One of the few outright false statements in Bill Clinton's speech last night was his claim that unemployment remains high because workers just don't have the right skills. "There are already more than 3 million jobs open and unfilled in America," he said, "Mostly because the people who apply for them don’t yet have the required skills to do them." The overwhelming weight of the research says this isn't what's happening.

Granted, Clinton is hardly alone in thinking that a "skills mismatch" — that is, a gap between the skills employers want, and the one unemployed people have — explains a lot of the jobs crisis. Nobel laureate Michael Spence, for instance, has espoused this view on occasion. This interpretation is opposed by Keynesians who insist that high unemployment is an outgrowth of low demand, and that if the economy as a whole were firing on all cylinders, unemployment would fall in turn.

Three studies responding to this debate have come out just this year. One, from the New York Fed, found that a skills mismatch caused only 1.5 points of the 5 point increase in unemployment after the financial crisis. This graph, from the Fed researchers, shows two different measures of the increase in unemployment caused by a skills mismatch. Both show that the mismatch is, at the moment, less than one percentage point:

Federal Reserve Bank of New York

Edward Lazear, a Stanford economist who served as chair of the Council of Economic Advisors for George W. Bush from 2006 to 2009, and James Speltzer of the Census Bureau, found (pdf) the same thing. Lazear and Speltzer develop an index of "structural mismatch," and found that it increases dramatically during downturns. What's more, it's currently at its prerecession level, and yet unemployment is still higher than it was. If the mismatch were raising unemployment, then this wouldn't be the case:

Edward Lazear and James Speltzer

Lazear and Speltzer also found that unemployment shows the same trends across education groups, indicating that what's causing unemployment is low demand, not that people don't have the skills necessary to be employed. If the latter explanation were true, unemployment wouldn't have fallen by as much for college graduates:

Source: Edward Lazear and James Speltzer

The third study, by Jesse Rothstein of the University of California at Berkeley, found, like Lazear and Speltzer, that unemployment rose about the same amount across all education groups. He also tries to figure out whether wages have increased. If employers are having a hard time finding skilled workers, they should increase their wages. Did they? No — not even for new job postings, and not even if you correct for changes in the number of people employed by each industry due to the Recession:

Jesse Rothstein

Rothstein also looked at the reasons behind the increase in long-term unemployment. It turns out that the historically high number of long-term unemployed people can be almost totally explained by the historically weak state of the labor market. If the skills argument that Clinton invoked was correct, there should be even more long-term unemployed people than the state of the labor market suggests. That isn't so.

And these are just the studies that came out this year. But they reflect a consensus among most economists that what we're seeing is unemployment due to slow growth and inadequate demand for goods and services, not due to a lack of skills. Investing in education may be a good goal on its own merits, but it won't solve the unemployment problem.