December 17, 2012
Last week the Fed announced that it would continue to maintain its zero interest rate policy until the unemployment rate fell below 6.5 percent. While the Fed has always targeted low unemployment in addition to low inflation as part of its legal mandate, this was the first time it had explicitly tied its monetary policy to an unemployment target instead of just an inflation target.
This decision has Robert Samuelson very worried. Samuelson warned that the last time the Fed tried to target both inflation and unemployment was in the 1970s and complains that this ended disastrously. Both parts of Samuelson’s claim are wrong.
In fact, in the decades since the 1970s the Fed has maintained a commitment to lowering unemployment in addition to inflation, even if its priority was always on the latter. It would be very difficult to explain the decision to lower interest rates in the 1995 and again in the recession in 2001, except by a concern over excessive unemployment. This concern is certainly reflected in the transcripts of the meetings from these years. The only difference between last week’s announcement and the Fed’s past actions was the decision to explicitly set an unemployment target for its monetary policy. In the past, analysts would have found it necessary to review the minutes and public statements by members of the Fed’s Open Market Committee to infer a target.
The notion that inflation just exploded out of the blue in the 1970s is also inaccurate, as an examination of price movements of that decade shows.
Source: Economic Report of the President (Table 63).
The chart shows the inflation rate over the last 40 years. The blue line shows the official CPI that is most widely used in public discussions and in most legal contracts. The red line shows the core CPI, which excludes energy and food prices. The green line shows an error adjusted core CPI. In the years prior to 1982, the official CPI used a homeownership component. This measure showed a much higher rate of inflation when mortgage interest rates spiked in the mid-70s and again in the late 70s. In 1982 the index was changed to include an owners’ equivalent rent component that imputes a rent for owner occupied homes. The green line applies this methodology to a core index going back to 1969, in effect showing what a core CPI using the current methodology would have looked like in that period.
As can be seen, there are two notable spikes in the 1970s. In the first, in 1975 the core error adjusted CPI hit 8.0 percent. In the second, the error adjusted core hit 10.0 percent in 1980. In both instances the run up in core prices was preceded by sharp jumps in food and energy prices. The major factor in both cases were big rises in oil prices, the first driven by an oil embargo led by Arab countries and the second by the cutoff of oil flows associated with the Iranian revolution. At the time the United States was far more dependent on oil than it is today, which means that a comparable spike in oil prices would have considerably less impact on prices than it would today.
The error in the CPI in the 1970s almost certainly contributed to the rise in the inflation rate during this period also. At the time, a large segment of the workforce had wage contracts that were explicitly tied to the official CPI. Many other contracts were also tied to the CPI. This means that if the CPI erroneously showed a higher rate of inflation, this error would be directly passed on in higher wages and prices, leading to higher actual inflation. (Alan Blinder and Janet Yellen made the same sort of argument about how much smaller changes in measurement had the effect of lowering actual inflation in the 1990s.) In short, there were two spikes in inflation in the 1970s, both around 4.0 percentage points using current methods, and both driven by extraordinary circumstances which are unlikely to be repeated.
However much Samuelson might be misguided in thinking that another spike in inflation might come up out of the blue, he is even more misguided in his assessment of the consequences. The bad story from the 1970s inflation was the recession of 1981-1982 that was used to bring down inflation. While it’s not clear that such a severe recession was necessary to bring inflation under control (inflation fell sharply everywhere in the world in these years, not just the United States), this recession was much less severe by most measures than the current downturn. While unemployment peaked at almost 11.0 percent in December of 1982, it was down to 8.3 percent by the end of the 1983 and 7.3 percent by the end of 1984. (This was in a context where full employment was thought to be over 6.0 percent unemployment.)
If we infer that in the unlikely event that we get another spurt of inflation that we will have to undergo a downturn similar to the 1981-82 recession to bring it under control, then we would still be in a better world than we are in today. In effect, Samuelson is telling a person suffering from an otherwise terminal case of cancer that the drug that would save him could lead to nausea. While Samuelson thinks this makes the choice of treatments a close call, most other people probably would not see it that way.
Comments