The Impact of GDP Revisions on Taylor Rule Estimations
Along with July's advanced estimate for second-quarter GDP, the annual revisions for previous GDP estimates were released. Revisions showed a dramatically lower path for GDP than had been previously estimated. In fact, after revisions, real GDP is now believed to still be below pre-recession levels.
This deeper dip in GDP is a more accurate picture of the actual economic conditions experienced throughout the recession. Less dramatically, inflation as measured by core PCE inflation was also revised.
We look at how these revisions could impact policy using what is known as the Taylor rule. The Taylor rule is one of the most common tools used to evaluate Fed policy because it suggests what the federal funds rate should be and compares it to actual rates to get some insight into monetary policy decision making. The traditional rule supposes that the Fed increases rates when inflation increases and decreases rates when the output gap gets larger (the output gap is the difference between potential and actual GDP).
We estimate the rule using the four-quarter change in the core PCE price index as our measure of inflation and a traditional calculation for our measure of the output gap. That traditional calculation shows by what percentage output is above or below its potential. Since potential output is unobservable, we use the Congressional Budget Office's August 2011 estimates for it. The revisions to GDP have so far had a very minimal impact on those estimates, so our output gap calculation has declined since 2010. It is possible that the CBO will revise down its potential estimates at some future date.
Results are shown in the chart below. They indicate that policy has been constrained by the zero lower bound since the end of 2008. Before the GDP revisions and without the zero lower bound restriction, the Taylor rule suggests that the federal funds rate should have been approximately 2 percentage points lower than it was, that is, around −2 percent. After the revisions, the large increase in the output gap suggests that the rate should be 3 percentage points lower than it is currently. That is, the revisions themselves would cause nearly a 1 percentage point drop in the rule's estimate for the interest rate.
At the August Federal Open Market Committee meeting, which occurred shortly after these revisions, the committee strengthened its forward-looking guidance and replaced its "extended period" language with "The Committee currently anticipates that economic conditions—including low rates of resource utilization and a subdued outlook for inflation over the medium run—are likely to warrant exceptionally low levels for the federal funds rate at least through mid-2013."
However, an almost unprecedented number of bank presidents dissented. "Voting against the action were: Richard W. Fisher, Narayana Kocherlakota, and Charles I. Plosser, who would have preferred to continue to describe economic conditions as likely to warrant exceptionally low levels for the federal funds rate for an extended period."
But the statement was not explicit about what the Committee meant by "low rates of resource utilization." There are many measures of this, but two common measures are the output gap, which we used in our estimation of the Taylor rule above, and the unemployment rate.
Narayana Kocherlakota was one of the dissenters and seems to favor the latter measure of resource utilization. He noted that "In particular, personal consumption expenditure (PCE) inflation rose notably in the first half of 2011, whether or not one includes food and energy. At the same time, while unemployment does remain disturbingly high, it has fallen since November."
This suggests that another way of looking at policy is to use the unemployment rate instead of the output gap in the Taylor rule. One reason that many prefer this specification is that potential output and thus the output gap is unobservable. But others prefer the output gap because they argue that policy needs to be more forward-looking and that unemployment has traditionally lagged changes in GDP.
With an output gap as currently estimated of almost 7 percent of GDP and an unemployment rate of 9.1 percent, both of these rules suggest a high degree of policy accommodation is warranted. But since unemployment was not revised when GDP was, the degree of policy accommodation called for with an unemployment-based Taylor rule did not change after the revisions. (There were minor changes in this rule since the estimate of core PCE inflation did change.)
As we were looking at the gap-based rule above, we noted that the revisions to GDP had almost no impact on the CBO's estimate of potential, so our output gap measure declined significantly. The opposite extreme would be where the revisions in GDP did lower potential GDP, which would help close the output gap. The unemployment-based Taylor rule is similar to that extreme.