January Labor Stats Show Lower Unemployment Along With Loss of Jobs

The math doesn’t seem to add up, and economists are the first to admit it. In January, the U.S. economy lost 20,000 jobs. The unemployment rate, meanwhile, dropped — from 10.0 percent down to 9.7 percent. Ordinarily, those two numbers move together, not in opposite directions.

A Drop Economists Themselves Call “Inexplicable”

The Economic Policy Institute, parsing the numbers the same day the Bureau of Labor Statistics released them, didn’t dress up the contradiction. “Given the slight decline in payroll jobs in January, there was an inexplicable decline in unemployment in January from 10% to 9.7%,” the group’s analysis stated bluntly, adding that “while the improvement in the unemployment rate is a welcome sign, at this point the drop can largely be attributed to the higher volatility of the much smaller household survey.” In plain terms: the unemployment rate and the jobs count come from two separate government surveys, one much smaller and noisier than the other, and this month, they simply told two different stories. Anyone hoping the falling rate signaled a real turning point was, in EPI’s assessment, reading more into a single month’s noise than the data actually supported.

The report carried its own separate gut-punch: revisions to prior months’ data showed the economy had actually lost 930,000 more jobs last March and 1.4 million more jobs last December than originally estimated. Total jobs lost since the recession began in December 2007 now stood at 8.4 million — a number stark enough that the U.S. labor market, EPI noted, started 2010 with fewer total payroll jobs than it had a full decade earlier, in January 2000, despite the labor force having grown by nearly 11 million workers in the intervening years.

Where the Losses — and the Rare Gains — Landed

The month’s job losses weren’t evenly spread. Construction shed 75,000 jobs, concentrated mostly in nonresidential building. Government payrolls dropped by 8,000 net, but that modest figure masked a sharper underlying split: state and local governments cut a combined 41,000 positions under mounting budget pressure, while the federal government actually added 33,000 workers — the vast majority of them temporary hires brought on for the 2010 Census, a one-time hiring bump that would reverse just as quickly once the count wrapped up later that year.

Compared to a year earlier, the improvement was real, even if it didn’t feel that way to anyone still searching for work. In January 2009, at the depths of the crisis, the economy had shed 779,000 jobs in a single month. Losing “only” 20,000 a year later represented genuine, if cold, progress.

A Jobless Recovery, Named as Such in Real Time

What’s notable, looking back at how this report was framed at the time, is that the phrase “jobless recovery” wasn’t a retrospective label applied years later — EPI used it explicitly in real time, writing that assuming the recession had technically ended the previous summer, “we are now well into a jobless recovery.” That’s a striking admission to make in February 2010: acknowledging the recession was over on paper while the labor market showed no comparable sign of turning around. The group’s own calculation of the “jobs gap” — the total number of jobs needed to both replace what was lost and absorb population growth since the recession began — put the figure at 11 million and rising, a number EPI argued demanded “bold action to create jobs” rather than confidence that the private sector would simply generate them on its own.

The Rate Would Keep Falling — For Reasons That Weren’t Entirely Reassuring

In hindsight, January’s confusing drop turned out to be the start of a pattern rather than a one-month fluke. The unemployment rate continued falling through the following year, hitting 9.0 percent by January 2011. But labor economists tracking the decline noted a persistent, uncomfortable asterisk: a meaningful share of that drop reflected discouraged workers leaving the labor force altogether — people who stopped counting as “unemployed” not because they found jobs, but because they gave up looking — rather than a labor market genuinely healing at the pace the headline number implied.

For related coverage of unemployment trends during this period, see our piece on the November 2010 jobless figures, or browse our full Labor & Economy archive.