value terms, much of the benefit
will accrue in the relatively distant future,
whereas the costs must be borne today.
As I have emphasized, there are a number of sta­
tistical problems that always will make it difficult
to demonstrate conclusively that lower inflation
will lead to faster growth. However, the evidence
in this study, while not conclusive, not only is
consistent with such a relationship but also sug­
gests that the quantitative magnitude of this effect
is sufficiently large that policymakers should lean
toward low inflation.
Brian Motley
Senior Economist

Ball, Laurence. 1993. "H ow Costly Is Disinflation? The
Historical Evidence." Federal Reserve Bank of Phil­
adelphia Business Review (November/December).
Mankiw, N. Gregory. 1992.
York: Worth Publishers.



_______________ , David Romer, and David N. W eil. 1992.
"A Contribution to the Empirics of Economic
Growth." Quarterly lournal of Economics (May).
Motley, Brian. 1993. "Growth and Inflation: A Cross­
Country Study." Working Paper 93-11. Federal
Reserve Bank of San Francisco.
Solow, Robert M. 1956. "A Contribution to the Theory
of Economic Growth." Quarterly Journal of Eco­
nomics (February).

Opinions expressed in this newsletter do not necessarily reflect the views of the management of the Federal Reserve Bank of
San Francisco, or of the Board of Governors of the Federal Reserve System.
Editorial comments may be addressed to the editor or to the author. . . . Free copies of Federal Reserve publications can be
obtained from the Public Information Department, Federal Reserve Bank of San Francisco, P.O. Box 7702, San Francisco 94120.
Phone (415) 974-2246, Fax (415) 974-3341.

^ /X

s ) i ua As m/
> e q o iin
^ 'Cr ja e pp oj u piu d
d d a Aa o a u

0ZLW> V 'ojspupjj

ZO xog




jo Huog
0 A J0 S 0 y |OJ0p©-J

queuj^Jods(] qzuoesey
Index to Recent Issues of F R B S F


W e e k ly L e tte r

93 -30
93 -40

Federal Reserve Independence and the Accord of 1951
C h in a on the Fast Track
Interdependence: U.S. and Japanese Real Interest Rates
N A FTA and U.S. Jobs
Japan's Keiretsu and Korea's Chaebol
Interest Rate Risk at U.S. Com m ercial Banks
W hither California?
Eco no m ic Impacts of Military Base Closings and Realignments
Bank Lending and the Transmission of Monetary Policy
Sum m er Special Edition: Touring the West
The Federal Budget Deficit, Saving and Investment, and Grow th
Adequate's not G oo d Enough
Have Recessions Becom e Shorter?
California's Neighbors
Inflation, Interest Rates and Seasonality
D ifficult Tim es for Japanese Agencies and Branches
Regional Com parative Advantage
Real Interest Rates
A Pacific Econom ic Bloc: Is There Such an Anim al?
N A FTA and the W estern Economy
Are W orld Incom es Converging?
M onetary Policy and Long-Term Real Interest Rates
Banks and M utual Funds

W alsh
H uh/Kim
Sherw ood-Call
Sherw ood-Call
Crom w ell
Crom w ell
Zim m erm an
Schm idt
Frankel/W ei
Schm idt/Sherw ood-Call
M oreno
Laderm an

The F R B S F W e e k ly L e tte r appears on an abbreviated schedule in June, July, August, and Decem ber.