Inflation: Causes, Costs, and Current Status
Inflation: Causes, Costs, and Current Status
Updated August 26, 2008
Specialist in Macroeconomics
Government and Finance Division
Gail E. Makinen
Economic Policy Consultant
Government and Finance Division
Inflation: Causes, Costs, and Current Status
Since the end of World War II, the United States has experienced more or less
continuous inflation. It would be difficult to find a similar period in American
history before that war. Indeed, prior to World War II, the United States often
experienced long periods of deflation. It is worth noting that the Consumer Price
Index in 1941 was virtually at the same level as in 1807.
During the last economic expansion, March 1991-March 2001, the inflation rate
remained low by post-World War II standards. This is true regardless of which of the
indexes is used to calculate the rate at which the price of goods and services rose. A
low inflation rate is especially significant since the U.S. economy was fully
employed, if not over fully employed, according to many estimates for the last three
years of that expansion. Yet, contrary to expectations, the inflation rate showed little
tendency to accelerate. Keeping an economy moving along a full employment path
without igniting a burst of inflation is a difficult policy task.
Because labor costs make up nearly two-thirds of total production costs, the rate
at which they rise is often regarded as an indication of future inflation at the retail
level. They tended to rise in the latter stage of the 1991-2001 expansion, moderate
during the subsequent contraction, recovery, and expansion.
Rather than measure inflation by using the rate at which prices overall are rising,
some economists prefer a measure that reflects primarily the systematic factors that
raise prices. This yields the “underlying” or “core” rate of inflation. Three measures
of this rate show that it accelerated during 2006 through 2008:1H. However, price
increases over this period have been especially sharp in food and energy, not included
in the core rate. Nevertheless, until recently, the United States has maintained a
relatively low rate of inflation during this expansion even as labor markets have
Why should the United States be concerned about inflation? This study reports
the distilled knowledge of economists on the real cost to an economy from inflation.
These are remarkably more varied than the outlays for “shoe leather,” long reported
to be the major cost of inflation (“shoe leather” being a shorthand term for the
resources that have to be expended on less efficient methods of exchanges).
The costs of inflation are related to its rate, the uncertainty it engenders, whether
it is anticipated, and the degree to which contracts and the tax system are indexed.
A major cost is related to the inefficient utilization of resources because economic
agents mistake changes in nominal variables for changes in real variables and act
accordingly (the so-called signal problem). This cost may not have been experienced
in the United States during the post-World War II era.
In troduction ......................................................1
Causes of Inflation.................................................2
The Economic Costs of Inflation......................................3
Inflation Costs in a Fully Indexed Economy.............................4
Inflation Costs in a Partially Indexed Economy...........................5
Inflation and Uncertainty............................................8
Economic Costs of Inflation: Summary................................8
The Measurement of Inflation........................................9
Changes in the Prices of Goods and Services........................9
Changes in Labor Costs........................................11
Some Component Parts of the CPI...............................12
The Underlying or Core Rate of Inflation..............................13
List of Tables
Table 1. Alternative Measures of the Rate of Inflation, 1998-2000:1H.......10
Table 2. Recent Quarterly Rates of Inflation...........................11
Table 3. The Rise in Labor Costs, 1998-2008..........................12
Table 4. The Change in the CPI and Its Major Components...............12
Table 5. The Underlying Rate of Inflation.............................13
Causes, Costs, and Current Status
Inflation is one of the differentiating characteristics of the U.S. economy in the
post-World War II era. Except for 1949 and 1955, prices, on average, have risen
each year since 1945. The cumulative effect of this inflation is staggering: the price
level has risen a little more than 1000% since the end of World War II.1
This was not true in the pre-World War II period. On the eve of that war, 1941,
the U.S. price level was virtually the same as in 1807. During the periods from 1846
to 1861 and 1884 to 1909, the United States experienced a near constant price level.
And in the 15 years from 1865 through 1879, the price level either remained constant
or declined. The principal periods of inflation between 1800 and 1941 were
associated with wars and the discoveries of gold and silver both here and abroad (and
with increased efficiencies in extracting both metals).
The experience with inflation during the 1991-2001 economic expansion was
reassuring in the sense that even as the unemployment rate declined to a 30-year low,
the inflation rate showed little tendency to accelerate.
