Unemployment Data Conundrum

Jan 25, 2013

The state unemployment rate dropped for the second month in a row in December. But the numbers have been volatile with intermittent highs and lows that sometimes don’t seem to add up. WNPR’s Sujata Srinivasan takes a look at why short-term figures may not always show the long-term picture. 

Last year, the state unemployment rate – calculated by the U.S. Bureau of Labor Statistics or BLS, through a survey that includes around 1,600 households – climbed from 7.8 percent in May to as high as 9 percent in August, dipping slightly in September, back to 9 percent in October, falling again through November and ending at 8.6 percent in December. The December data shows people left the labor force–perhaps because they were discouraged and had given up looking for work. Connecticut Labor Department economist Andy Condon is working with the BLS to try and find answers to what appear to be contradictions in last year’s data. 

“It’s a pretty unusual pattern to see large decreases in the number of employed with a high percentage of those people leaving the labor force. That’s a very unusual pattern and the possible explanations just don’t make any sense to us.” 

Condon offers two possibilities. One, a major shift in the demographic composition of the workforce where aging employees simply choose to retire. If this is the case, the payroll survey – a different  indicator that totals the jobs created or lost during the month – does not show that these positions are being filled. The other reason could be that something in the household survey sample has changed. The state Department of Labor used to conduct a survey of around 6,000 businesses in the state and compute the results. Now, it’s produced centrally by the BLS.

“The Bureau of Labor Statistics largely took over the function of data collection through centralized collection points and they almost entirely took over the job of statistical estimation of job counts. That has resulted in an estimate that is purely statistically based as opposed to what we did here in the state and it takes little less advantage of local economic information that our analysts might have had and could provide.”

Condon says this calls for a slightly different way of interpreting the monthly jobs data.

“Our monthly job estimates were a little bit smoother when the states were producing that information and that’s because we were able to use information that we knew to confirm or fail to confirm certain data changes that may have looked unusual. That can no longer be done. So the estimates tend to be a little bit more volatile. Few more ups and downs. And so what we have recommended is that instead of overly focusing on one month’s results, to take a look at recent short-term trends and cycles.”

The preliminary monthly jobs estimates are revised and then benchmarked annually to state unemployment insurance and tax records in March. Economists say these revisions and other factors leading to volatility are why it’s important to look at more variables to get the long-term picture. 

Alissa DeJong, economist at the Connecticut Economic Resource Center – or CERC.

“The state numbers – the monthly and the quarterly unemployment and labor force estimates that come out are very volatile. It is difficult to look at them independently and to draw a lot of major conclusions.” 

So DeJong looks at other data in tandem. 

“What I like to do is look at those monthly indicators and complement them with what the national indicators are saying as well as what other indicators are saying. What does consumer confidence look like? What do retail sales look like? And set those monthly indicators in a more holistic context to get a picture of what’s going on.”

But even the long-term data points to a rough road. The biggest factor of concern is a contraction in the state’s labor force, which shrunk by around 45,000 last year. Meanwhile, job growth was flat, compared to 7,800 jobs added in 2011. The private sector has regained just 36 percent of jobs lost during the latest recession. 

Economist Pete Gioia at the Connecticut Business and Industry Association.

“The only bright lining in the report was that we did see that there were 800 jobs added in manufacturing and 500 in construction and mining. But overall this shows an economy that isn’t even in first gear right now. It’s really stuck in neutral.”

Many sectors are struggling to recover, but staffing firms say they are seeing growth in some areas. Recruitment firm Robert Half International has hired more employees in Hartford to meet the need of client-companies in the state, who are hiring in the IT, accounting, finance and technology space. 

Staffing manager Kelleigh Marquard.

“What we’re seeing right now in the Greater Hartford marketplace is definitely a dual job market. Last year was a better year than the year before. I think companies are more optimistic and our clients are telling us that they need our help when it comes to finding specialized skills.”

Condon says he is yet to see this trend in the overall jobs data. But the state Labor Department is expecting the jobs number to increase significantly when the benchmark revision process is completed. That figure, says Condon, could be as high as 10,000.

For WNPR, I’m Sujata Srinivasan