Are Americans working less because the government is paying them not to work? A large number of people seem to think this. Obviously the idea is popular on the right -- recall Mitt Romney’s infamous “47 percent” speech in 2012. But a surprising number of academic economists and even more left-leaning pundits have picked up the idea.
For example, Casey Mulligan, an economics professor at the University of Chicago and a blogger for the New York Times, has argued that a number of programs designed to soften the impact of the Great Recession ended up giving people an incentive to stay unemployed. This is the thesis of his 2012 book ``The Redistribution Recession,'' whose title is sure to appeal to the political right, but whose basic idea -- that means-tested benefits constitute implicit high taxes on the poor -- isn't that controversial, as this Paul Krugman blog post shows.
Another proponent of this thesis is Kurt Mitman, a macroeconomist at Stockholm University. Kurt’s models predict that ending extended unemployment-insurance benefits at the beginning of this year should increase both employment rates and labor force participation in the U.S. economy. I bet Kurt that this wouldn't happen, solely on the grounds that I think most macro models are barking up the wrong tree. So far, unemployment has continued to fall since the policy change, though labor force participation hasn't risen much -- in other words, it looks like we may split our bets.
A third version of this idea came from a surprising source -- Jordan Weissmann of the Atlantic (now of Slate), a generally center-left pundit. In March 2013, Weissmann wrote that Social Security Disability had become a “secret welfare program,” that in effect pays Americans not to work.
Economists are like most people -- they like simple stories. They also like effects that they understand. And the idea that taxes are an incentive not to work is a simple, uncontroversial idea. So it’s no surprise that many researchers, pundits and politicians would be attracted to the notion that our long, grinding economic stagnation is a policy error -- that good intentions ended up as bad policies.
But there are some serious problems with this thesis, as even a casual look at the data will reveal. The main problem is the behavior of wages.
If government programs are paying people not to work, then that should put upward pressure on real wages. This is intuitively obvious -- if the government is paying you to take a vacation, you will raise the price you demand to go back to work. Paying people not to work creates a supply shortage, and supply shortages increase prices.
But when we look at the data, that’s not what we see happening. After rising strongly in the early 2000s, real wages have basically been unchanged since the 2008 financial crisis. That presents a puzzle for the unemployment-as-vacation hypothesis put forward by Mulligan, Mitman and others. Now, of course it’s possible that other forces intervened, pushing down real wages just as government redistribution programs pushed them up. But it would be a pretty big coincidence that these other factors hit at the exact same time as the recession!
When you break up the wage data by percentile, it looks even worse for the vacation thesis. Christopher Ingraham of the Washington Post gives a snapshot. Wages have risen for the top earners, whose unemployment rates have stayed fairly low, while the bottom half of earners, for whom unemployment rates have spiked, have seen their real wages fall. Lower-wage earners are exactly the people that theories like Mulligan’s and Mitman’s predict will take a vacation when the government pays them not to work -- after all, if you make $200,000 a year, quitting your job to sit on your couch and collect welfare isn't very attractive.
Economics 101 says that when the price of something and the quantity produced both fall, demand, not supply, has fallen. In America, the price of labor and the quantity of labor have both fallen and stayed low since 2009. That is a hint that the government’s welfare programs are having only a minimal impact on the number of Americans with jobs. Whatever caused us to stagnate for five years and counting, it probably wasn’t welfare.
So what is it? Well, I don’t think anyone really knows why these long stagnations happen. I suspect that it has something to do with the aftermath of financial crises, but the link is not yet well understood.
To contact the author on this story:
Noah Smith at firstname.lastname@example.org
To contact the editor on this story:
James Greiff at email@example.com