Does the fed abhor inflation?

Current discussions of monetary policy have the Fed being virtually single-minded in its resolve not to let the inflation bogey out of the bottle. Yet this concern is relatively recent, dating from 1992 when the CPI inflation rate moved below 4 percent to its present level. Just a half-decade earlier,

disinflation policies pushed the double-digit inflation rate of the early 1980s to less than 2 percent in 1986, from which it jumped to 3.7 percent the following year. It remained in the 4-5 percent range until 1992, due in large part to the fear of a repeat of the unemployment experience of the early 1980s associated with the deceleration of inflation, a situation seen as a reaffirmation of the Phillips curve.

If asked about policy in those years, the usual response was something like, "Of course we don't want to see a higher inflation rate, but the unemployment consequences of bringing the rate again below 2 percent are too high." In other words, a 4-5 percent inflation rate was regarded as optimal, because policy was to have neither a higher nor lower inflation rate.

With the inflation rate now less than half of that, is there reason to push it lower, say to zero percent, as recently argued by William Poole (1999)? On the basis of the U.S. experience, I demonstrate that even lower rates of inflation-in other words, deflation-and the negligible interest rates they imply do not portend disaster for monetary policy and for the economy. In particular, I argue that the liquidity trap notion implied by deflation should not be a shibboleth for inflationary policies.

The Fear of Low Inflation

In the period 1800-1913, consumer prices fell 42 percent, an average annual deflation rate of one-half of a percent. Included in this was the inflation of the Civil War and the 30 year post-Civil War deflation of 46 percent. From 1913 to 1999, consumer prices rose 1,568 percent. This period included the sharp deflation in the early 1920s and 1930s as well as the inflation in World War I and II. Of more interest is the post-World War II experience and the contrast to the earlier period. Between 1800 and 1946, consumer prices rose less than 15 percent, an annual inflation rate of less than a tenth of a percent. In the 1947-99 period, prices rose 640 percent.

The reason underlying the 640 percent price rise is not hard to uncover; it is principally concern about unemployment, in particular, a desire not to repeat either the miseries of the 1930s or the Federal Reserve's culpability in those times (Wheelock 1998). The use of inflation as grease rather than sand in reducing unemployment later became formalized by the Phillips curve. This was most clearly seen in the1970s, a decade characterized by J. Bradford DeLong (1997) as "America's Peacetime Inflation," for it was here that expansionary monetary policy was used to cope with supply shocks, as called for by frameworks based on the Phillips curve.

The fall in inflation to its current low levels has been captured in low interest rates. And that is a source of anxiety because of the fear that a recession when interest rates are "too close to the zero bound ... gives the Fed [in]sufficient room to ease-that is, to cut rates" (Poole 1999: 454). Hence, the implication is that the Fed should have a bias toward modestly rising prices, that is, it should inflate at least at current rates.

This concern over a zero interest rate bound is the topic of several recent papers (Clouse et al. 1999; Favero, Giavazzi, and Flabbi 1999; Krugman 1998; McCallum 2000; Reifschneider and Williams 1999). In these studies, deflationary monetary policy is generally not seriously considered because there appears to be general agreement that it would wreak havoc on financial institutions as debtors default due to their inability to carry the increasing real burden of their debts. This would spill over to financial institutions who similarly would be pushed to bankruptcy because of loan defaults. This of course is an extension of the Ben Bernanke (1983) thesis, taken from Irving Fisher (1932).

One theme that emerges is the popular 1930s' concept of a liquidity trap, the notion that even aggressive monetary actions would not stimulate recovery because they could not reduce interest rates. In the current transmission channel view emphasizing interest rate targeting, the fact that rates do not and indeed cannot then fall, there can be no short-term economic stimulus. The clear implication is that monetary policy is not and cannot be effective.

Paul Krugman (1998) reintroduced the liquidity trap possibility into contemporary discussion in his ruminations about the longrunning recession in Japan. Because short-term interest rates are virtually zero there, monetary policy cannot reduce real interest rates through pushing the nominal rate down. There is, however, another monetary option, one for which there is impressive historical precedent.

Deflation and Recovery

Suppose the economy is in a recession, interest rates are zero, prices are falling, central bank credit is declining, loans by banks similarly are falling, and excess reserves in the banking system are extraordinarily high and mounting. What is the prognosis for the economy over the next year? This is not just an idle intellectual exercise.

