One of the great recent puzzles in economics has been the absence of inflation, particularly in the years after the Great Recession. Some recent research explains why inflation has been so muted for so long. This work also suggests that inflation will be of diminishing concern in the future. Those conclusions will be hard for many to swallow, so let me explain.

Americans under age 50 will have no meaningful memory of inflation, while Americans over 60 will have sharp and unpleasant memories of its effects. Fixed mortgage rates provide a good example. A 1981 home buyer faced a 30-year-fixed mortgage rate of 18.5 percent, but a decade later, in 1991, it was half that rate. That same mortgage is today at 2.8 percent, perhaps a record low. Inflation affects nearly everything from the price of goods and services to credit card rates and wages. It is not something that affects just a few products or commodities. In that way, inflation reduces the standard of living of families. Inflation also slows economic growth by imposing a de facto tax on savings while shifting wealth from those who save to those who borrow.

The reemergence of inflation risk is not an idle worry. Inflation helped push Hitler to political prominence, and ushered in dictatorships from Argentina to Zimbabwe. Inflation is not just a minor financial phenomenon. The underlying cause of inflation is an excess supply of money in an economy. The most acute examples involve governments or central banks increasing the money supply to bolster a lagging economy.

The huge stimulus payments and easy money policies in the wake of the Great Recession caused inflation fears. More than a decade ago, I penned columns warning about the growing risk of inflation, a concern shared by most economists. Thankfully, our worries didn’t materialize, but that failure of our predictive models unleashed a torrent of new research on the topic. In my judgement, that work has largely solved a decade-long puzzle.

Inflation is caused by excess money, but for it to actually occur, markets for goods and services must experience price increases. Labor markets also respond with wage increases. However, today’s labor markets are in the midst of historic change. Automation has replaced many jobs, particularly those that have routine tasks, such as assembly line work. These differed from less routine types of jobs in their response to recessions. Within that modest fact lies the key to the puzzle.

Workers who perform more abstract tasks tend to change jobs more readily, which makes those labor markets more fluid. Workers in more routine tasks change jobs less frequently, making their labor markets less fluid, or more rigid. One study of European workers found that the labor markets for non-routine workers saw three times as much fluidity or job changes as those for routine workers. For example, consider which worker is more likely to change employers this year, a web designer or an assembly plant worker.

There are many reasons why this labor market fluidity may exist. Perhaps the workers who perform abstract tasks are more comfortable changing jobs. Maybe the business that employs lots of non-routine workers has larger capital investments, and so is more reluctant to risk suspend operations due to staffing shortages. Whatever the reason, this is simply an empirical fact.

Over the past three decades, labor markets have created far more non-routine or abstract jobs while eliminating routine jobs. For example, 81 percent of all the new jobs created in the U.S. since the end of the Great Recession went to college graduates, whose occupations comprise more abstract tasks. This makes labor markets far more fluid, but there are regional variations in these changes. The best laboratory for this is Europe, where there are differences in both the mix of occupations and inflation. Two economists at the French Central bank studied this in depth.

By controlling for industry differences in countries, and changes to policy by individual central banks, these authors were able to isolate the effects of labor market fluidity on inflation. They found that the more polarized the workforce, the less inflation was observed. Because Europe had both a sovereign debt crisis and a Great Recession over the past 15 years, there were also two different sets of impacts to help calibrate their findings.

There are other hints that underlying conditions in labor markets may be holding down inflation.

Two economists in the U.S. (Amanda Weinstein at Akron, and Carlianne Patrick at Georgia State) found the unemployment rate was far lower in cities with a higher mix of abstract occupations. In cities with a higher share of workers performing routine tasks, the unemployment rate was higher, and more volatile during recessions. This implies, but does not conclude higher labor market fluidity in the abstract occupations.

What is not yet clear is why this labor market fluidity helps keep inflation in check. It may be that businesses face fewer shocks to profitably due to more fluid labor markets, and thus are less inclined to raise prices. Or, it may be that the more fluid labor markets occur mostly in firms that already enjoy some monopoly power. Larger firms tend to change prices less frequently, even when others around them do so. At least, that is what my doctoral dissertation reported. Thus, if we are indeed living in a more monopolized world, there will be fewer inflationary shocks.

There are two things about this most readers will find interesting. The first is that this provides a pretty clear insight on how research works. Economic models of inflation worked well for decades, then their predictions failed. This caused theorists and empiricists to head back to their blackboards and computers. The second interesting issue is that if this holds true, we should expect less inflationary pressure in the future.

There is much discussion about inflation in today’s political and economic news. We’ve just been through an historic economic crisis accompanied by an historic fiscal and monetary stimulus. Several products, like lumber or used automobiles have seen big price swings. But, if we look at consumer inflation over the last two decades, this is nothing but a modest adjustment period. Right now, that is what the evidence suggests we should expect.