In the final year of the 1982-1990 expansion, the Consumer Price Index (CPI)
rose 6.1% followed by a 3.1% increase in 1991. During the first 10 full years of the
1.7%, 1.6%, 2.7%, 3.4%, and 1.6%. For the years 2002-2007, the rate has been
2.4%, 1.8%, 3.3%, 3.4%,2.6%, and 4.1%.2 During first half of 2008, while the CPI
for all commodities rose at an annual rate of 8.5%, the food component rose at an
annual rate of 6.4% and the energy component rose at an annual rate of 60.5%.
Inflation can be defined as a sustained or continuous rise in the general price
level or, alternatively, as a sustained or continuous fall in the value of money.
Several things should be noted about this definition. First, inflation refers to the
movement in the general level of prices. It does not refer to changes in one price
1 In this and the following paragraph, all changes in the price level are as measured by the
Consumer Price Index.
2 The rate of rise in the CPI per year is measured on a December over December basis rather
than a year over year basis.
relative to other prices. These changes are common even when the overall level of
prices is stable.3 Second, the rise in the price level must be somewhat substantial
and continue over a period longer than a day, week, or month.4
Causes of Inflation
There has been practically no period in American history in which a significant
change in the price level has occurred that was not simultaneously accompanied by5
a corresponding change in the supply of money. This has led to a widely held view
that “inflation is always and everywhere a monetary phenomenon resulting from and6
accompanied by a rise in the quantity of money relative to output.”
Although this view is generally accepted, it is, in fact, consistent with two quite
different views as to the cause of inflation.
In one view a more rapid rate of money growth plays an active role in inflation
and results either from mistaken policies of the Federal Reserve or the Federal
Reserve subordinates itself to the fiscal requirements of the federal government and7
finances budget deficits through money creation. According to this view, the control
of inflation rests with the Federal Reserve and depends upon its willingness to limit
the growth in the money supply.
An alternative view comes in several versions. They have in common a belief
that the major upward pressure on prices comes from activities which would produce
a fall in real output. A favorite candidate is the attempt by organized labor to obtain
increases in real wages. Other activities include the monopolistic pricing behavior
of OPEC, major crop failures or changes in the terms of international trade produced
3 Especially troublesome for the definition of inflation is how to define a rise in the price of
an important commodity such as oil. Since it enters as an important input into the
production process as well as being a final product, it may cause many other individual
prices to rise. Is this a rise in relative prices or is it more appropriately defined as inflation?
Economists differ on how to describe this phenomenon. Some blame OPEC for the inflation
of the 1970s and early 1980s. Others treat this as a rise in relative prices and attribute the
inflation of the period to misplaced policies of the Federal Reserve.
4 The words “somewhat substantial” cannot be defined precisely. All of the major price
indexes have a number of shortcomings, such as only imperfectly correcting for changes
in the quality of the goods and services contained in the index. For that reason, relatively
low rates for inflation (e.g., plus or minus 2% or less) are often taken to be equivalent to
price level stability.
5 Perhaps the only exception to this statement is the inflation during the Korean War of
6 Milton Friedman, What Price Guideposts in Guidelines: Formal Controls of the
Marketplace, Aliber, Robert and George Schultz, ed. (University of Chicago Press, 1966),
7 Examples of Federal Reserve policies that are likely to produce inflation are those that fix
rates of interest too low or that support unrealistic foreign exchange values of the dollar.
by a decline in the foreign exchange value of the dollar. The decline in real output
that these activities produce will, in general, lead to rises in unemployment. To
prevent unemployment from increasing, in one version of this alternative, the Federal
Reserve is seen to pump up demand by easing the growth of the money supply. In
the process it ratifies the rise in the price level. Thus, in this version, while a growth
in the money supply is necessary to ratify the upward movement in the price level,
it is not the cause of the rise in prices.
It is interesting to speculate what would happen if the Federal Reserve refused
to expand demand in the face of the rise in unemployment. Presumably, after a
protracted period, the additional unemployment would lead to a fall in wages, costs,
and other prices. Over the longer run, output would return to its previous level or
growth path, the price level would fall back to its previous level and only relative
prices and wages would be different. Thus, while the Federal Reserve has the power
to curb inflation, it is unlikely to exercise this power in the face of a large runup in
In another extreme variant, what the Federal Reserve does is really irrelevant.
Should it refuse to expand what is conventionally called money to pump up demand
in the presence of these developments that reduce output, money substitutes under
the guise of credit will emerge that will allow demand to grow and the price increases
to be ratified. This variation, interestingly, precludes excessive money growth from
causing inflation for it also holds that the Federal Reserve cannot force too much
money on the economy. Inflation, then, cannot be a case in which too much money
is chasing too few goods.8
The first two explanations for inflation find many adherents among American
economists, whereas the third is more common among some British economists.
The Economic Costs of Inflation
Economists often discuss jointly the costs to an economy from unemployment
and inflation since for much of the period since the late 1950s it was generally
believed that a long run tradeoff existed between the two.9 While the cost of
unemployment was well articulated, the cost of inflation was relegated to “shoe
l eat her.”10
8 This view is commonly held by economists associated with the late Nicholas Kaldor and
systematically explained by him in “Monetarism and U.K. Monetary Policy,” Cambridge
Journal of Economics, vol. 4 (1980), pp. 293-318.
9 A comprehensive discussion of the costs of inflation can be found in Fischer, Stanley, and
Franco Modigliani, “Towards an Understanding of the Real Effects and Costs of Inflation,”
Weltwirtschaftliches Archiv, vol. 114, no. 4 (1978), pp. 736-787.
10 Before financial institutions could pay explicit interest on deposits that function as money,
economists believed that individuals and businesses would shift their wealth into savings-
type deposits on which interest was paid. Because of this, they would have to make more
frequent trips to banks to obtain money. This involved primarily a cost of shoe leather, as
The high U.S. inflation rate of the late 1960s, 1970s and early 1980s, caused
economists to rethink the costs of inflation to an economy. What follows is a
distillation of those efforts.
Describing the costs to an economy from inflation can be confusing for several
reasons. First and foremost there is the confusion over the cost to the economy
versus the cost to specific individuals. Costs to individuals may not impose a burden
on the economy because they are in the nature of a redistribution of either income
and/or wealth. What is lost by some is gained by others. Nevertheless, some of these
redistributions can have real effects.
Second, some of the costs of inflation are permanent in the sense that so long
as the inflation continues the costs will be incurred. Others are only transitory and
arise as the economy moves from one inflation rate to another or because the rate of
inflation itself is variable.
Third, some costs are incurred only because the inflation is unanticipated while
other costs arise even when the inflation is fully anticipated. Finally, some costs
occur only because of the absence for one reason or another of appropriate
safeguards, for example, the absence of indexed contracts.
Inflation Costs in a Fully Indexed Economy
As an introduction to understanding the costs imposed on an economy by
inflation, consider first an economy that is completely indexed for inflation. Thus
every conceivable contract is adjusted for changes in the price level including those
for debt (bonds and mortgages) and wages and salaries; where taxes are imposed only
on real returns to assets, where tax brackets, fines and all payments imposed by law
are indexed, where the exchange rate is free to vary and there are no legal restrictions
imposed on interest rates, etc.
In this economy, the distinction between anticipated and unanticipated inflation
is unimportant except if the inflation rate is high and the indexed adjustments are not
continuous. Then real costs can occur. However, for analytical purposes, assume
that all individuals perfectly anticipated the inflation and that the indexed
adjustments are continuous.
In this economy, inflation can impose only two real costs: the less efficient
arrangement of transactions that result from holding smaller money balances and the
necessity to change posted prices more frequently (the so-called menu costs).
The first of these, entailing the rearrangement of transactions due to the higher
costs of holding money, is the one cost uniformly identified in the text books as “the
cost of inflation.” It is worth considering what is involved.
shoes wore out more frequently because of the increased number of trips. Hence the often
expressed view that the primary cost of inflation to an economy was “shoe leather.”
Both individuals and businesses hold money balances because it allows each to
arrange transactions in an optimum or least cost way (e.g., for business this involves
paying employees, holding inventories, billing customers, maintaining working
balances, etc.) and to provide security against an uncertain future. Holding wealth
or assets in a money form, however, is not costless. A measure of the so-called
opportunity cost is the expected rate of inflation, a cost that rises because wealth can
be held in alternative forms whose price or value rises with inflation.