The economy about which I am talking is not the present-day Japanese economy, but that of the United States in 1937-39. A very sharp depression began in May 1937 and lasted through May 1938. The economic data for the period appear in Table 1. The dates in the table correspond to the beginning of the depression, May 1937; the trough, May 1938; and the last month of a 27-month fall in prices, August 1939. The June 1941 column represents the month in which prices returned to where they were at the recession's start.2 Except for the real rate and the rate on short-term treasuries, all data are indexed at 100 in May 1937.

In the depression, industrial output fell 33 percent.3 Unemployment increased to 19 percent in 1938, from the previous year's 143 percent. Wholesale prices declined 11 percent during the depression and continued to fall another 3 percent over the next 15 months, until August 1939, thereby falling 14 percent from the start of the depression two and a quarter years earlier.4 After increasing a paltry 0.7 percent during the depression, Federal Reserve Credit contracted 5 percent during that first 15 months of recovery. The interest rate on 3-month Treasuries averaged two-tenths of a percent, falling from 41 to 5 basis points in the 1937-38 depression. Thereafter, it averaged 4 basis points in the subsequent 15 months. Bank loans also declined, falling 14 percent overall, of which 3 percent occurred after the May 1938 trough. The decline in business loans marched with the fall in total loans as they also fell 3 percent after the trough. The excess reserves of the banking system at the end of the depression were $2.5 billion. This was 172 percent higher than at the beginning, as banks sought to rebuild their reserve positions to levels prevailing prior to the three-step doubling of reserve requirements beginning in August 1936. The $2.5 billion in excess reserves was about five times higher than their peak at the nadir of the Great Depression, 1929-33. They subsequently more than doubled, rising to $5.2 billion in the next 15 months.

Anyone viewing the economy in late spring 1938 would have foreseen a continuation of the sharp slide that began late the previous spring. Had there been a contest for "most pessimistic economic outlook," that period certainly would have been a prime candidate for making it to the contest's finals.

IMAGE TABLE 13

TABLE 1

The economy however did not emulate 1929-33. It rebounded sharply over the next five quarters, the ones in which prices continued falling. In that period, industrial production increased 34 percent, an annual rate of 26 percent. Real GNP rose 11 percent, an annual rate of 9 percent. What happened? How could an economy with essentially zero short-term interest rates, deflation, decreased bank lending, restrictive Federal Reserve policy, and increasing excess reserves rebound so dramatically? The answer to a large extent was monetary expansion. The M2 and MI money supplies increased 13 and 18 percent, respectively, beginning in late spring 1938, after declines of 3 and 5 percent in the depression. The money supply increases were not due principally to Fed policy. Though reserve requirements were reduced in early spring 1938, Federal Reserve credit was unchanged during the depression and then actually declined.

The principal driving force undergirding the monetary expansion was the rise in the monetary base, and this was due to the massive inflows of gold from Europe. The base expanded 28 percent during the 1938-39 recovery phase, up markedly from its 6 percent increase in the depression. The gold stock rose 27 percent, up from its 4 percent increase in the depression, thus contributing almost entirely to the expansion of the monetary base in the recovery.'

The economy continued to expand as the monetary base and money supply continued to grow. In the fall of 1939, the economy was back to where it had been a decade earlier, in autumn 1929, as the Great Depression was in its early stage. Prices reversed their downward move, getting back to their 1926 = 100 base level in late 1942. Judicious policy would then have ceased being expansionary. By that time, however, monetary policy was no longer concerned with the behavior of prices. Its principal focus was war finance, in particular with keeping interest rates low. The exigencies of World War II shifted monetary policy and the Fed away from its traditional concerns. With the end of the war and the widely held belief that a postwar repeat of the 1930s might well be in store, monetary policy turned increasingly to a predisposition toward inflation as a vehicle to palliate unemployment.

Will the Fed's recently found abhorrence of inflation be discarded in the next economic downturn? Will the liquidity trap and fear of deflation be sufficient reason for the Fed to abandon its low inflation policy? The recent evidence suggests the tenuousness of the Fed's commitment to low inflation. The historical evidence from the late 1930s indicates that aggressive monetary expansion can move the economy back to trend, even when deflation is present. That inflation came later in the buoyant recovery was due to the Fed's perhaps understandable unwillingness to take action in light of the developing war climate. It was certainly not because of ignorance as to how it can be prevented.

