The Anatomy of Slow Recovery

BERKELEY – Between 1950 and 1990 – the days of old-fashioned inflation-fighting downturns engineered by the United States Federal Reserve – America’s post-recession unemployment rate would fall on average 32.4% over the course of a year from its initial value toward its natural rate. If the US unemployment rate had started to follow such a path after peaking in the second half of 2009, it would now stand at 8.3%, rather than 8.9%.

Unfortunately, none of the net reduction in the US unemployment rate over the past year came from increases in the employment-to-population ratio; all of it came from declining labor-force participation. Unemployment has fallen from 10.1% over the course of the past 18 months, but the employment-to-population ratio has remained stuck at 58.4%. Perhaps it would be better if unemployed people who could have jobs – and who at full employment would have them – were actively looking for work rather than out of the labor force completely.

If you take that view, between 1950 and 1990, the US employment-to-population ratio would rise an extra 0.227% annually on average for each year that the unemployment rate was above its natural rate. If the US employment-to-population ratio had started to follow such a path after its 2009 peak, the current ratio would be 59.7%, rather than 58.4%. (In that case, we would be experiencing “morning in America,” rather than the current state of economic malaise.)

This is, I think, the best metric to use to quantify the decidedly sub-par pace of today’s jobless recovery in the US. It is not out of line with other American yardsticks: since the output trough, real GDP has grown at an average rate of 2.86%/year, barely above the rate of growth of the US economy’s productive potential. And it is not out of line with the experience of other rich economies, whether Japan or in Europe.