EH.net is owned and operated by the Economic History Association with the support of other sponsoring organizations.
Pierre L. Siklos, Wilfrid Laurier University
What is Deflation?
Deflation is a persistent fall in some generally followed aggregate indicator of price movements, such as the consumer price index or the GDP deflator. Generally, a one-time fall in the price level does not constitute a deflation. Instead, one has to see continuously falling prices for well over a year before concluding that the economy suffers from deflation. How long the fall has to continue before the public and policy makers conclude that the phenomenon is reflected in expectations of future price developments is open to question. For example, in Japan, which has the distinction of experiencing the longest post World War II period of deflation, it took several years for deflationary expectations to emerge.
Most observers tend to focus on changes in consumer or producer prices since, as far as monetary policy is concerned, central banks are responsible for ensuring some form of price stability (usually defined as inflation rates of +3% or less in much of the industrial world). However, sustained decreases in asset prices, such as for stock market shares or housing, can also pose serious economic problems since, other things equal, such outcomes imply lower wealth and, in turn, reduced consumption spending. While the connection between goods price and asset price inflation or deflation remains a contentious one in the economics profession, policy makers are undoubtedly worried about the existence of a link, as Alan Greenspan's "irrational exuberance" remark of 1996 illustrates.
Historical and Contemporary Worries about Deflation
Until 2002, prospects for a deflation outside Japan remained unlikely. Prior to that time, deflation had been a phenomenon primarily of the 1930s and inextricably linked with the Great Depression, especially in the United States. Most observers viewed Japan's deflation as part of a general economic malaise stemming from a mix of bad policy choices, bad politics, and a banking industry insolvency problem that would simply not go away. However, by 2001, reports of falling US producer prices, a sluggish economy, and the spread of deflation beyond Japan to China, Taiwan, and Hong Kong, to name a few countries, eventually led policy makers at the US Federal Reserve Board to publicly express their determination at avoiding deflation (e.g. See IMF 2003, Borio and Filardo 2004). Governor Bernanke of the US Federal Reserve raised the issue of deflation in late 2002 when he argued that the US public ought not to be overly worried since the Fed was on top of the issue and, in any event, the US was not Japan. Nevertheless, he also stressed that "central banks must today try to avoid major changes in the inflation rate in either direction. In central bank speak, we now face "symmetric" inflation risks."1 The risks Governor Bernanke was referring to stem from the fact that, now that low inflation rates have been achieved, the public has to maintain the belief that central banks will neither allow inflation to creep up nor permit the onset of deflation. Even the IMF began to worry about the likelihood of deflation, as reflected in a major report, released in mid-2003, that assessed the probability that deflation might become a global phenomenon. While the risk that deflation might catch on in the US was deemed fairly low, the threat of deflation in Germany, for example, was viewed as being much greater.
Deflation in the Great Depression Era
It is evident from the foregoing illustrations that deflation has again emerged as public policy enemy number one in some circles. Most observers need only think back to the global depression of the 1930s, when the combination of a massive fall in output and the price level devastated the U.S. economy. While the Great Depression was a global phenomenon, actual output losses varied considerably from modest losses to the massive losses incurred by the U.S. economy. During the period 1928-1933 output fell by approximately 25% as did prices. Other countries, such as Canada and Germany, also suffered large output losses. Canada also experienced a fall in output of at least 25% over the same period while prices in 1933 were only about 78% of prices in 1928. In the case of Germany, the deflation rate over the same 1928-1933 period was similar to that experienced in Canada while output fell just over 20% in that time. No wonder analysts associate deflation with "ugly" economic consequences. Nevertheless, as shall see, there exist varieties of deflationary experiences. In any event, it needs to be underlined that the Great Depression period of the 1930s did not result in massive output losses worldwide. In particular, seminal analyses by Friedman and Schwartz (1982), and Meltzer (2003), concluded that the 1930s represented a deflationary episode driven by falling aggregate demand, compounded by poor policy choices by the leadership at the US Federal Reserve that was wedded at the time to a faulty ideology (a version of the 'real bills' doctrine2). Indeed, the competence of the interwar Fed has been the subject of considerable ongoing debate throughout the decades. Disagreements over the role of credit in deflation and concerns about how to reinvigorate the economy were, of course, also expressed in public at the time. Strikingly, the relationship between deflation and central bank policy was often entirely missing from the discussion, however.
