Father and Son Moving in to New House by Chromatica, veer.com

Horace Greeley, were he alive today, would be disappointed to see so many Americans not taking his famous advice—“Go West!”—to heart.

U.S. workers are not only moving west at a declining rate, but they are becoming less likely to move east, north, and south as well. The U.S. Census Bureau recently released data on geographic mobility in the United States that shows the rate at which workers move to different states and counties is still historically low. This now-three-decade-long trend has some important implications for the U.S. economy, even if we aren’t exactly sure what’s causing it.

One of the advantages of the U.S. labor market has long been that its high levels of labor mobility let workers move to a new area if job prospects dry up. If there are fewer jobs in Indiana, you can move to California. In fact, a 1992 paper by economists Olivier Blanchard, now of the Peterson Institute for International Economics, and Lawrence Katz of Harvard University shows that mobility was the key way people responded to losing a job.

But that’s less the case now, according to new research by economists Andrew Foote of the U.S. Census Bureau, and Michel Grosz and Ann Huff Stevens of the University of California, Davis. Looking at the impact of mass layoffs, the researchers find that workers are now twice as likely to exit the labor force as they were before the Great Recession of 2007–2009. Moving is still the largest response to being laid off, but it has become increasingly less important. The three economists’ finding is very similar to the findings of a working paper by  several International Monetary Fund researchers.

The declining rate of workers moving–and its changing importance in the labor market–raises the question of what’s actually causing the decline. One obvious culprit is the changing demographics of the United States: Older workers are less likely to move. But as Adam Ozimek of Moody’s Analytics points out, there are analyses of this declining migration rate that have accounted for this change and still found a significant decline in labor mobility. Ozimek points to research suggesting a role for occupational licensing in pushing down the rate in geographical moves. Although he notes that the decline isn’t well enough understood at this point to make definitive conclusions.

The research we do have, however, suggests other potential explanations—although some make more sense than others. For example, research by economists Raven Molloy and Christopher Smith of the Federal Reserve Board of Governors, and Abigail Wozniak of the University of Notre Dame shows that the decline in mobility is intimately linked with changes in the labor market. Workers are less likely to move somewhere because the gains to getting a new job in any location have declined as well. In this case, geographic mobility tracks job-to-job mobility. Of course, this just shifts the question to why job-to-job mobility has declined. It could be because workers are better sorted to their jobs so they don’t need to move, or because of a declining demand for workers—a collapsing of the job ladder.

The answer of this now-slightly-different question remained unsolved. But at least in the search for the answers, we have some guidance on the right set of questions to ask.