One of the perplexing economic questions these days is why wage growth has been so slow despite the longest economic expansion in U.S. history (now in its 11th year).

By conventional wisdom, tight labor markets should be raising wages much faster than is occurring. The logic is simple. With low unemployment, workers can quit their jobs and find something better. Companies have to boost wages to attract new workers or keep the ones they’ve got. This is Economics 101.

The trouble is that Econ 101 isn’t behaving as expected. Corrected for inflation, average hourly earnings in November were up a meager 1.1 percent from the previous November, according to the Bureau of Labor Statistics. Moreover, wage growth has decelerated recently. “What puzzles me most … is the slowdown in the last couple of months,” says Elise Gould of the Economic Policy Institute, a left-leaning think tank.

Wages are crucial to the economic outlook because they fuel consumer spending, which is about 70 percent of the economy (gross domestic product). If weak wages curb Americans’ shopping, the economy could slow further or even fall into a recession, which is usually defined as two consecutive quarters of GDP decline. The political implications for President Trump and his rivals are obvious.

What gives? There are many theories. Unfortunately, most have serious shortcomings.

Let’s start with the Phillips Curve. Named after its discoverer — economist A.W. Phillips — it holds that as unemployment declines, wage gains rise. You will hear it said that the Phillips Curve has straightened out; wages don’t rise with falling unemployment. But this is not an explanation. It’s simply a way of describing the basic problem: the disconnect of wages from unemployment.

It’s also worth noting that the death of the Phillips Curve has been greatly exaggerated. If you look at the underlying wage data, you discover that, as the U.S. economy recovered from the Great Recession of 2007-2009, wages did pick up as the Phillips Curve predicted. To simplify slightly: When unemployment was high, wage growth declined almost to 1 percent annually; as the recovery strengthened, wage growth improved to about 3 percent now.

But the figures I’ve just given you are “nominal” — that is, they don’t correct for inflation. When you make the correction, the Phillips Curve does flatten out. The annual gains are small.

So the question remains: Why are wage gains so weak when unemployment (3.5 percent in November) is so low?

My favorite theory is the hangover from the Great Recession. Th e economic collapse was so sudden and so frightening (monthly unemployment reached 10 percent) that it changed the way workers and employers thought and behaved. Workers cared less about the last pennies on their paychecks and more about keeping their jobs. They were willing to make sacrifices on ­wages.

Meanwhile, employers were also more cautious. They resisted big raises that would make them more vulnerable in the next recession.

Sounds plausible. Unfortunately, it’s not entirely consistent with some other facts. If people worry more about job security, you would expect there to be fewer job quits. But that didn’t happen. Quits and hires did drop when the economy was weak, according to another BLS survey, but recovered when the economy revived.

Another theory is that the growing costs of employer-paid health insurance have squeezed take-home pay, because employers regard medical benefits as a substitute for pay. Economist Mark Zandi of Moody’s Analytics says this is probably occurring but is too small to explain the current behavior of the Phillips Curve.

A more likely candidate would be people in the “prime labor force” — generally considered to be those between 25 and 54 — who don’t now have jobs. The employment-to-population ratio of this group is 80.3 percent, about the level it was in 2007 but lower than in 2000. These people represent a potential new source of labor supply that could be holding down wages, says EPI’s Gould.

The most truthful answer to nearly all of these questions is: We don’t know. We do know that wage growth matters, even if some of its mechanisms are still a mystery.

Read more from Robert Samuelson’s archive.