FRBSF Economic Letter
2000-24 | August 11, 2000
Twenty years ago, the policy problem facing many central banks was obvious–inflation was too high and needed to be reduced. Identifying the problem did not mean it was easy to solve. Debate centered on the potential real output and unemployment costs of reducing inflation, on whether a gradual disinflation or a more rapid one would be least costly, and on how central banks could credibly commit to carrying through a disinflation once begun.
Today, the policy issues facing many central banks are less clear-cut. Inflation has been brought down to low levels, and a clear priority is to ensure inflation remains low. The success in maintaining inflation at low levels, however, has led to suggestions that monetary policy should be directed towards a more ambitious goal–achieving a stable price level. Inflation is the rate at which the price level changes; with low but positive rates of inflation, the price level is steadily increasing. The U.S. has enjoyed low and stable rates of inflation in recent years, averaging just under 2.5% per year during the last five years. However, even these low average rates of inflation lead to sizeable increases in the price level over time. This is illustrated in Figure 1, which shows the price level (measured by the Consumer Price Index, left scale) and the inflation rate (right scale) over the last forty years. Since 1985, the U.S. price level has risen by more than 50%, even though inflation has averaged barely over 3% during this period.
A policy of price stability strives to keep the price level constant. This Letter reviews the arguments for and against pursuing price stability.
One of the chief arguments for price stability is that it makes planning for the future easier. When there is inflation, planning for the future is made more complicated by the need to forecast the future level of prices. For example, people planning for retirement must factor into their calculations the fact that prices are likely to be much higher when they retire. If the price level were stable, people could know that the general level of prices when they retire would be the same as it is today. People who are now thirty-something might not retire until 2035, but if inflation averages just 2.5% over the next 35 years, prices will have more than doubled over that time.
If people could be assured that inflation would be constant, their planning would be as easy as with a constant price level. Unfortunately, the historical record indicates that inflation does not remain constant. It fluctuates, and even fluctuations within a narrow range can increase uncertainty about the future. Suppose a household planning to retire in 30 years would like to have a retirement income equivalent to $75,000 at today’s prices. If inflation averages 2% over the next 30 years, they will need an income at retirement of $136,000, while if inflation averages just slightly more, say 3%, they will need an income at retirement of $182,000. Planning for retirement would be simplified if the price level remained stable, and individuals would not be forced to bear the unnecessary uncertainty associated with low but positive inflation.
If inflation remains relatively stable, the reduced uncertainty offered by a policy of price stability is likely to be small, however. For example, suppose inflation averages 2% per year, and almost always remains within a range of 1 to 3%. Fluctuations in inflation of this order of magnitude would only add a margin of error in forecasting the price level over 30 years of about plus or minus 2%.
A second argument in favor of price stability emphasizes the ways inflation can interfere with the operation of the price system. In a market economy, prices play a critical role, helping to allocate society’s scarce resources efficiently. For example, when a good becomes scarcer, as oil has in recent months, its price rises. This signals consumers to cut back on their consumption of that good. As a result of consumers making rational decisions about how to spend their income, the economy as a whole economizes on this scarce good. By responding to shifts in market demand and supply, the price system serves to direct the allocation of resources in the economy, playing the role of Adam Smith’s “Invisible Hand.”
How does inflation interfere with this process? One way is by making it more difficult to distinguish relative price changes from changes in the general price level. It is relative prices that matter for the allocation of resources. When the price of one good rises or falls relative to other goods, consumers and producers respond. If all prices rise together, relative prices remain unchanged; no adjustments in spending or production decisions by households and firms are called for. But when there is general inflation, and especially when it is variable, it may be hard to disentangle relative price changes from changes in the overall price level.
There is a second reason that inflation may interfere with the functioning of the price system, one that is a consequence of the fact that many prices and most wages adjust slowly, remaining fixed for extended periods of time. Wages, for example, are often adjusted only once a year. If inflation is positive, wage-setters and price-setters will need to be forward-looking, taking into account their estimate of average inflation over the period that wages and prices will be fixed. When inflation is expected, a firm that sets prices once a year will need to fix prices at a higher level than a firm that plans to change prices every month. That way, the price measured relative to the general level of prices will be about right over the whole year–but it will be too high relative to general prices at the start of the year, falling relative to general prices over the course of the year as general prices rise, and ending the year at a level that is too low relative to the general level. With individual wages and prices adjusting at different times over the year, inflation causes a greater dispersion in relative prices. This dispersion is inefficient–it does not reflect changes in relative scarcity of different goods, but instead it simply reflects the random rate at which different wages and prices are adjusted across the economy. This unnecessary dispersion could be eliminated if the general level of prices remained stable.