The Debt-Deflation Problem
The prevailing ideology treated the occasional deflation as one that acted as a necessary spur for economic growth, a symptom of economic health, not one indicative of economic malaise. However, there were notable exceptions to the chorus of views favorable to deflation. Irving Fisher developed what is now referred to as the "debt-deflation" hypothesis. Falling prices increase the debt burden and adversely affect firms' balance sheets. This was particularly true of the plight faced by farmers in many countries, including the United States, during the 1920s when falling agricultural prices combined with tight monetary policies sharply raised the costs of servicing existing debts. The same was true of the prices of raw materials. The Table below illustrates the rather precipitous drop in the price level of some key commodities in a single year.
Commodity Prices in the U.S., 1923-24
|Commodity Group||May 1924||May 1923|
|Clothes and Clothing||187||201|
|Fuel and Lighting||177||190|
|Chemicals and drugs||127||134|
Source: Federal Reserve Bulletin, July 1924, p. 532.
The Postponing Purchases Problem
Hence, a deflation is a harbinger of a financial crisis with repercussions for the economy as a whole. Others, such as Keynes, also worried about the impact of deflation on aggregate demand in the economy, as individuals and firms postpone their purchases in the hopes of purchasing durable goods especially at lower future prices. He actually advocated a policy that is not too dissimilar to what we would refer to today as inflation targeting (e.g., see Burdekin and Siklos 2004, ch. 1). Unfortunately, the prevailing ideology was that deflation was a purgative of sorts, that is, the price to be paid for economic excesses during the boom years, and necessary to establish to conditions for economic recovery. The reason is that economic booms were believed to be associated with excessive inflation which had to be rooted out of the system. Hence, prices that rose too fast could only be cured if they returned to lower levels.
Not All Deflations Are Bad
So, are all deflations bad? Not necessarily. The United Kingdom experienced several years of falling prices in the 1870-1939. However, since the deflation was apparently largely anticipated (e.g., see Capie and Wood 2004) the deflation did not produce adverse economic consequences. Finally, an economy that experiences a surge of financial and technological innovations would effectively see rising aggregate supply that, with only modest growth in aggregate demand, would translate into lower prices over time. Indeed, estimates based on simple relationships suggest that the sometime calamitous effects that are thought to be associated with deflation can largely be explained by the rather unique event of the Great Depression of the 1930s (Burdekin and Siklos 2004, ch. 1). However, the other difficulty is that a deflation may at first appear to be supply driven until policy makers come to the realization that aggregate demand is the proximate cause. This seems to be the case of the modern-day episodes of deflation in Japan and China.
Differences between the 1930s and Today
What's different about the prospects of a deflation today? First, and perhaps most obviously, we know considerably more than in the 1930s about the transmission mechanism of monetary policy decisions. Second, the prevailing economic ideology favors flexible exchange rates. Almost all of the countries that suffered from the Great Depression adhered to some form of fixed exchange rates, usually under the aegis of the Gold Standard. As a result, the transmission of deflation from one country to another was much stronger than under flexible exchange rate conditions. Third, policy makers have many more instruments of policy today than seventy years ago. Not only can monetary policy be more effective when correctly applied, but fiscal policy exists on a scale that was not possible during the 1930s. Nevertheless, fiscal policy, if misused, as has apparently been the case in Japan, can actually add to the difficulties of extricating an economy out of deflationary slump. There are similar worries about the US case as the anticipated surpluses have turned into large deficits for the foreseeable future. Likewise the fiscal rules adopted by the European Union severely hinder, even altogether prevent some would say, the scope for a stimulative fiscal policy. Fourth, policy-making institutions are both more open and accountable than in past decades. Central banks are autonomous and accountable and their efforts at making monetary policy more transparent to financial markets ought to reduce the likelihood of serious policy errors as these are considered to be powerful devices to enhance credibility.
