Worker glut--or shortage?

The unemployment rate for construction workers in December was 16 percent, not seasonally adjusted, nearly double the rate for all workers. Yet the number of unemployed construction workers has fallen to the point that some companies may soon have trouble attracting the particular workers they need. How can this be?             First, a quick data lesson: Every month, the Bureau of Labor Statistics (BLS) issues a variety of figures about employment and unemployment. The employment data are from a “payroll survey” of records filed each month by nearly all employers. The resulting report gives a generally accurate count of the number of workers on company payrolls. But it does not weed out double-counting of workers who may be on two payrolls, nor does it count self-employed (and “off-the-books”) workers. National and state employment data are presented in both unadjusted and seasonally adjusted form. Seasonal adjustment accounts for normal weather- and holiday-related variations in employment. These variations are particularly important in construction, and monthly data that are not seasonally adjusted should only be compared with the same month in prior years, not to other months. Unemployment data are derived from a monthly “household survey” of 60,000 individuals who are interviewed and classified by the industry they last worked in. The sample is weighted to reflect the entire working-age population. Nevertheless, the sample size—while very large by polling standards—is only big enough for BLS to seasonally adjust the total, not industry data. (The December rate for all workers was 8.3 percent, not seasonally adjusted, while the seasonally adjusted figure was 8.5 percent.) BLS converts the raw numbers into an estimate of how many individuals in the entire population are unemployed. The number of people who are listed as unemployed former construction workers was 1.3 million in December 2011. That was a decline of 700,000 from two years earlier. Unfortunately, on balance, hardly any of those workers have gone back to construction jobs. The payroll survey showed that construction companies actually employed 100,000 fewer workers in December 2011 than two years before, not seasonally adjusted. In other words, 600,000 former construction workers left the industry over the past two years. They may have landed jobs in other industries, gone back to school, retired or just dropped out of the work force. (Some may have gone into business for themselves, but the number of individuals in the household survey who said they were “self-employed, nonagricultural workers” also dropped—by nearly 500,000—over the past 24 months.) To be sure, the number of unemployed construction workers still exceeds industry hiring requirements by a large margin. Another BLS data set, the Job Openings and Labor Turnover Survey (JOLTS), shows that the industry typically hires between 200,000 and 300,000 workers per month in November through February, implying that there are far more unemployed workers than the industry will be hiring in the next few months. But in April through June 2011, hiring ranged from 440,000 to 500,000 per month. Even a modest upturn in construction activity in 2012 will elevate hiring further and narrow the gap between openings and available workers. Historically, U.S. workers—especially construction workers—have been very mobile. But recent Census data show interstate mobility dropped to an all-time low in 2010. That suggests contractors in states with low unemployment and increasing demand for construction workers may have more trouble attracting them than in the past. North Dakota is at the top of the list. As of November, it had the lowest unemployment rate (3.4 percent, seasonally adjusted) and had an 18 percent jump in construction employment from November 2010. Other states that have had large construction employment gains in the last year and below-average unemployment rates are Oklahoma (6 percent increase in construction jobs, 6.1 percent unemployment rate) and Wyoming (4 percent construction job gain, 5.8 percent unemployment rate). In short, the construction industry as a whole still has far too many unemployed workers. But demand for workers should pick up in 2012, and the pool of available, qualified workers will be much shallower in some states than the national industry unemployment data suggest.