Here's an odd economic mystery. Americans are moving around the country a lot less frequently than they used to. For example: In the early 1990s, 3 percent of Americans moved across state lines each year. Today, the rate is half that.

A rare sight. (Getty)

Economists have offered all sorts of theories for the drop in mobility. Perhaps it’s because America’s aging, and the elderly don’t move as much. Or maybe it’s because there are more two-earner households around — it’s harder to swap cities when both adults need to sort out their job situations. Maybe it’s the recession and the rise in underwater mortgages. Or perhaps high moving costs are to blame.

But none of these theories appears to fit the data. That's the argument of an intriguing new paper (pdf) from Raven Molloy, Christopher L. Smith and Abigail Wozniak of the Federal Reserve Board. The authors find, for instance, that demographics can't explain the drop, since Americans of all ages and groups are moving less.

Instead, they suggest that Americans are moving less because Americans are switching jobs less frequently than they used to — and this was true even before the Great Recession hit. Occupational mobility is down. The rate at which people quit their jobs keeps falling. Moves between firms are down. And these trends match up well with the fall in mobility, at both the individual and state level.

Of course, that just raises a new question: Why aren't Americans switching jobs as often as they used to?

The authors consider a number of potential hypotheses, from the increase in two-earner households to the rise in health-care costs. But none of these seems satisfactory. For instance, they note, the number of two-earner households hasn't increased significantly since the 1980s.

Instead, their working hypothesis is that "internal labor markets" within companies have changed in such a way that workers are less likely to leave. Perhaps that's because workers are better able to match up with an employer from the outset — thanks to Internet searching, say — so they're more likely to stay with their company longer.

There's also some evidence that more jobs require company-specific skills, making it hard for workers to transfer elsewhere. And there's evidence that workers have been getting steadily more anxious about quitting their jobs  — again, a trend that was happening even before the recession hit.

Whatever the explanation, the authors find that the wage gains associated with switching jobs have fallen over time, suggesting that workers now have less to gain from moving to a different company or job. If true, that's a huge deal, they note, since for a long time "economists have surmised that changing employers is a main channel of individual-level wage growth." Increasingly, it appears, that's no longer true.

Further reading:

— Neil Irwin recently argued that the U.S. labor market won’t be healthy until people feel like they can quit their jobs. But it's possible that there are reasons besides the weak economy for why the number of people quitting their jobs has been declining.

— Here's an older post on why Americans aren't moving. The paper discussed here argues that people have less need to move because jobs are no longer as location-specific as they used to be. But the Federal Reserve Board paper discussed above argues that this theory can only explain part of the data.