از رکود بزرگ به تورم بزرگ : وابستگی به مسیر و سیاست پولی

There is substantial narrative evidence that the shadow of the Great Depression may have influenced the conduct of U.S. monetary policy during the 1970s. In this paper, we estimate central bank reaction functions for the United States and 12 other countries over the 1970s to examine the relationship between the magnitude of the Great Depression and the response of central banks to output gaps and inflation during the Great Inflation. The main finding is that countries which suffered the most during the 1930s had monetary policy reaction functions that responded substantially more aggressively to output gaps during the 1970s.

The period from 1965 to 1979, known as the Great Inflation, has been described as “America's only peacetime outburst of inflation” (DeLong, 1997, p. 247) as well as “the greatest failure of American macroeconomic policy in the postwar period” (Mayer, 1999, p. 1). In recent years, many potential reasons for this long and persistent increase in U.S. inflation have been proposed. Some of these explanations emphasize bad luck associated with exogenous non-policy shocks while others focus on the conduct of monetary policy.2 While macroeconomists have long been interested in modeling how the Federal Reserve responds to economic conditions, following Taylor (1993), it has become common practice to characterize Federal Reserve behavior in terms of policy reaction functions where the central bank sets the federal funds rate in response to deviations of the inflation rate from a desired target, as well as in response to deviations of output from potential. Many authors contend that an important cause of the increased inflation was that the Federal Reserve did not respond aggressively enough to inflation prior to 1979. Using the reaction function framework, Taylor (1999) argues that before 1979 the Federal Reserve did not raise nominal interest rates more than one-for-one when inflation increased.3 As a result, real interest rates decreased when inflation increased which fueled further spending and higher inflation.4Judd and Rudebusch (1998) and Clarida, Gali, and Gertler (2000) reach similar conclusions for the pre-1979 period, using modified policy rules that take into account the partial-adjustment dynamics of interest rates and forward-looking behavior. All of these papers conclude that an important reason for the inflation during the 1970s was that the Fed did not react aggressively enough to inflation: the inflation-response coefficient was less than one prior to 1979, but substantially greater than one after 1979. The view that the Federal Reserve did not respond vigorously enough to inflation before 1979 has been questioned, however. Using only data available to policymakers at the time, Perez (2001) finds that the response to inflation was roughly similar between the pre-1979 and post-1983 periods. In both periods, the inflation-response coefficient is greater than one. Similarly, Orphanides (2004) reports that the Federal Reserve responded more than one-for-one to inflation in both the 1966–1979 and 1979–1995 periods, based on real-time data. Orphanides (2004), however, does find that the Federal Reserve responded much more aggressively to output gaps before 1979. While an aggressive response to output gaps is stabilizing with perfect information, Orphanides, Porter, Reifschneider, Tetlow, and Finan (2000) shows that a large response to output gaps can be destabilizing and lead to higher inflation in the presence of measurement error and noisy information. In particular, the authors find that the failure of policymakers to recognize the productivity slowdown that began by the early 1970s caused policymakers to systematically overestimate potential output and output gaps. The one-sided measurement error in the 1970s, along with the large response of policymakers to these mismeasured output gaps, led to excessive monetary ease, which contributed to the Great Inflation. The underlying cause as to why monetary policy “was too activist in reacting to perceived output gaps,” before 1979 is not examined in detail by Orphanides (2004, p. 151). DeLong (1997), Wheelock (1998), and Mayer (1999), however, all emphasize the impact of the Great Depression on policymakers during the 1970s.5DeLong (1997, p. 250) believes “the truest cause of the 1970s inflation was the shadow cast by the Great Depression” that caused policymakers to view any level of unemployment as “too high.” DeLong (1997, p. 257) contends that “if 4 percent unemployment had turned out to be the natural rate, the cry would have arisen for a reduction in unemployment to 2 percent.” In other words, the expansionary bias of the Federal Reserve, due to the memory of the Great Depression, led to the Great Inflation. Similarly, Wheelock (1998) contends that the Great Depression undermined the ideological commitment to the gold standard. As a result, during the 1970s the U.S. was able to abandon fixed exchange rates and pursue overly expansionary policy. The influence of the Great Depression on the conduct of monetary policy during the 1970s is also echoed by a member of the Federal Open Market Committee (FOMC) during that time. In a 1996 interview, Lyle Gramely said “[n]ow one can say looking back that all of us were suffering from the after-effects of having lived through the Great Depression …”.6 While the influence of the shadow of the Great Depression on monetary policy in the 1970s makes intuitive sense, none of these authors provide empirical evidence to support their assertions. In this paper, we use international data from both the 1930s and 1970s to investigate any potential connections between the severity of the Great Depression and central bank behavior during the Great Inflation. While the Great Depression of the 1930s was a worldwide phenomenon, there was also substantial international variation in its severity. In addition, other countries in the rest of the industrialized world also experienced, to a greater or lesser extent than the U.S., a burst of inflation during the 1970s (see Table 1). We estimate monetary policy reaction functions for the U.S. and 12 other advanced economies during the Great Inflation period. If the “shadow of the Great Depression” hypothesis is correct, we expect that countries where the depth and/or duration of the Great Depression were more severe should have postwar monetary policy reaction functions that respond more to output gaps. Although DeLong (1997) clearly emphasizes that the Great Depression caused policymakers to react strongly to real factors in setting interest rates, the impact of the Great Depression may have also caused central banks to react less aggressively to inflation. As a result, we also test whether there is any differential response to inflation across countries in the 1970s based on the severity of the Depression in the 1930s.The empirical evidence presented in this paper is consistent with the Great Depression hypothesis. We find that countries which suffered the most during the 1930s had central banks that responded significantly more to output gaps during the 1970s. This result holds for the entire sample, as well as for the subset of countries in the G-10, G-7, and G-3. There is also some evidence that the more severe was the Great Depression, the less central banks responded to inflation during the 1970s. This result, however, is only apparent for the G-3 countries, and it is not evident in the full sample. Finally, the length of time on the gold standard during the 1930s is also associated with central bank reactions to output gaps during the 1970s. Scholars have long recognized that the longer countries remained on the gold standard longer during the 1930s, the more severe was the Great Depression.7 We find that length of time on the gold standard in the 1930s is associated with both a more severe Great Depression and a more aggressive response to output gaps in the 1970s.8

The evidence presented in this paper strongly suggests that the shadow of the Great Depression likely influenced monetary policymakers during the 1970s. Countries that experienced larger and more persistent declines in output during the 1930s had central banks that responded more aggressively to output gaps during the 1970s. One limitation of this research, due to data unavailability, is that the central bank reaction functions are estimated using revised rather than real-time data. As long as there is no differential relationship between the policy response to output gaps and policymakers’ ability to measure such gaps across countries, the reported results can be thought of as unbiased measures of the true response. As Orphanides et al. (2000) have shown for the U.S., an aggressive response to output gaps in the presence of measurement errors and noisy information can be destabilizing and lead to higher inflation. If policymakers in other countries also failed to recognize the productivity slowdown that began by the early 1970s, then this would have led to a systematic overestimation of potential output and output gaps in other countries as well. This coupled with the relatively large central bank responses to output gaps could have contributed to the Great Inflation, particularly in those countries that experienced substantial output losses during the 1930s.