Why do people work?

That question is at the center of the conservative case against anti-poverty programs. Republicans like Rand Paul conclude that policies like disability insurance or the Earned Income Tax Credit take away a key motivation — putting food on the table — that propels people to look for work. Thus these policies must be reducing labor supply and economic growth.

Liberals often don't confront this point head-on, arguing instead that it's unjust for people to starve because they're out of work. It's an inevitability, given that conventional understandings of market capitalism require around one out of 20 people to be unemployed at all times.

This is a good point, but the conservative argument is worth confronting on the merits. While there is an inherent trade-off between work and economic output, the story is not so simple as conservatives make out. Austerity — which often requires cutting anti-poverty programs — also kills labor supply.

For an example of the conservative position, let's go to Daniel Mitchell, who wrote up some new findings from the National Bureau of Economic Research:

The mid-1990s welfare reform apparently helped labor supply by pushing recipients to get a job. Disability programs, by contrast, strongly discourage productive behavior, while wage subsidies such as the earned-income credit ostensibly encourage work but also can discourage workforce participation for secondary earners in a household. [The Federalist]

There is some surface plausibility to this argument. Social Security reduced poverty among the elderly by 71 percent, but in so doing probably also reduced the number of old people working. On some margin, there is a trade-off between work and poverty reduction, because a lot of jobs suck and people will quit them if they can.

However, it leaves a great deal out. Most critically, it doesn't consider the business cycle. At the bottom of the Great Recession, for instance, the ratio of job seekers to job openings was nearly seven to one. That means it was mechanically impossible for six out of seven unemployed people to get jobs then. In order for "pro-work" welfare reform to have a prayer of working, the jobs you're pushing people into actually have to exist.

In other words, when there is a recession, fiscal and monetary stimulus is the way to preserve labor supply, and austerity is the way to destroy it. But if you refuse to accept the logic of aggregate demand, as Mitchell did back in the very pit of the Great Recession, you're stuck arguing that soup kitchens caused the Great Depression.

The international context presents an even more obvious problem. The conservative account of anti-poverty programs straightforwardly implies that the larger the welfare state, the lower the labor force participation rate (that is, the fraction of people who are working or actively looking for a job). If people don't have to work due to generous government benefits, then they won't work.

This doesn't remotely comport with the evidence. In point of fact, by developed world standards, the U.S. welfare state is extremely stingy and our labor force participation rate is quite low. Take Sweden, for instance. It boasts the welfare benefits of Ayn Rand's deepest nightmares: universal health and dental insurance, 480 days of paid parental leave per child, a monthly child benefit of about $120 up through age 16, two weeks sick leave, government pension at age 65, and so on.

Overall, if we look just at market incomes, then Sweden has about the same market poverty rate as the U.S. — but its welfare benefits cut the actual poverty rate down to half that of the U.S. That's the scale of transfers we're talking about, and other Nordic nations do even better. Yet Sweden's labor force participation rate was 64.1 percent as of two years ago, more than a percentage point better than the U.S. rate, which has been hovering below 63 percent for the last couple years.

Again, at some point there has to be a trade-off between work and output. In decades previous, the U.S. beat European nations in labor force participation, because those nations chose relatively more free time as they became richer, instead of maniacally ratcheting up GDP for its own sake.

But correct macroeconomic policy also matters a great deal. If there is a catastrophic collapse in aggregate demand that is not fixed for years and years, that's also going to burn up labor supply — in a way that is both cruel and pointless.