When inflation occurs or when the rate of inflation rises, holding money
becomes more costly. Individuals and businesses then attempt to get by with less
money (for businesses this may mean billing customers more frequently, paying
employees more frequently, etc.). This means that least cost transactions patterns are
no longer least cost. The new patterns are less efficient — they use more time or
more resources to effect a given transaction. In addition, holding smaller real money
balances also reduces the security money provides against an uncertain future.
The magnitude of this cost has been reduced in the United States in recent years
because financial institutions can now pay interest on a variety of deposits that
function as money. Thus, the primary cost of inflation on money holding applies to
currency on which no interest is paid. To the extent, however, that financial
institutions are slow to raise interest rates in tandem with inflation, deposit holders
will economize on holding deposits and arrange transactions less efficiently, thereby
imposing a short-run cost on the economy.
The other cost imposed by inflation in a fully indexed economy is the so-called
menu cost which involves the extra time and resources that are used in adjusting
prices more frequently in an environment where prices are rising. These additional
costs are incurred mainly with goods and services that are sold in nonauction
markets. It does not apply to auction markets where prices change more or less
continuously in response to shifts in supply and demand.
Inflation Costs in a Partially Indexed Economy
Very few economies are fully indexed, even those in which inflation is severe.
In the United States, indexation is incomplete. As such, inflation can impose costs
even if it is fully anticipated. A case in point involves the arrangements for levying
taxes. Taxes are levied in several instances on nominal as opposed to real income.
As a result, the interaction of inflation and taxation can impose real effects on an
economy by altering the incentives to work, save, and invest. Several examples
should suffice to explain what is involved.
Consider first an individual who, in a non-inflationary period, earns a real rate
of interest of 5% and who pays taxes of 30% on this income. The aftertax real rate
of interest is 3.5% (i.e., 5% - 30% x 5 %). Now, assume that a 10% rate of inflation
is expected over the one year term of the loan. As a result, the market rate of interest
rises to 15% (composed of a real rate of 5% and an expected inflation rate of 10%).
At a tax rate of 30%, the aftertax rate of return falls to 0.5%.11 To the extent that
saving is responsive to the real aftertax rate of return, taxing nominal yields as is
done in the United States, discourages individual savings.12 (The existing empirical
evidence for the United States suggests that private sector saving is quite insensitive
to the aftertax rate of return.)
Next, consider what happens to the real aftertax rate of return on business
capital during an inflation. For tax purposes, the depreciation of business plant and
equipment is based on actual or historic costs. During an inflation, charging
depreciation based on historic cost raises the nominal profits of businesses and the
basis on which corporate profits taxes are levied. As a result, the aftertax real rate
of return falls and this discourages businesses from adding to their stock of plant,
equipment, and structures — the bases for future economic growth.13 14
Third, to the extent that income tax brackets are not indexed or not indexed
completely, inflation in a progressive income tax system can reduce the real aftertax
income for wage and salary earners creating a disincentive to work.
During the 1980s, the U.S. tax code was rewritten to adjust the tax brackets for
inflation as well as to reduce the level and progressivity of the federal income tax.
As a result, inflation has a much reduced interaction with federal taxes in reducing
aftertax real income.
Several private sector practices also interact with inflation to produce real
economic effects. The first is the continuation of level payment nominal mortgages
for financing housing. This practice front loads the real cost of a mortgage during an
inflation and, as a result, it discourages the purchase of homes, especially by younger
Second, business firms continue to record all data in terms of the dollar even
though the real purchasing power of this important unit of measure varies
11 The aftertax real rate is equal to: 15% - 4.5% (which is 30% of 15%) = Aftertax nominal
yield of 10.5% - 10.0% inflation = 0.5% real aftertax yield.
12 The possibility arises that the interaction of inflation and the taxation of nominal rates of
return will produce negative aftertax real rates of return.
