Falling Unemployment Rate Should Never Be Bad News

COMMENT

If you were to step back and look at the U.S. unemployment rate, you would see that it moves in big, sweeping arcs, up and down. Yes, there can be small month-to-month reversals, but it has a clearly defined trend for an obvious reason: it tracks the business cycle.

Is this time different? Is the decline in the jobless rate from a high of 10 percent in October 2009 to 7.6 percent last month a case of right direction, wrong reason, and therefore misleading?

That was the assertion two weeks ago when the Labor Department reported jobs data for March. Employment, as measured by the household survey, fell 206,000. The only reason the unemployment rate declined 0.1 percentage points was that 496,000 people dropped out of the labor force: They stopped looking for work. As a rule, an expanding economy draws more people into the labor force in the hope of finding a job. So what seemed like good news on the surface was really bad news, we were told.

That assessment may be true in the short term, but in the long run it matters very little why the unemployment rate rises or falls. The “what” trumps the “why.” The wrong reason eventually becomes the right reason. The trend is what matters.

It’s hardly the first time that a decline in the household survey’s measure of employment, which is more volatile than the monthly tally of businesses, caused the unemployment rate to fall for the wrong reason. There have been 22 such instances in the past 30 years, according to a study by the International Strategy & Investment Group, Inc. Seven of those occurred during the current expansion.

Similarly, there have been 47 cases in the last three decades where the unemployment rate rose for the wrong reason, ISI found. Employment increased but was swamped by the number of labor-force entrants. It turns out that regardless of the internals, the unemployment rate contains more information than any of the other components in the jobs report, the ISI study found.

The weakness of other parts of the March jobs data only accentuated the unemployment-rate debate. There is reason to believe that the 88,000-increase in non-farm payrolls (derived from a separate survey of businesses) is understating the labor market’s recovery, according to Joe Carson, director of economic research at AllianceBernstein in New York. That’s because hard data from the U.S. Treasury on withholding taxes suggest more income than what’s implied by recent employment reports.

For the first six months of fiscal 2013, withheld income taxes are 9.2 percent higher than the same period last year, compared with an increase of 2 percent to 4 percent, smoothed for tax changes, in personal income, Carson said.

Withheld taxes tend to run above the Commerce Department’s measure of personal income, but this year’s gap, even adjusted for the Jan. 1 tax increase on top earners, is “unusually wide,” he said.

Although the Treasury does accept donations, the government doesn’t withhold taxes on income that people don’t earn – at least not yet. The additional income implied by withholding tax data has to come from one of three main sources: More employees, longer hours or higher wages. When the payroll data are reconciled with the Labor Department’s Quarterly Census of Employment and Wages, released with a six-month lag, we will have a better idea of the source of that income.

The demographic component of the labor-force participation rate may be more pronounced as the baby boomers retire, but the cyclical component should continue to drive the direction of the unemployment rate.

What if the softer economic data at the end of the first quarter are a sign that the business cycle is turning? The last place you would look for clues is the jobless rate, which holds the distinction of being a member of the Index of Lagging Indicators.

It’s hardly the only indicator to find itself challenged for being right for the wrong reason. My favorite indicator, the yield curve, is another example of the what mattering more than the why.

In the middle of 2006, the yield curve inverted: The federal funds rate was higher than the yield on the 10-year Treasury note. Unlike unemployment, the spread between the two rates is a leading economic indicator. In fact, it’s the leader of the leaders.

Most folks pooh-poohed the inverted yield curve as a harbinger of recession, which it is. Hey, just because it has been right every other time, why put your faith in something if you can concoct a good reason to discount it? Federal Reserve Chairman Ben Bernanke and his predecessor, Alan Greenspan, both touted the “savings glut” as a reason for low long-term rates even as the funds rate ratcheted higher.

In the end, the what (shape of the yield curve) trumped the why (savings glut). The lead time was unusually long, but the spread gave us an early warning sign of recession. I suspect the unemployment rate, although lagging, is giving us a correct signal now.