So Why Did Unemployment Fall?

Kash decides to be the first to post on AB’s new Hook ‘em Horns color scheme – so it’s time for California Angrybear to stop whining about USC’s loss and start whining about the weak labor market. Let me provide another snippet from the BLS report:

The employment-population ratio held at 62.8 percent in December, 0.4 percentage point higher than a year earlier. The labor force participation rate, at 66.0 percent, was unchanged over the year.

While it is true that the increase in employment per the Household Survey was somewhat higher than the disappointing increase per the Payroll Survey, the employment-population ratio has flat lined at 62.8%. So why did the unemployment rate fall? Simple – because the labor force participation rate fell from 66.1% to 66.0%. But BLS decides to report the change over the entire year, which is an excuse for me to update one of my favorite graphs – reporting the labor force participation rate and the employment-population ratio over a period longer than just a year. In fact, let’s graph both ratios over the past decade.

It is true that the employment-population ratio has risen a bit since the summer of 2003, which is one reason why the unemployment rate has fallen. But the other reason has to do with the decline in the labor force participation rate. If one compares the labor market situation today to that of five years ago, both ratios are significantly lower. Now we have had a debate as to what these developments mean in terms of whether we or close to full employment or not. Some of us on the left think much of the decline in the labor force participation rate is due to a discouraged worker effect, while some on the right would argue that some of the decline is attributable voluntary reductions. Perhaps the reality lies somewhere in the middle, but one thing is clear: a smaller percentage of Americans have jobs today than they did five years ago even if the Administration’s minions try to tout the current labor market as the best ever.

Update: Michael Darda finds reasons not to be as disappointed as Kash was:

The unemployment rate dropped to 4.9 percent from 5 percent, which is consistent with rising real wages over time. At the same time, 12-month job growth in the department’s household survey (a broad employment gauge that includes start-ups and the self-employed) remains considerably larger (217,000) than the 12-month job gains recorded by the more-often cited establishment (or payroll) survey (168,000). History shows that such gaps typically unwind with stronger payroll job growth.

Darda makes two empirical claims here: (1) real wages rise when the unemployment rate is around 5%; and (2) whenever the increase in the household survey employment figure exceeds the increase in the payroll survey employment – the latter is about to accelerate. Of course, he provides neither evidence nor context to either claim.

Darda also makes this prediction:

As the recovery deepens and broadens during 2006, I expect the unemployment rate to drop further …

Of course, he does not tell us if his forecast is predicated on a further decline in the labor force participation rate.

Update II: Mark Thoma provides a very nice follow-up on the discouraged worker effect as he realized that the BLS report deserves more credit than I gave it after my initial skim read of it.