13 The taxation of nominal profits may also encourage business to opt for shorter-lived
capital during an inflation. In addition, since interest expenses are deductible for tax
purposes, inflation encourages businesses to finance expansion by the use of debt as
opposed to equity. This can impart an element of instability to the financial structure of the
14 Inflation can also influence some public decision-making because it leads to a
misrepresentation of the reported statistics on which these decisions are made. Specifically,
the Federal budget deficit tends to be overstated because the inflation premium in interest
rates that represents the repayment of principal, is reported as interest expense in both the
federal budget accounts and the GNP accounts. To the extent that public concern centers
on the current operating outlays of the federal government, true interest outlays are
considerably less than currently reported in the budget and, thus, the current operating
deficit is much smaller than reported in the federal budget document.
considerably over time. This practice has the potential for distorting the real
profitability of business over time as well as the valuation of other relevant
magnitudes. Since these nominal magnitudes are frequently used as the basis for
borrowing and lending decisions, they have the potential for seriously distorting
In this section, the real effects of inflation are analyzed in an environment where
it is unanticipated and where the economy relies on nominal or unindexed contracts.
In this situation, an important effect of inflation is to redistribute both income and
wealth. It would be a mistake, however, to conclude that because gainers and losers
cancel, there can be no real effects from inflation.
To see one such real effect, consider what happens to the interest bearing public
debt. Inflation reduces the real value of the public debt and with it the real value of
the wealth of the private sector, the ultimate owners of most of that debt. Thus,
inflation redistributes wealth from the private to the public sector. But who
constitutes the public sector? These are the taxpayers who also happen to be the
members of the private sector, some of whom own the debt.
Thus, redistribution reduces the real value of the taxes needed to service this
debt and the reduction is most beneficial for the younger workers in the current
population and for future generations. As a result of the fall in real tax burden, their
real disposable income rises, both today and in the future. They are thus able to save
more while older workers and retirees will, no doubt, have to reduce their
consumption for while they are faced with a large wealth loss, they gain very little
from the reduced tax burden. Thus, the redistribution of wealth between the private
and the public sectors is really a redistribution between generations that could have
an effect on the rate of capital formation.
Perhaps the most serious effect of unanticipated inflation in a market economy
is its potential to make the price system malfunction and misallocate resources.
Those who live in market economies are apt to take its functioning for granted. They
may fail to appreciate or understand the vital role that prices perform in such a
system. As standard textbooks in economics teach, the price system determines what
is produced, how it is produced, and to whom the output is distributed.
For the price system to perform these functions efficiently, producers must be
able to discern a change in real or relative prices from a change in nominal prices
which essentially leaves all relative prices unchanged. Only with the former will it
be profitable to alter production. A similar phenomenon holds for workers. A rise
15 These distortions could be minimized if the dollar was defined as a real unit of account
(e.g., defined in terms of a standard basket of commodities). Several proposals for doing
so have been put forth. See Warren L. Coats, Jr., In Search of Monetary Anchor: A “New”
Monetary Standard, IMF Working Paper, 1989.
in money wages may bring forth a greater quantity of labor time if workers are
convinced that this is a rise in real wages, that is, money wages relative to prices.16
It is easier for producers and workers to discern these changes in real prices and
wages if the price level is stable or if the inflation rate is constant. It is more difficult
when the rate of inflation is rising and/or more variable. Under these circumstances
market economies are apt to have “signal” problems. That is, producers and workers
mistake changes in nominal prices and wages for changes in corresponding real
magnitudes and act accordingly. The resulting changes in output and labor time are
inefficient and would not have occurred but for the mistakes in perception. A “signal
extraction” problem may not have arisen in the United States. The evidence
available is insufficient to come to such a conclusion.
Inflation and Uncertainty
Empirical studies completed in the 1970s support the view that inflation is
associated with greater uncertainty about future prices and that the degree of
uncertainty rises with the rate of inflation.17
Rising uncertainty about future prices is believed to produce several possible
“real” effects. First, individuals appear to shift from buying assets denominated in
nominal terms (e.g., bonds) to so-called real assets such as residential structures,
land, precious metals, art work, etc. Because some of these assets are in fairly fixed
supply, the resulting capital gain produced by the shift could conceivably raise
private sector wealth by a sufficient amount to cause a fall in the saving rate. Second,
to compensate for the perceived greater uncertainty, lenders appear to require a
greater real reward for supplying funds for investment. Third, contracts tend to be
The first two developments lead to rising real interest rates which tend to reduce
the rate of investment and capital formation. The third development leads
businessmen to prefer shorter lived assets.