The major argument against pursuing price stability, as opposed to low inflation, is that overall economic instability might increase. To understand why a policy aimed at price stability might lead to more volatile inflation and output, suppose an increase in oil prices leads to a temporary rise in inflation. Under a zero inflation policy, the Fed must ensure that inflation is reduced back to zero. Under such a policy, however, the price level is left permanently higher by the temporary bout of inflation. For example, if inflation has been zero and then rises for one year to 2% before being brought back to zero, the general price level will be left 2% higher than its initial level. In contrast, under a policy of price stability, the Fed would need to bring the price level back down to its initial level–a temporary period of inflation would need to be followed by a deflation, a fall in prices. This makes inflation more variable–it might go from 2% one year to a negative 2% the next. If prices and wages are slow to fall, an economic slowdown might be required to produce the required deflation. Greater inflation variability and output variability could then result under a policy aimed at price level stability.
While this argument is plausible, work by Svensson (1999) suggests it is not always correct. Contrary to the case suggested by the simple example, the variability of inflation under either price level or inflation targeting will depend on how quickly policymakers try to bring the price level (the inflation rate) back to its target. In the example, it was assumed that the price level would be brought back to its target level one period after an inflation shock. This caused inflation to gyrate from +2% to -2%. If the price level were brought back on target more gradually, the volatility of inflation and the real economy would be less. For example, suppose again that inflation jumps to 2% for one year. If the price level returns gradually to target over the next five years, the fluctuations in inflation and output could be significantly reduced. Spreading the adjustment out over several years means that the price level remains away from its target longer, but the eventual return to the initial price level maintains the advantages of longer-run predictability of general prices. In principle, monetary policy could keep average prices stable around a constant level with little or no increase in the volatility of either the rate of inflation or in real economic activity.
Under a policy that aims for either zero inflation or for price stability, monetary policy’s ability to help stabilize the real economy may be hindered by the fact that market interest rates cannot be less than zero. In a severe recession, monetary policy could push market rates down to zero, but no lower. Some economists have argued that a positive average rate of inflation, by raising the average level of market interest rates, has the benefit of reducing the chances that the zero bound on market interest rates would ever pose a real constraint on policy. However, if the public believes the central bank will maintain long-run price stability, a severe recession with falling prices will generate expectations of future inflation, since prices will have to rise sometime in the future to bring the price level back on target. Even with market interest rates at zero, expectations of future inflation reduce real interest rates and help stimulate the economy.
A second argument against pursuing price stability hearkens back to the role of the price system in a market economy. One of the important prices in the economy is the price of labor–the real wage. When macroeconomic disturbances affect the economy, adjustments in the average real wage provide one of the mechanisms that helps restore full employment. For instance, if the supply of labor grows more rapidly than demand, real wages need to fall to prevent unemployment from rising. If the price level is kept stable, real wages can fall only if nominal wages decline. Since nominal wages seem to adjust slowly, and many economists argue that they are particularly slow to fall, the real wage may fail to respond quickly enough to maintain full employment. Focusing on price stability would, in this case, lead to greater fluctuations in employment. If nominal wages are slow to adjust, real economic activity may be more stable if the price level is allowed to change in the face of economic disturbances. A rise in the price level that reduces real wages may aid in restoring full employment more quickly than would be the case if the adjustment had to come exclusively through a fall in nominal wages.
Twenty years ago, the objective of policy was to bring inflation down. Now that inflation has been reduced and has remained at relatively low levels for several years, the more ambitious goal of price stability has been suggested. Long-run price stability would reduce one source of uncertainty that households and firms face when they plan for the future. The benefits of long-term price stability, however, would be offset if a policy of price stability increased the volatility of inflation and the real economy. Whether increased economic instability would occur, though, is an empirical issue, one that economic theory alone cannot determine.
Carl E. Walsh
Professor of Economics, UC Santa Cruz,
and Visiting Scholar, FRBSF
Svensson, Lars E.O. 1999. “How Should Monetary Policy Be Conducted in an Era of Price Stability?” In New Challenges for Monetary Policy, Federal Reserve Bank of Kansas City. Proceedings (August) pp. 195-259.
Opinions expressed in FRBSF Economic Letter 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. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.
Please send editorial comments and requests for reprint permission to
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120