Nonprofit, nonpartisan journalism. Supported by readers.


Minnesota loses 13,600 jobs; unemployment rate falls to 5.9% — huh?

Those seemingly conflicting statements are factually accurate — as far as the reported statistics go — but highlight problems some see with the way data for state employment numbers are gathered.

By one measure, Minnesota’s jobs market is fully recovered from the depths of the recession and is at an all-time high. But by another measure, the state showed its biggest one-month drop in jobs since June 2009.

Those conflicting statements are factually accurate — as far as the reported statistics go — highlighting a dilemma facing state labor-force number-crunchers and a growing disagreement with changes made by the U.S. Bureau of Labor Statistics (BLS) last January in how state employment numbers are gathered.

First, to the numbers.

The Minnesota unemployment rate, based on survey data from 1,700 households, fell by half a point in November to a seasonally adjusted 5.9 percent, the lowest since October 2008. That’s well below the U.S. rate of 8.6 percent, according to figures released today by the Minnesota Department of Employment and Economic Development (DEED).

Article continues after advertisement

The separate survey of 3,500 employers showed job losses of 13,700 across virtually every sector except education and health services, which gained 1,300 jobs, and mining and logging, which remained flat. October figures were revised to reflect an additional 1,200 jobs lost during that month.

The drop in the unemployment rate may signal good news for the economy, but because the three-month average unemployment rate fell below 6.5 percent, about 14,000 long-term unemployed Minnesotans will not be eligible for extended unemployment benefits currently being debated in Congress. Even if Congress passes the extension, those Minnesotans currently receiving extended benefits will see them ending in January.

The two separate surveys show the state has lost 22,900 jobs over the past three months, even while the unemployment rate was falling from 7.2 percent to 5.9 percent during that period.

“There’s been a serious degradation in the quality of estimates [coming from the BLS], said Steve Hine, DEED’s director of Labor Market Information. “Looking at Minnesota data, I very clearly see it. But Minnesota is only one state.”

The beef that Hine and fellow state labor officers have is that the BLS started conducting employer surveys on its own last January. Saying it could save $5 million by not contracting with each state labor office to gather the survey data, the BLS in Washington felt it could do the same job at lower cost. For Minnesota, that meant a savings of about $100,000 and the elimination of 1.5 positions, which Hine said his department absorbed through early retirements.

As a result of the change, Hine said the BLS often misses what’s happening in the states and has to come back later to revise numbers, usually upward.

He cites an example where construction employment was up, according to an employer survey, but the BLS felt it was too optimistic so it “smoothed” the data, applying a trailing 12-month average employment. Such an approach, coming out of a recession, risks understating what is actually happening on the ground, Hine said.

In another instance, the University of Minnesota did not respond to the employer survey in time to be included so its impact on education employment was understated for that month. Once the survey data came in, the previous month’s figures were revised. When the state was conducting the survey, they would have recognized the significance of the gap in U of M data and followed up before releasing numbers, he said.

Hine, who sits with other state labor officials  on a policy-making body for the national employment statistics system, said the group plans to study the quality of the year’s data by looking at the magnitude of monthly revisions.

Article continues after advertisement

They plan to complete the study next spring.

Even if the household survey data do more accurately reflect what’s going on in the economy, it still shows the state faces some significant challenges for what the BLS calls “workforce underutilization.”

The “official” unemployment rate (known as U3 in BLS-speak) is the number most often cited in news reports and comes from the household survey conducted by the Census bureau on behalf of BLS. That survey counts as “unemployed” individuals 16 and older who have actively sought work in the previous four weeks but are not working. There are other measures of what the BLS calls “labor underutilization” as well.

The largest category is the “underemployed,” meaning workers who currently work fewer hours than they would like. This number swelled during the recent recession and remains high, accounting for 5.5 percent of the total U.S. workforce and 5.2 percent in Minnesota for the 12- month period ending in September – the latest available statistics.

“Discouraged workers” are those who have given up looking for work “because they believe no jobs are available for them,” according to the BLS. In November, 1.1 million people in the U.S. fit that category, a decrease of 186,000 from a year earlier. Discouraged workers accounted for 0.6 percent of the workforce nationally and 0.3 percent in Minnesota.

Another category is the “marginally attached,” those who have not looked for work in the past four weeks for such reasons as school attendance or family responsibilities. This category accounted for 0.9 percent of the U.S. workforce and 0.7 percent in Minnesota.

Adding together the various measures produces a Minnesota labor underutilization of 13.1 percent, about 340,000 people. That ranks 40th among the states. The U.S. rate of labor underutilization stood at 16.2 percent for the same period, about 25 million people.