Economic Costs of Inflation: Summary
What is the cost of inflation? It is customary in text books to answer this
question in terms of a situation where the rate of inflation is anticipated by all market
participants who can either continuously re-contract or in which everyone is protected
16 The key word in this explanation is “may,” for a rise in real wages has both a substitution
and an income effect. The substitution effect will cause workers to substitute work for
leisure while the rise in real income will make leisure a more attractive option to working.
Whether the quantity of labor time increases or decreases as the real wage rises will depend
on which effect is stronger.
17 There are numerous studies that provide support for this view. Some are cited in the
references in the Fischer and Modigliani study on which this section is based.
from inflation through indexation. In this world the cost to an economy from
inflation is the increased resource cost from conducting transactions with reduced
holdings of money — popularly termed “shoe leather” costs. If the inflation is
serious, this cost is by no means trivial.18
However, inflations are seldom perfectly anticipated. In this situation, perhaps
the most serious real effect comes from the ability of rising prices to jam the price
signals that are so important to the smooth and efficient functioning of a market
economy. Evidence suggests that this may not have been a problem for the United
States in post-World War II era.
In general, the cost of inflation to an economy will be larger the higher the rate
of inflation, the more variable the rate, the less it is anticipated, the greater is the
uncertainty it causes, and the less indexed is the economy.
The Measurement of Inflation
Changes in the Prices of Goods and Services
The rise in the general level of prices, the essence of inflation, is measured by
using a price index. Ideally, the price index used should be broad based and one in
which the individual prices are weighted to indicate their importance to the economy.
For purposes of this study, three separate price indexes are used. The first two
are very broad based and derived from the measurement of the nation’s gross
domestic product (GDP). They differ in the quantities that are used to weight the
prices. The first uses side-by-side year quantities (that move every year) and is
called, the chain weight deflator. The second uses current year quantity weights and
is called the implicit price deflator.
The third index is the Consumer Price Index (CPI), which prices a “market
basket” of goods and services purchased by an urban family, a market basket whose
individual items are weighted by how much the urban family spent on them in a base
year period — currently 1982-1984.
In Table 1, the inflation rate given by each index is shown for the past 11 years.
The rate of inflation measured by the two price indexes derived from the GDP
accounts are the same.
18 It should be noted that these are not “once-and-for-all” costs. They will be incurred as
long as the inflation continues. Thus, a correct measure of these costs is their present
Table 1. Alternative Measures of the Rate of Inflation,
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008:1H
GDP-Cha i n 1.1 1.6 2.2 2.4 1.7 2.2 3.2 3.5 2.8 2.6 1.8
Implicit 1.1 1.6 2.2 2.4 1.7 2.2 3.2 3.5 2.8 2.6 1.8
CPI 1.6 2.7 3.4 1.6 2.4 1.8 3.3 3.4 2.6 4.1 8.5
Source: U.S. Departments of Commerce and Labor.
Notes: CPI is on a December over December basis while the GDP deflators are on a 4th quarter overth
4 quarter basis. For 2008, all indexes are at annualized rates for the first half year.
The average annual compound rate of inflation over the 1996-2007 period was,
according to both the chain weighted and implicit price deflator for GDP, about
The data in Table 1 also show that the low point for the inflation rate during this
The data on Table 1 also reveal that over this period, the inflation rate has never
fallen to zero, even during the economic downturns of 2001 and 1990-1991.19 Why?
A simple answer from mainline economic theory is that the unemployment rate was
never held high enough for long enough. This requires some elaboration.
The key to understanding the theory is to see what happens to the inflation rate
when the actual unemployment rate differs from the full employment rate. Whenever
the actual unemployment rate is above the full employment rate, the resultant slack
will cause the inflation rate to fall. As the inflation rate falls, the expected rate of
inflation should also fall if economic agents believe the government is sincere in its
efforts to end inflation (i.e., that the government will not reverse its policy in the face
of rising unemployment). As inflation expectations fall so will wage demands and
falling wage demands will bring about a lower unemployment rate. Ultimately, the
economy will move back to full employment at a zero inflation rate or a stable price
level. Thus, the important steps in the sequence are (1) a convincing government
policy to reduce the inflation rate to zero; (2) toleration of an above normal rate of
unemployment; and (3) the adjustment of inflation expectations and wage demands
to the lower rate of inflation. Obviously this did not happen during the 2001
downturn or that during 1990-1991. The Federal Reserve reversed policy and
expanded demand before the inflation rate fell to zero. How long and by how much
the Fed would have had to tolerate the U.S. unemployment rate above its full
19 Given that the CPI imperfectly measures the “true” rate of inflation, a stable price level
or a “true” zero rate of inflation is thought to prevail when the inflation rate as measured by
the CPI falls within a range of from 0.5% to 2.0% - this being the possible range of error in
the current CPI.
