The government’s April jobs report produced some happy headlines and a big stock market rally. The dismal March jobs tally was revised upwards from under 100,000 new jobs to a still feeble 138,000. In April, the economy created 165,000 jobs. The nominal unemployment rate dropped all the way from 7.9 percent to 7.5 percent.

But look a little deeper and you’ll appreciate just how crummy these numbers are.

The typical new job pays far less than the jobs that have been lost. We are still down a net 2.8 million jobs from the number of people who were employed in 2007 before the recession started. All told, there are 22 million Americans either unemployed or under-employed—looking for full-time work and not finding it.

One telling indicator is the very low percentage of people who are in the labor force. Before the recession, in 2007, the employment-to-population ratio was above 63 percent, down slightly from its peak of over 64 percent in 2000. Since the great collapse, the ratio has been stuck at just over 58 percent. And despite the improvement in the official unemployment rate, the percentage of people in the labor force hasn’t budged during the more than three years of “recovery.”

It would be comforting to think that the labor force number is low because baby boomers are retiring. But the one age group whose labor force participation is actually increasing is people over 65—they can’t afford to retire.

Economists across the spectrum expect that the mediocre rate of job creation will continue or worsen as the sequester squeezes more tightly in coming months. The nonpartisan Congressional Budget Office projects that the combination of the sequester and the increase in payroll taxes that was part of the January budget deal (plus the token tax hike on the top one percent) will cut economic growth this year from what could have been 3 percent to just 1.5 percent. You don’t produce very many good jobs with that kind of feeble growth.

One of the great myths of our era is that the problem is a “skills mismatch.” Superficially, the fact that college grads do so much better than high school dropouts lends credence to this view.

Blaming the lousy economy on worker skills conveniently accomplishes three goals for political elites. It blames workers themselves for the poor condition of the economy as exacerbated by bad policy. It scapegoats public schools and builds support for the voucher and charter movement. And it diverts attention from the perverse macroeconomic policies of budget tightening—now a goal shared by both the President and his Republican opposition.

But when you think about it, not much changed about the skills that the economy demanded between 2008 and 2013. The workers who had jobs and skills in 2008 did not abruptly lose them. What changed, of course, was the financial collapse.

Here is history’s great counter example. In 1940, conservative critics and orthodox economists were making just the same arguments: new technology was displacing human workers. Too few workers had the necessary skills for the modern economy.

That was why, supposedly, unemployment remained stuck about 12 percent despite all of the employment programs of World War II.

But then the Japanese attacked Pearl Harbor. In six months, the government entered $100 billion worth of war production orders, mostly financed by debt. And guess what? Unemployment disappeared.

Imagine if government today spend a comparable scale of money—trillions in today’s dollars—on the public infrastructure and green conversion that our economy needs. People would gain skills on the job.

Consider, too that most of the World War II economic stimulus came from building products that added nothing to the domestic economy except for the workers’ wages. Imagine if all that money went not to make war but to rebuild the economy.

So let’s not cheer the feeble jobs numbers, and let’s recall that belt-tightening will only make the economy worse. Maybe—maybe—the president contemplating his legacy will reverse course and save us from a decade of stagnation.