One explanation for accommodative monetary policy in the 1970s is that the Federal Reserve responded less aggressively to inflation. Using real-time estimates of future inflation, I estimate Taylor rule type reaction functions for monetary policy. Contrary to results that rely on final, revised data, I show that the FOMC’s intended response to inflation was similar during both periods. Policy may have turned out to be accommodative largely because of the FOMC’s difficulty in forecasting inflation or because the Federal Reserve’s inflation target was higher in the 1970s, but not because they responded less aggressively to inflation.
It is a widely held belief that monetary policy was accommodative towards inflation in the 1970s and hawkish towards inflation after Paul Volcker took office as the Chairman of the Federal Reserve in 1979. This may have been due to an insufficiently strong response of the federal funds rate to inflation. Estimated forward-looking monetary policy reaction functions consistently show that the Federal Reserve’s response to future inflation was not strong enough. Using different specifications, Mehra (1999) and Clarida, Gali, and Gertler (2000) estimate that from 1960 to the second quarter of 1979, the federal funds rate increased by less than one percentage point for a one percentage point increase in expected inflation (dFFR/dπ < 1). This implies that the real federal funds rate actually falls when inflation rises, creating an accommodative monetary policy. Taylor (1999) describes why this constitutes poor formulation of monetary policy and suggests why the European Central Bank should use a policy rule with a response to inflation of greater than one:
In fact, a simple way to characterize the better monetary policy performance in the United States in the 1980s and 1990s compared with the 1960s and 1970s is that this response coefficient increased from below this stability threshold to above the threshold. If the European Central Bank chooses to use the short term interest rates as its instrument, then I believe that having a response coefficient greater than one will be the first step to achieving good performance. (Taylor 1999, p. 12)
In this paper, I show that the intention of the Federal Open Market Committee (FOMC) was to have the federal funds rate respond to changes in expected inflation in a nonaccommodative fashion.
The Federal Reserve Board Staff prepares data describing the state of the economy prior to each FOMC meeting. Some of this data, compiled in what is called the Greenbook, are purely historical data covering the path of real output, unemployment, inflation, etc., over time. However, a rich source of information is the Staff’s estimates of current and future values for these economic indicators. Perez (1999) suggests that researchers should use these real-time estimates to properly analyze the formulation of monetary policy. Athanasios Orphanides 1997, Orphanides 1999 and Orphanides 2000 has pioneered the use of real time data when studying the formulation of monetary policy. In particular, Orphanides (2000) introduces real-time data into the standard Taylor rule for monetary policy to compare the policy prescriptions using real-time and revised data. He shows that using the coefficients suggested by Taylor (1993), there is little difference between the FFR suggested by real-time data and revised data and thus the Federal Reserve’s accommodative policy cannot be attributed to its not using the Taylor rule. This paper differs from Orphanides (2000) in that I estimate a monetary policy reaction function with real-time data using the Taylor rule framework in an attempt to determine whether the Federal Reserve intended policy to be accommodative in the 1970s.
Both Perez (1999) and Mayer (1999) support the use of real-time data to proxy for the information set used by FOMC participants. Therefore, in this paper, I use real-time estimates of future inflation and the current output gap to estimate Federal Reserve reaction functions. In the next section, I describe the real-time data set and the construction of real-time estimates of the output gap. Next, I estimate Taylor rule reaction functions to investigate the degree to which monetary policy was intended to be accommodative towards expected inflation. I show that while the actual policy that was formulated by the FOMC was accommodative in the late 1970s, in that the response of the federal funds rate to changes in expected inflation was less than one, and became nonaccommodative after 1983, the intended policy of the FOMC was nonaccommodative in reacting to inflation over the time periods 1975.1 to 1979.2 and 1983.1 to 1993.4. Further there appears to be no change in the intended formulation of monetary policy over these time periods other than a decrease in the target inflation rate.1 Finally, I discuss possible reasons that the actual policy of the FOMC was accommodative. This is likely due to failure to accurately predict future inflation or a high inflation target prior to 1979.