employment level is uncertain. In light of actual experience, a number of economists
believe the Fed reversed course too soon.
Because annual measures of the change in the price level can conceal some large
quarterly variations, the annualized quarterly rates of inflation according to the chain
weighted GDP index and the CPI are given for the period beginning 1997 in Table
Table 2. Recent Quarterly Rates of Inflation
(in percentages at annual rates)
1999 2000 2001 2002 2003
GDP- 1.7 1.5 1.6 2.2 3.1 1.3 2.1 2.2
200420052006 2007 2008
Cha i n 3.7 3.8 2.3 3.2 4.0 2.1 4.1 3.7 3.5 2.7 2.8 2.2 4.1 2.1 1.5 2.8 2.6 1.1
CPI 3.5 4.8 1.9 3.4 2.3 4.0 5.2 3.2 2.2 4.9 3.6 -2.5 4.2 4.6 2.7 5.3 4.0 5.0
Source: U.S. Departments of Commerce and Labor.
Changes in Labor Costs
Because labor costs comprise nearly two-thirds of the value of final output,
some economists believe that they are an important determinant of the rate of
inflation. However, changes in the rate of growth of labor costs must be read with
care. Wage increases can be driven by productivity increases, tight labor markets,
and inflation and fears of inflation. One way to determine the force or forces driving
wage increases is to examine what happens to per-unit labor costs. To this end, in
Table 3, the two major measures of labor cost, a comprehensive measure of wage
and benefit costs, the employment cost index and per-unit labor costs in the nonfarm
business sector, are shown.
The growth rate of both measures of labor cost generally showed a tendency to
accelerate during the expansions of the 1980s and 1990s as labor markets tightened.
Subsequent recessions and growing unemployment had a depressing effect on the rise
in both measures. During the current expansion beginning in 2002, the rate of
increase in both measures is fairly steady even as the unemployment rate has fallen.
Table 3. The Rise in Labor Costs, 1998-2008
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Per unit labor costa126.96.36.199.4-0.50.30.92.188.8.131.52
ind e x
Source: U.S. Department of Labor.
a. In the nonfarm business sector on a year over year basis. For 2008, annualized rate for the first
b. For private industry on a December over December basis.
Some Component Parts of the CPI
Not all prices rise at the same rate over time. In fact, a characteristic of market
economies is frequent changes in relative prices. These changes would occur even
in the absence of inflation and are due to such factors as changes in tastes and the
introduction of new production technologies. In Table 4, the rise in the prices of the
major component parts of the CPI are contrasted with the rise in the overall CPI. A
noticeable development over the course of the decade shown in the table has been the
great variability in the relative price of energy and the rise in the relative price of
medical services. During 2007 and the first half of 2008, food prices have risen at
an accelerated rate.
Table 4. The Change in the CPI and Its Major Components
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
CP I 1 .6 2.7 3 .4 1.6 2 .4 1.8 3 .3 3.4 2 .6 4.1 8 .5
Food 2.3 1 .9 2.8 2 .8 1.5 3 .6 2.6 2 .6 2.1 4 .9 6.4
Shelter 3 .3 2.5 3 .5 4.1 3 .2 2.2 2 .7 2.6 4 .1 3.1 3 .9
Medical 3.4 3 .7 4.2 4 .7 5.0 3 .7 4.3 4 .3 3.6 5 .2 3.7
Source: U.S. Department of Labor.
Note: Measured on a December over December basis. For 2008, it is the annualized rates over the
first half of the year.
