Helping the unemployed move might not help them find a job

August 24, 2013

Do we need to encourage people to move in order to get the economy going and unemployment down? In a recent list of 14 conservative-friendly ways to get the unemployed back to work (covered by Wonkblog here), Michael R. Strain of AEI proposes we “offer UI-funded relocation subsidies for the long-term unemployed.” Elsewhere he writes that “an employment program should include a relocation subsidy to help the long-term unemployed move from high-unemployment areas to low-unemployment areas.”

This proposal could be helpful to people, although the specifics do matter. Back in 2010, the Hamilton Project proposed a mobility bank to make mobility loans, where “loans from the proposed mobility bank could not be discharged in bankruptcy court.” Getting the government into the business of making bankruptcy-exempt loans to people in desperate situations strikes me as problematic.

But besides the specifics, what does this policy say about the situation our country currently faces? One thing to note about a recession is that economic activity is decreased across all markets. It usually isn’t a matter of the normal churn of some places doing well and others poorly.

A lot of media coverage looks at the absolute level of unemployment (e.g. “10 States With Ridiculously Low Unemployment”), ignoring the fact that the natural rate of unemployment will vary from state to state based on a variety of influences. What we want to see is how unemployment has evolved since the Great Recession started.

So let’s do that. Here’s the ratio of unemployment by state, comparing unemployment now in July 2013 vs. unemployment in November 2007, right before the recession started (click for a bigger image):

The average is 1.56, which means that across all states unemployment is a little more than one-and-a-half times what it was in 2007. The median is 1.52, and the standard deviation is 0.26. You can play around with the data set here.

More interesting, there’s only one state, North Dakota, where unemployment is where it was pre-crisis.

While this is slowly going down, is it going down across the board? Maybe something has become broken in the recovery, with some states recovering quite well and other stuck in a high-unemployment situation. Given the regional nature of the housing bubble, this could be true. If that is the case, helping people move is a priority.

To see that, let’s see if unemployment is going down proportionately to where it increased. Here’s a graph of the rising change in unemployment from November 2007 through October 2009 (when national unemployment peaked at 10 percent), against the declining change in unemployment from then to now.

I was actually surprised by how clean this graph turned out. The linear regression has a t-stat of -6.46 and an r-squared of 0.45. Or in English, if you want to predict where unemployment has been falling you should look where it went up in the first place.

This is important because the proposal Strain has in mind would likely go to people who are at the back of the job queue -- long-term unemployed with low savings (which proxies for being involved in low-wage labor markets). These are populations that will have trouble finding work, and so it is likely that they’d just move to be at the end of the queue of another state.

To the broader point, there’s plenty of evidence indicating that mobility hasn’t fallen in the recent recession. People have wisely looked to see whether housing lock is an issue, and from a variety of evidence it doesn’t seem likely. People who have work opportunities locked in but face severe credit constraints would benefit, although I don’t see the evidence that consumer credit constraints as a whole are binding on the economy.

(As a technical aside, I’d like to comment on another of Strain’s proposals: “Permanently reduce the payroll tax and pay for the reduction by, say, increasing Medicare and Social Security eligibility ages.” Most economists calling for more action purposely want to avoid something called Ricardian Equivalence, or the idea that consumers will ignore stimulus because they know they’ll just have to cover it later. This proposal seems engineered to cause Ricardian Equivalence to happen, which might be fun for researchers but unlikely to help now.)

I applaud Strain for arguing that the situation in the labor market requires serious action by those on the right. But our eye should be on proposals that benefit the economy as a whole -- aggregating consumer demand through fiscal spending, monetary policy and the zero-lower bound, and getting to the bottom of the problems in the housing market -- rather than help the relative situation of specific individuals. In July 2013, a half-decade into this crisis, that is still the fundamental problem we face.

Show Comments

Get Wonkbook in your inbox

Sign up for our morning economic policy primer.

Most Read Business
Next Story
Neil Irwin · August 23, 2013