Parallels between the 1930s and Today
Nevertheless, in spite of the obvious differences between the situation today and the ones faced seven decades ago, some parallels remain. For example, until 2000, many policy makers, including the central bankers at the Fed, felt that the technological developments of the 1990s might lead to economic growth almost without end and, in this "new" era, the prospect of a bad deflation seemed the furthest thing on their minds. Similarly, the Bank of Japan was long convinced that their deflation was of the good variety. It has taken its policy makers a decade to recognize the seriousness of their situation. In Japan, the debate over the menu of needed reforms and policies to extricate the economy from its deflationary trap continues unabated. Worse still the recent Japanese experience raises the specter of Keynes' famous liquidity trap (Krugman 1998), namely a state of affairs where lower interest rates are unable to stimulate investment or economic activity more generally. Hence, deflation, combined with expectations of falling prices, conspires to make the so-called 'zero lower bound' for nominal interest rates an increasingly binding one (see below).
Two More Concerns: Labor Market and Credit Market Impacts of Deflation
There are at least two other reasons to worry about the onset of a deflation with devastating economic consequences. Labor markets exhibit considerably less flexibility than several decades ago. Consequently, it is considerably more difficult for the necessary fall in nominal wages to match a drop in prices. Otherwise, real wages would actually rise in a deflation and this would produce even more slack in the labor market with the resulting increases in the unemployment rate contributing to further reduce aggregate demand, the exact opposite of what is needed. A second consideration is the ability of monetary policy to stimulate the economy when interest rates are close to zero. The so-called "zero lower bound" constraint for interest rates means that if the rate of deflation rises so do real interest rates further depressing aggregate demand. Therefore, while history need not repeat itself, the mistakes of the past need to be kept firmly in mind.
Frequency of Deflation in the Historical Record
As noted above, inflation has been an all too common occurrence since 1945. The table below shows that deflation has become a much less common feature of the macroeconomic landscape. One has to go back to the 1930s before encountering successive years of deflation.3 Indeed, for the countries listed below, the number of times prices fell year over year for two years or more is a relatively small number. Hence, deflation is a fairly unusual event.
Table 2: Episodes of Deflation from the mid-1800s to 1945
(year record begins)
|Years of persistent deflation/Crisis|
|Australia (1862)||5||BC: 1893
BC, CC: 1924-26
1931-35, BC: 1931
|Canada (1914)||2||CC: 1891,1893, 1908, 1921, 1929-31
|Denmark (1851)||9||1882-86, BC: 1885, 1892-96, BC: 1907-08
1921-32, BC, CC: 1921-22, 1931-32
|Finland (1915)||1||BC: 1900,1921 1929-34, BC, CC: 1931-32|
|France (1851)||4||CC, BC: 1888-89, 1907, 1923, 1926
1932-35, BC: 1930-32
|Germany (1851)||8||1892-96, BC, CC:1893, 1901,1907
1930-33, BC, CC: 1931, 1934
|Italy (1862)||6||1881-84, BC, CC: 1891, 1893-94, 1907-08, 1921
1930-34, BC: 1930-31, 1934-35
|Japan (1923)||1||CC, BC: 1900-01, 1907-08, 1921
1925-31, BC, CC: 1931-32
|Netherlands (1881)||6||1893-96, BC, CC: 1897, 1921
CC, BC: 1935, 1939
|Norway (1902)||2||BC, CC: 1891, 1921-23
1926-33, BC CC: 1931
|New Zealand (1908)||1||BC: 1920, 1924-25
1929-33, BC, CC: 1931
Notes: Data are from chapter 1, Richard C.K. Burdekin and Pierre L. Siklos, editors, Deflation: Current and Historical Perspectives, New York: Cambridge University Press, 2004. The numbers in parenthesis in the first column refer to the first year for which we have data. The second column gives the frequency of occurrences of deflation defined as two or more consecutive years with falling prices. The last column provides some illustrations of especially persistent declines in the price level, defined in terms of consumer prices. In italics, years with currency crises (CC) or banking crises (BC), are shown where data are available. The dates are from Michael D. Bordo, Barry Eichengreen, Daniela Klingebiel, and Maria Soledad Martinez-Peria, "Financial Crises: Lessons from the Last 120 Years," Economic Policy, April 2001.
Is There an Empirical Deflation-Recession Link?
If that is indeed the case why has there been so much concern expressed over the possibility of renewed deflation? One reason is the mediocre economic performance that has been associated with the Japan's deflation. Furthermore, the foregoing table makes clear that in a number of countries the 1930s deflation was associated with the Great Depression. Indeed, as the Table also indicates for countries where we have data, the Great Depression represented a combination of several crises, simultaneously financial and economic in nature. However, it is also clear that deflation need not always be associated either with a currency crisis or a banking crisis. Since the Great Depression was a singularly devastating event from an economic perspective, it is not entirely surprising that observers would associate deflation with depression.
