Watching the Republican and Democratic conventions, you might have thought there was nothing the two parties could agree on. But you’d be wrong. They have adopted the same theory -- at least in public -- as to why unemployment remains so high. The only problem? That theory is wrong.

Jonathan Newton/The Washington Post

You heard it in former President Bill Clinton’s comments. “Of course, we need a lot more new jobs, but there are already more than 3 million jobs open and unfilled in America, mostly because the people who apply for them don't yet have the required skills to do them.” In other words, the hold-up isn’t that employers don’t want to hire, but that they can’t find the workers they need.

Among economists, this is known as the “structural” theory of unemployment. A better name might be the “math-and-ATMs” theory. The idea is that unemployment is high because workers don’t have the right skills for today’s industries. Either they didn’t learn them in school -- the “math” side -- or they learned good skills but new technologies made them irrelevant -- “the ATMs” side. The way to fix it is to give workers the skills they need to compete in the modern economy.

This is in contrast to the “nobody-is-buying-anything” theory of unemployment. Here, the problem is that indebted consumers aren’t spending as they usually do, which means businesses aren’t hiring, which means consumers have less money to spend, which is creating a vicious cycle of economic stagnation. The solution, at least in the short term, is stimulus -- the government steps in and buys things, or hands out tax cuts so consumers and businesses can buy things, or somehow helps consumers get out from under their housing debt. Once that happens and the economy is humming along, government backs off and pays down its own debts.

Politicians tend to prefer the math-and-ATMs theory of unemployment because the solutions are the policy equivalent of motherhood and apple pie. You get to talk about better schools and worker retraining and fostering new industries. That’s a whole lot more popular than talking about further stimulus. But it’s also a whole lot less likely to work.

Stanford’s Ed Lazear chaired George W. Bush’s Council of Economic Advisers. So you might expect him to be the last economist who would provide aid-and-comfort to the pro-stimulus camp (though note that Bush himself pushed for and signed the Economic Stimulus Act of 2008). But in a recent study with James Spletzer, he demolished the foundations of the math-and-ATMs theory.

Lazear and Spletzer constructed what they call an “industrial mismatch” index that compared vacancies and job openings in different industries. If the economy had suddenly developed a strong preference for one set of skills over another, we should see job openings in some industries skyrocket even as unemployment falls. But that didn’t happen. “The industrial mismatch index in late 2011 is at the same level as before the 2007-2009 recession,” they conclude.

This is good news. It’s much harder to solve a math-and-ATMs problem than a nobody-is-buying-anything problem. Retraining workers takes a long time and is hard to do. Improving schools only helps the next generation. The kind of jobless problem that politicians seem to want is, weirdly, the kind they can’t solve anytime soon.

The unemployment problem they have, conversely, is the kind they could solve, or at least ameliorate, right now. The Obama administration even has a pretty good plan to do so: The American Jobs Act, which includes an expanded payroll tax cut, more infrastructure investment, better jobless insurance, a tax cut for firms that hire new workers, aid to state and local governments, and a program to rebuild schools and foreclosed properties. The law would cost around $450 billion, which the Obama administration proposes to pay for by closing tax breaks for richer Americans. Independent economists estimate it would create around two million jobs over the next two years.

And note that passing more stimulus doesn’t mean you can’t try and upgrade the skills of your labor force, too. We can and should retrain workers and improve schools and do all the other math-and-ATMs stuff while addressing the immediate crisis.

But the Obama administration has stopped mentioning the American Jobs Act. It didn’t appear in Obama’s convention speech, for instance. And the Romney campaign doesn’t have anything even like the American Jobs Act, even though Romney had a pretty substantial stimulus proposal in the 2008 campaign. So we’re in a situation where both campaigns know -- or at least once knew -- what the jobless problem is and what needs to be done about it, but neither wants to propose the policies that the economy needs, as opposed to the policies that voters want.

Which is all to say that our unemployment problem really isn’t of the math-and-ATMs or nobody-is-buying-stuff variety. It’s of the politicians-are-failing variety. And there’s no policy solution to that.