(AP Photo/Jose Luis Magana)

Are wages stagnant because the economy was bad yesterday or because it's bad today? Yes.

The problem is that even though unemployment is a normal-ish 5.7 percent, wage growth is still a below-normal 2.2 percent. It could be that companies aren't increasing wages to make up for the fact that they wanted to cut them during the recession, but couldn't. That's what the Federal Reserve calls "pent-up wage deflation." Or it could be that companies aren't increasing wages due to the fact that there's more slack than the unemployment rate is letting on. And that's why the Federal Reserve is, for example, paying close attention to part-time workers who want full-time jobs but can't find them.

What does it matter which it is? Well, if the problem is in the past, then the Fed can ignore low wage growth and start raising rates. But if the problem is still with us, then the Fed can't ignore it and might have to put off its plans to start normalizing policy in June.

Now you don't have to be an economist to know that people don't like taking pay cuts. It's pretty rational. If you have fixed costs, like a mortgage or student loans, then there's only so much less that you can afford to make. But it's not just that workers resist lower pay. It's that firms don't like to force it, either. That's because pay cuts make workers unhappy, and unhappy workers are less productive ones. So, as economist Truman Bewley found out when he actually asked people, companies think pay cuts are bad for business. (And that's just as true in Indian villages as it is in U.S. mega-cities).

It's no surprise, then, that a lot of workers got exactly zero wage growth during the Great Recession and not-so-great recovery. Companies "should" have cut wages, but didn't feel like they could. So instead, firms kept wages right where they were. You can see some of this in the chart below from economists Martin Beraja, Erik Hurst, and Juan Ospina. They compared how much unemployment changed with how much wages did in each state between 2007 and 2010. The result, as you might expect, is that the more joblessness went up, the less wages did—staying flat, or even falling a little, in the hardest-hit states.

Source: Beraja, Hurst, and Ospina
Source: Beraja, Hurst, and Ospina

Now wait a minute. Take a look at that chart again. Even during the bust, wages were still rising in all but what had been the bubbliest of states. So it's hard to see how today's low wage growth could be making up for yesterday's zero wage growth that "should" have been negative when there wasn't that much zero wage growth to begin with. It's even harder to see how that could be the case when you consider Goldman Sachs found that the cities and industries that increased wages the least right after the crisis tended to increase them the most a few years later. Companies, in other words, really do try to avoid wage cuts, but despite that, there aren't many "pent-up wage cuts" left over now.

It's just the unemployment, stupid. Or maybe the underemployment. Between people who can't find the full-time jobs they want, people who haven't been able to find any jobs after looking for at least six months, and people who think things are so hopeless that they've given up looking for now, there are a lot more people than normal stuck on the margins of the labor force. And these "shadow unemployed," according to the Fed, exert just as much downward pressure on wages as the regular unemployed. Put it all together, and wages haven't recovered because the economy hasn't fully recovered.

The only mystery is the one we make.