Tuesday, September 27, 2011

Looking At Another Fed Hawk's Argument

Last week I looked at Fed President Fisher's argument for not engaging in further Fed policy easing.  While part of his argument made sense (the Fed had already done all it could do) part of it did not (that regulatory burdens were hindering a hiring increase).  Today I want to look at Minneapolis Fed President Narayana Kocherlakota's argument for not engaging in further Fed policy easing.
Much of my discussion so far has been a look back over the past three and a half years, since the start of the Great Recession in December 2007. My assessment is that, despite some profound economic shocks, the FOMC—led by Chairman Bernanke—has successfully met its price stability mandate by engaging in imaginative forms of monetary accommodation and thereby helped lower the unemployment rate. Now I’d like to turn to my assessment of the most recent round of FOMC decision-making. To put this part of my talk in the proper context, I want to ask another key question: How did the FOMC achieve its success over the past three and a half years with regard to price stability?

The answer, I believe, is that the FOMC consistently made choices in response to changes in short-term economic conditions that were designed to support its medium-term objectives. Getting these choices right is certainly more of an art than a science. With that said, economists have suggested a number of rules that tell central banks how to respond to changes in economic conditions so as to keep inflation near some target level. I generally find these rules useful in guiding policymaking, and especially so when they arrive at the same recommendation. (Unfortunately, that’s not always the case.)

But here’s one instance in which most of the rules do deliver the same recommended course of action. Suppose the FOMC observes an increase in available measures of inflationary pressures and a decrease in labor market slack—that is, the gap between maximum employment and observed employment. Then many monetary policy rules would recommend that the FOMC not ease policy further and in fact consider reducing the level of monetary policy accommodation. That recommendation—don’t ease further if you’re doing better on your mandates—makes sense to me.
With that recommendation in mind, let’s go back to November 2010. At that date, the FOMC took a significant policy step by announcing its intention to buy $600 billion of longer-term Treasury securities. Until the most recent meeting in August, this was the last major policy step undertaken by the Committee. What did available measures of inflationary pressures and labor market slack—the “mandate dashboard”—look like back in November?

In terms of inflation, I generally think that core inflation does a better job of tracking underlying inflationary pressures, because it does not include the highly volatile and transitory fluctuations in food and energy prices. In November, PCE core inflation over the preceding 12 months had been less than 1 percent and had decelerated throughout the year. Of course, a good mandate dashboard should also include some measure of the future course of inflationary pressures. Here, it is worth noting that, even with the large-scale asset purchase in place, FOMC participants expected core inflation to remain very low: less than 1.3 percent over the upcoming calendar year of 2011.

In terms of labor market slack, I’ve argued elsewhere that it’s hard to find reliable measures of this key variable.8 But the FOMC statement makes specific reference to the unemployment rate as a gauge of labor market slack, and so I’ll use that measure on my notional mandate dashboard. The unemployment rate was 9.8 percent in November 2010. With the help of the large-scale asset purchase, FOMC participants expected it to fall to about 9 percent a year hence.
 Let's stop and take a more thorough look at the situation in 2010.   First, note there are two variables he's looking at: inflation (as expressed by the PCE core inflation number) and labor market slack (as defined by the unemployment rate).  At this time, inflation is running at less than 1.3% and unemployment is at 9.8%.  From his perspective, inflation is low and labor slack is high, so it's OK for the Fed to act.
How had the mandate dashboard changed in August 2011? PCE core inflation rose sharply: From December 2010 through July 2011, the annualized core PCE inflation rate was over 2 percent. FOMC participants did not submit forecasts of core PCE inflation in August. However, the most recent Survey of Professional Forecasters, done before the August FOMC meeting, predicted that core PCE inflation will average 1.7 percent in 2011 and 1.6 percent in 2012. It seems clear that inflationary pressures were higher in August than in November. My own current forecast for core PCE inflation is even higher than the SPF’s—I expect that it will average around 2 percent per year over 2011 and 2012.

What about labor market slack? The unemployment rate was 9.1 percent in August 2011, as opposed to 9.8 percent in November 2010. Again, we don’t have FOMC participant projections available from the August meeting. However, the Survey of Professional Forecasters predicts that unemployment will be 8.6 percent in just over a year’s time. Going into the August FOMC meeting, my own forecast for unemployment was a little more optimistic, in the sense that I do expect unemployment to be under 8.5 percent by the end of next year. But, even with the more pessimistic SPF forecast, labor market slack is smaller than in November 2010, when the FOMC expected unemployment to remain around 9 percent in a year’s time.

So, measures of past and forecasts of future inflationary pressures were higher in August than at the time of the FOMC’s last major policy move in November. Measures of current labor market slack and expectations of future labor market slack were smaller in August. The monetary policy rules that I described earlier would suggest, again, “Don’t ease further if you’re doing better on your mandates.” Indeed, they’d recommend that the level of policy accommodation be reduced.
Flash forward to the last meeting when core PCE was over 2% and unemployment had moved lower to 9.1% with further projected increases in PCEs and declines in unemployment over the next 6-12 months.  The question now becomes: is this an appropriately high level of inflation to be concerned about higher prices and a declining unemployment rate to think more can not be done?  As I've noted, inflation is running at hot levels, so I think this is a valid concern.  Most importantly, I'm concerned about food prices maintaining increased pressure on the CPI number.  However, arguing there is a sufficient drop in unemployment to indicate a decreasing of labor market slack is a non-starter -- especially considering the length of time of many of the unemployed.  Simply put, unemployment is still way too high to considering making this argument viable.  As such, I don't think it holds up as indicating slack in decreasing in a sufficient degree.

I will add that I think if he had stuck with inflation, and then pointed to overall CPI plus food prices (which are running very hot as noted at the above link) he would have been on sound ground.