By Noah Smith

Bloomberg Opinion

With the economy strong and unemployment low, why is wage growth so sluggish? Lots of economists and pundits are debating this vexing question. When the labor market gets tight, wages are supposed to rise faster. Instead, median wage growth is slower than it was back in 2016:

The most benign explanation is that there’s no mystery here — total compensation, which includes both wages and benefits, may be accelerating:

The first quarter of 2018 did see substantial compensation increases — an annualized rate of almost 4 percent. But one quarter doesn’t make a trend. In 2017, compensation growth was running at about 2.5 percent. That’s lower than in the early 2000s, even though more prime-age Americans are at work now than then.

Another benign explanation is that despite extremely low unemployment, the economy still isn’t really at full employment yet. The Great Recession lasted so long that many workers simply gave up looking for jobs — these people were classified not as unemployed, but as out of the labor force altogether. Some argue that when we take this shadow unemployment into account, the recovery — and the associated wage growth — are right on track. However, even in this picture, 2017 looks a bit weak.

Also, using total compensation instead of wages might not be a good idea, because benefits might be increasing due to factors unrelated to the business cycle, such the rapid rise in health-care costs. If this is the case, then the disparity between now and the early 2000s increases — wage growth in early 2018 has been equal to or lower than the trough of the early 2000s business cycle. There’s also a possibility that some of the people who dropped out of the labor force during the Great Recession weren’t really unemployed, but were just people who decided not to have formal jobs anymore by working under the table or in the black market. If that’s true, then using prime-age employment overstates the unemployment rate, meaning that wage growth is even slower than it ought to be at this point in the cycle.

So perhaps things aren’t OK. It’s possible that structural forces, unrelated to the business cycle, may be putting long-term downward pressure on wages.

One such factor might be what economists call monopsony, or concentrated market power. Evidence is piling up that employers in the U.S. are able to hold down wages because it’s hard for workers to find new jobs at higher pay in the area. If this power is greater now than in past years, it could be restraining wages, as Nobel economist Paul Krugman explains in an excellent blog post. Other structural factors — increased use of noncompete agreements, and the continued decline of unions — might be increasing employers’ power to avoid raising pay. The idea that employer power is holding down wages is becoming more popular.

But it’s also interesting to note that a similar phenomenon is happening in Japan. As observers of Japanese Prime Minister Shinzo Abe’s economic program have noted, the labor market there is even better than in the U.S. Yet wages are even more stagnant.

It’s perfectly possible that employer power is increasing in Japan as well. But Japan has very different policies and practices on antitrust, noncompetes and unions. So it’s curious to see the same macroeconomic mystery in that country as in the U.S. It might be a coincidence, or it might be the result of structural factors other than employer power.

International trade will seem like an obvious suspect to many — after the 2000s, the story of cheap Chinese labor holding down wages in the developed world has been burned into many people’s brains. However, Chinese labor costs have soared during the past few years and no other super-efficient, large-scale, low-wage manufacturing country has emerged. Also, China’s effect on rich-world wages was the greatest in the first decade of the new century, when the U.S. had faster wage growth than now.

The retirement of highly paid baby boomers might be more a promising suspect. This especially makes sense in Japan, where raises were traditionally based on seniority. When well-paid people leave the workforce, it tends to drag down the national average, even if entry-level wages are still going up.

Yet another factor might be slow productivity growth. Though wages don’t always track productivity, there is a relationship. Productivity often tends to slow down as an economy reaches full employment, since the least productive workers also tend to be the last ones to be hired. But productivity may also be slowing worldwide due to technological stagnation. If so, that’s likely to put a drag on wages.

For now, the reasons for slow wage growth will be remain a mystery — in fact, debate as to whether it even constitutes a mystery will continue to be lively. But in the meantime, the best approach is just to address the problems that might be holding down wages. Decreasing employers’ power over their workers and increasing economy-wide productivity growth are worthy goals, no matter where we are in the business cycle.

— Smith is a Bloomberg Opinion columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion