Initial claims: Green shoots or false signal?

Current economic reports are grim, with indicators such as capacity utilization and housing starts at levels consistent with a deep contraction. Yet, there is a widespread view among analysts that the official end date of this recession is not that far off, sometime in late 2009. If this consensus is correct, then there should be some signs in the next few months of bottoming out in the economy. One harbinger of economic improvement can be found in the weekly reports on initial claims for unemployment insurance.

Current economic reports are grim, with indicators such as capacity utilization and housing starts at levels consistent with a deep contraction. Yet, there is a widespread view among analysts that the official end date of this recession is not that far off, sometime in late 2009. If this consensus is correct, then there should be some signs in the next few months of bottoming out in the economy.

Fed Chairman Ben Bernanke has described these potential first signs as the "green shoots" of economic recovery. One harbinger of economic improvement can be found in the weekly reports on initial claims for unemployment insurance. These figures, released each Thursday for the prior week, are often volatile, so data watchers typically rely on a four-week moving average to smooth out the series.

The table (at Economy@W. P. Carey) illustrates the predictive power of the initial claims reports. In five previous recessions, the recessionary peak in the four-week moving average of initial claims has directly preceded the official end date of the recession, as set by the National Bureau of Economic Research.

It is important to recognize that a decrease in initial claims may not bring a decrease in total unemployment or the unemployment rate. For example, in the period after the recession end point in November of 1982, unemployment rates increased to 10.8 percent and stayed above 10 percent for the next six months, but initial claims by the newly unemployed fell in each of these months (and continued to decrease for 17 consecutive months).

How can unemployment rates go up when initial claims for unemployment are falling? As long as job growth is weak, unemployment rates will be high and rising as long as the number of workers newly losing jobs exceeds the number that are newly hired. Employment growth is necessary to reduce the ranks of the unemployed, and in a post-recession period slow growth keeps unemployment rates high.

Have initial unemployment claims peaked in the current recession? Analysts answer this question with a strong "maybe." The four-week moving average was 659,500 for the week ending April 4, and the average declined to 651,000 for April 11. The weekly reported numbers of new claims (not averaged) peaked in the week of March 28 at 674,000, and the weekly figures have decreased for two consecutive weeks.

The 674,000 figure for March 28 is the highest level of reported claims so far in this recession. It is early — way early — to determine if the four-week average of new claims for unemployment has peaked and is on the way down. The figures released for the next few Thursdays will tell us a lot more about whether we are seeing green shoots or merely false signals on the economy.