Conclusion

The recent experience of Japan with its depressed economy and short-term interest rates of essentially zero appears to be classic archetype of the liquidity trap. As such, the prospect of the U.S. experiencing such a situation has been used as justification for policies of moderate inflation, in order to keep nominal short-term rates from hitting their zero lower bound. That is, the prospect of a liquidity trap seems to be another) rationale for inflationary policies.

The evidence dealing with the recovery from the deep 1937-38 depression, one that carried all signs of a liquidity trap framework, demonstrates vividly that expansionary monetary policy is not emasculated by near-zero interest rates.

In his recent article, Poole (1999: 461) pointed out that monetary policy is not fundamentally concerned with "controlling of nominal interest rates. [Rather, it] is reflected in the growth of the money stock and, ultimately, the rate of inflation." The experience of 193738 gives clear evidence of this.

The liquidity trap notion should not be a shibboleth for inflationary policies, moderate that they may be.

FOOTNOTE

1Pre-1947 data are taken from Historical Statistics (1975: 210-11).

FOOTNOTE

2Wholesale prices rose to their July 1929, immediate pre-Great Depression, level in February 1942.

3As for aggregate output, real GNP declined 11 percent behveen 1937:2 and 1938:1. This 15 percent annual rate was larger than any yearly decline in the 1929-33 period. The data are from Balke and Gordon (1986).

4Since prices rose a bit in the first two months of the depression, the actual price decline was 15 percent over the period.

FOOTNOTE

5Due to a reclassification of business loans at the end of 1937, there is a break in the series; hence the behavior of such loans over the entire period is not comparable.

6They rose for another year, reaching $6.9 billion in October 1940.

FOOTNOTE

7A formal analysis of the role of monetary expansion as the principal factor spurring recovery from the 1937-38 depression is in Burbidge and Harrison (1985).

8The effective gold sterilization policy in the first nine months of 1937 contributed to the depression, by reducing the base's growth in the face of increases in the demand for the base arising from increases in reserve requirements.

REFERENCE

References

REFERENCE

Balke, N.S., and Gordon, R.J. (1986) "Appendix B: Historical Data" In R.J. Gordon (ed.) The American Business Cycle. Chicago: University of Chicago Press: 781-850.

Bernanke, B.S. (1983) "Nonmonetary Effects of the Financial Crisis in the

REFERENCE

Propagation of the Great Depression." American Economic Review 73 (June): 257-76.

Burbidge, J., and Harrison, A. (1985) "An Historical Decomposition of the Great Depression to Determine the Role of Money." journal of Monetary Economics 16 (July): 45-54.

Clouse, J., et al. (1999) "Monetary Policy When the Nominal Short-Term Interest Rate Is Zero." Board of Governors of the Federal Reserve System, unpublished.

REFERENCE

DeLong, J.B. (1997) "America's Peacetime Inflation: The 1970s." In C. Romer and D. Romer (eds.) Reducing Inflation: Motivation and Strategy, 247-80. Chicago: University of Chicago Press.

Favero, C.A.; Giavazzi, F.; and Flabbi, L (1999) "The Transmission Mechanism of Monetary Policy in Europe: Evidence from Banks' Balance Sheets." NBER Working Paper 7231 (July).

Fisher, I. (1932) Booms and Depression.s. New York: Adelphi.

Krugman, P. (1998) "It's Baaack! Japan's Slump and the Return of the Liquidity Trap." Brookings Papers on Economic Activity 2: 137-205.

REFERENCE

McCallum, B.T. (2000) "Theoretical Analysis Regarding a Zero Bound on Nominal Interest Rates." NBER Working Paper 7677 (April).

Poole, W. (1999) "Is Inflation Too Low?" Cato Journal 18 (Winter): 453-64.

Reifschneider, D., and Williams, J. (1999) "Three Lessons for Monetary Policy in a Low Inflation Era." Board of Governors of the Federal Reserve System, unpublished.

U.S. Bureau of the Census (1975) Historical Statistics of the United States. Washington, D. C.: Government Printing Office.

REFERENCE

Wheelock, D.C. (1998) "Monetary Policy in the Great Depression and Beyond: The Sources of the Fed's Inflation Bias." In M. Wheeler (ed.) The Economics of the Great Depression, 127-69. Kalamazoo, Mich.: W.E. Upjohn Institute for Employment Research.

AUTHOR_AFFILIATION

Frank G. Steindl is Regents Professor of Economics and Ardmore Professor of Business Administration at Oklahoma State University.