A “Now Hiring” sign appears in a storefront display window in Philadelphia.

One of the more concerning trends of the U.S. labor market in the post-Great Recession era is the seeming disconnect between the demand for labor in the economy and the actual hiring of workers. This divergence is often depicted by the difference in the growth rate of job openings (a sign of labor demand) and the number of hires in the Job Openings and Labor Turnover Survey data complied by the U.S. Bureau of Labor Statistics. One explanation of this trend is a “skills-gap” in the economy where workers just don’t have the requisite skills to get hired for the jobs that are open. Another diagnosis is that the 2007-2009 recession led employers to “up-skill” jobs—require higher skill levels for a given job—in the face of abundant labor. A new paper offers a slightly different interpretation of this upskilling argument.

The new paper, by economists Brad Hershbein of the Upjohn Institute and Lisa Kahn of Yale University, looks at how skill requirements for job openings changed from 2007 to 2010 and then to 2015. The economists used data from Burning Glass Technologies, which according to the firm contains the near universe of job openings posted online. Hershbein and Kahn use this data is see how skill requirements on job postings in different metropolitan statistical areas change depending on how hard their labor market was hit by the Great Recession.

What they find is that in areas that were hit hard by the recession experienced more upskilling than areas where the impact of the recession was more mild. In other words, the firms in areas where there were relatively more unemployed people were requesting higher skill requirements for posted jobs. This would seem to fit quite well with the common upskilling story that employers are taking advantage of a pool of reserve labor.

But their results don’t necessarily indicate this is the only factor in upskilling. Not only do firms in these areas have higher requirements on job openings, they also appear to hire workers with more skills. At the same time, these same firms appear to be investing more in an effort to retool their firms. To Hershbein and Kahn, these signs all point to firms in these areas adjusting to the recession by increasing investment in technology and skills as part of a longer-term trend in the economy. The two authors also point out this trend is continuing despite a tightening labor market in recent years as seen in the data up to the end of 2015.

Of course, it’s not clear that the areas hardest hit by the Great Recession have fully recovered, as other research shows. But while cyclical factors in the economy may have and might continue to play an important role in upskilling by employers, the role of longer-term trends is something that perhaps we shouldn’t ignore in this debate as well.