The Underlying or Core Rate of Inflation
Although the actual rate of inflation can provide much useful information to
policymakers on the state of the economy, it can also be misleading since it responds
to both systematic and random forces. The latter can best be understood by reference
to the food component of the CPI. An unusual cold spell in Florida in January that
damages a substantial part of the fresh produce crop can send food prices and the CPI
soaring. A similar effect can be produced by an unusually wet summer in the
midwest. Alternatively, an unusually good combination of rain and sunshine can
produce a bountiful harvest and lower prices. Energy prices are also susceptible to
such random effects associated with events as turmoil in major oil producing nations.
To minimize the confusing signals that could arise from the use of the actual
rate of inflation, some economists prefer to use a price index to measure inflation that
reflects basically only systematic forces. For some economists this can be achieved
by using the CPI less its food and energy components. (It should be noted that food
and energy represent about 25% of the current CPI.) Others want to use a moving
average either of the CPI itself or of the current CPI less its food and energy
components. The use of a moving average is based on the belief that if there are
random factors that influence the actual inflation rate, they have an average value of
zero. Hence, the use of a moving average should minimize their influence.
In Table 5, three measures of the so-called trend, underlying, or core rate of
inflation are presented. The first (CPI,4QMA) is simply a four-quarter moving
average of the CPI. The second (CPI-F+E) is the actual rate of inflation from the CPI
stripped of its food and energy components. The third (CPI-F+E,4QMA) is the rate
of inflation calculated from a four-quarter moving average of the second index.
Table 5. The Underlying Rate of Inflation
(annualized quarterly rates of change)
1999 2000 2001 2002 2003
CPI,4QMA 1.7 2.2 2.2 2.7 3.2 3.3 3.3 3.6 3.5 3.3 2.8 1.6 1.1 1.4 1.6 2.3 2.9 2.0 3.5 2.4
CPI-F+E 1.6 2.4 0.9 2.5 2.9 2.9 2.5 2.4 2.9 2.6 2.6 2.6 2.4 2.1 2.1 1.8 1.1 0.8 1.7 1.0
C P I-F + E , 2.0 2.1 1.8 1.8 2.3 2.3 2.7 2.7 2.2 2.7 2.6 2.6 2.6 2.4 2.3 2.1 1.7 1.5 1.5 1.0
2004 2005 2006 2007 2008
12 34 12 34 12 34 12 34 12 3 4
CPI,4QM 1.3 2.8 2.6 3.5 3.1 2.9 3.8 3.7 3.7 3.9 3.5 2.0 2.4 2.6 2.2 4.2 4.2 4.3
C P I-F + E , 1.5 1.7 1.9 2.0 2.3 2.1 2.1 2.2 2.1 2.5 2.7 2.6 2.7 2.1 2.1 2.3 2.4 2.3
Source: Computations by the authors based on Bureau of Labor Statistics data.
Inflation can impose a real cost on society in terms of the efficiency with which
the exchange mechanism works, by distorting the incentives to save, invest, and
work, and by providing incorrect signals that needlessly alter production and work
effort. Because of this, policymakers should be concerned with the ongoing rate of
inflation and any tendency for it to accelerate. An additional reason for concern
arises because efforts to reduce the rate of inflation have often been associated with
economic downturns. It should not be forgotten that the double- digit inflation of the
early 1980s was reduced only through an economic downturn during which the
unemployment rate rose to its highest level since the depression of the 1930s. It is
argued that the tendency for the inflation rate to accelerate in the late 1980s was a
major reason why the Federal Reserve tightened monetary policy, which was an
important factor causing the recession of 1990-1991. Inflationary developments
subsequent to that recession have been encouraging. The inflation rate has shown
either no or only a modest tendency to rise as unemployment came down. Using
various measures, the inflation rate for the period of 1993-2000 was low by standards
of the preceding decade. Much of the upward pressure on prices during 1999-2000,
2002, 2004, 2005, 2006, 2007, and the first half of 2008 has come from energy
prices. Likewise, the decline in the inflation rate during 2001 and 2003 was due to
a large fall in energy prices. Over the period from the mid-1990 until 2005, the
inflation performance by the U.S. economy was the best since the 1960s.
Nevertheless, the acceleration in energy prices during 2007 and the first half of 2008
is expected to have at least a short run effect on other prices because energy is an
important input cost in determining those prices. In addition, food price increases
over the period 2007-2008:1H have been especially sharp. During the first half of