But is this necessarily so? After all, the era roughly from 1870 to 1890 was also a period of deflation in several countries and, as the figure below suggests, in the United States and elsewhere, deflation was accompanied by strong economic growth. It is what some economists might refer to as a "good" deflation since it occurred at a time of tremendous technological improvements (in transportation and communications especially). That is not to say, even under such circumstances, that opposition from some quarters over the effects of such developments was unheard of. Indeed, the deflation prompted some, most famously William Jennings Bryan in the United States, to run for office believing that the Gold Standard's proclivity to create deflation was akin to crucifying "mankind upon a cross of gold." In contrast, the Great Depression would be characterized as a "bad" or even "ugly" deflation since it is associated with a great deal of slack in the economy.
Prices Changes versus the Output Gap, 1870s and 1930s
Notes: The top figure plots the rate of CPI inflation for the periods 1875-79 and 1929-33 for the United States. The bottom figure is an estimate of the output gap for the U.S., that is, the difference between actual and potential real GDP. A negative number signifies actual real GDP is higher than potential real GDP and vice-versa when the output gap is positive. See Burdekin and Siklos (2004) for the details. The vertical line captures the gap in the data, as observations for 1880-1929 are not plotted.
Whereas policy makers today speak of the need to avoid deflation their assessment is colored by the experience of the bad deflation of the 1930s, and its spread internationally, and the ongoing deflation in Japan. Hence, not only do policy makers worry about deflation proper they also worry about its spread on a global scale.
If ideology can blind policymakers to introducing necessary reforms then the second lesson from history is that, once entrenched, expectations of deflation may be difficult to reverse. The occasional fall in aggregate prices is unlikely to significantly affect longer-term expectations of inflation. This is especially true if the monetary authority is independent from political control, and if the central bank is required to meet some kind of inflation objective. Indeed, many analysts have repeatedly suggested the need to introduce an inflation target for Japan. While the Japanese have responded by stating that inflation targeting alone is incapable of helping the economy escape from deflation, the Bank of Japan's stubborn refusal to adopt such a monetary policy strategy signals an unwillingness to commit to a different monetary policy strategy. Hence, expectations are even more unlikely to be influenced by other policies ostensibly meant to reverse the course of Japanese prices. The Federal Reserve, of course, does not have a formal inflation target but has repeatedly stated that its policies are meant to control inflation within a 0-3% band. Whether formal versus informal inflation targets represent substantially different monetary policy strategies continues to be debated, though the growing popularity of this type of monetary policy strategy suggests that it greatly assists in anchoring expectations of inflation.
Borio, Claudio, and Andrew Filardo. "Back to the Future? Assessing the Deflation Record." Bank for International Settlements, March 2004.
Burdekin, Richard C.K., and Pierre L. Siklos. "Fears of Deflation and Policy Responses Then and Now." In Deflation: Current and Historical Perspectives, edited by Richard C.K. Burdekin and Pierre L. Siklos. New York: Cambridge: Cambridge University Press, 2004.
Capie, Forrest, and Geoffrey Wood. "Price Change, Financial Stability, and the British Economy, 1870-1939." In Deflation: Current and Historical Perspectives, edited by Richard C.K. Burdekin and Pierre L. Siklos. New York: Cambridge: Cambridge University Press, 2004.
Friedman, Milton, and Anna J. Schwartz. Monetary Trends in the United States and the United Kingdom. Chicago: University of Chicago Press, 1982.
Humphrey, Thomas M. "The Real Bills Doctrine." Federal Reserve Bank of Richmond Economic Review 68, no. 5 (1982).
International Monetary Fund. "Deflation: Determinants, Risks, and Policy Options "Findings of an Independent Task Force." April 30, 2003.
Krugman, Paul. "Its Baaaaack: Japan's Slump and the Return of the Liquidity Trap." Brookings Papers on Economic Activity 2 (1998): 137-205.
Meltzer, Allan H. A History of the Federal Reserve. Chicago: Chicago University Press, 2003.