Parsing the Federal Reserve's Flimsy Forecasts

Federal Reserve policymakers met last week and again lowered their forecast of economic growth, while upping their 2012 employment forecast.

Earlier this year, at their April meeting, members of the Federal Open Market Committee (FOMC) released their central tendency forecast of real GDP growth for the fourth quarter of 2012 over the fourth quarter of 2011. It amounted to a mid-point of 3.85 percent. At their June meeting they lowered that 2012 estimate to 3.5 percent. At last week's meeting the economic growth forecast was again reduced, this time to 2.7 percent.

The revisions in the committee's unemployment rate projections were also dramatic. In April the forecast for the fourth quarter of 2012 was 7.75 percent. In June it was raised to 8.0 percent, and last week it was raised again, this time to a mid-point of 8.6 percent. (With such large revisions it's hard to take seriously the FOMC's forecasts of unemployment and growth for 2013 and beyond.)

In contrast, FOMC forecasts of headline and core inflation for 2012 have not changed much, staying within a central tendency range that does not bump above two percent. That's hardly surprising since the committee's inflation forecasts reflect the presumption that the Fed's actions will keep prices stable. (Keep in mind that the committee is forecasting selected consumer price statistics, not necessarily reality - but that's another subject for another time.)

In just seven months the FOMC's projection of the 2012 unemployment rate has been raised by nearly a percentage point, which prompts one to take a closer look at the assumptions behind the committee's latest forecast.

Unfortunately, the Fed doesn't publish FOMC projections of employment, weekly hours, or productivity, which are key to understanding and evaluating the forecasts of unemployment and output. So we will have to fill in the holes.

Nonfarm business sector productivity (output per hour) rose at a 3.1 percent annual rate in the third quarter of this year, following a poor first half. Annual increases amounted to 4.1 percent in 2010 and 2.3 percent in 2009. The Federal Reserve Bank of New York, in its Liberty Street Economics research blog, has published a productivity forecast for 2012 of about two percent.

Growth of the working-age population will add about 1.5 million new jobseekers to the labor force next year - a one percent increase. Weekly hours are not expected to change much.

Thus, the FOMC's forecast of economic growth next year is pretty much eaten up by the employment required just to hold the unemployment rate constant, after taking account of likely productivity growth. To lower the unemployment rate from 9.0 percent currently to the 8.6 percent forecast for the fourth quarter of next year would require an employment increase of more than a half million over and beyond what looks to be implicit in the FOMC's economic growth forecast. And that's assuming that the labor force participation rate doesn't rise. If it does, even higher employment growth would be required to accommodate the increased number of jobseekers.

So the bottom line is that the FOMC's projected GDP is not only unlikely to lower the unemployment rate next year, but could increase it.

This brings up another issue: why doesn't the FOMC publish its forecasts of employment?

The Fed's dual legislative mandate is to promote stable prices and maximum employment. Consistent with that mandate, the FOMC publishes inflation forecasts, but it does not publish employment forecasts. Instead of employment, the Fed forecasts the unemployment rate, but unemployment is an imperfect - and misleading - proxy for employment.

It's well known that not all joblessness is included in the official unemployment rate in weak labor markets. Many of the jobless leave the labor force and retreat to the sidelines causing the official unemployment rate to be understated. The proportions of official unemployment and "hidden" unemployment vary from cycle to cycle.

There is also the difficulty of trying to determine the structural component of unemployment. An increase in structural joblessness, a major concern currently, can't be accurately measured and by its nature is beyond the direct influence of monetary policy. If structural unemployment is mistaken for cyclical unemployment, monetary policymakers could become too accommodating and spur price inflation.

Finally, the definition of unemployment, particularly its job-seeking requirement, makes that data series more prone to measurement error than the relatively straightforward measures of employment. In the collection of data during household interviews, respondent errors occur. Analyses of response errors have shown that reported unemployment is understated in weak labor markets.

There is one school of thought that says the Fed's mandate should be restricted to price stability since monetary policy cannot affect long-term unemployment. But as long as the Fed does have an "employment" mandate, the FOMC should substitute employment for unemployment in its forecasts, not to mention its policy debates.

In his press conference last week Federal Reserve chairman Ben Bernanke said the Fed has contained inflation but has "fallen short on the unemployment side." Mr. Chairman, unemployment is not the Fed's job.