Ed Leamer Argues the Unemployment Problem is Mostly Structural (I Disagree, and Policy Can Help in Any Case)
Here's how I see it. As I said, I am convinced by the evidence showing there's a large cyclical component to the unemployment problem, but I could be wrong (and so could Leamer and others -- I think they are). So which is the bigger error, not helping struggling households who could be helped, or trying to extend a hand when it's not going to do much good? I'd rather make the mistake of trying to help when it isn't necessary instead of leaving people stranded when help is possible.
But suppose unemployment is, in fact, mostly a structural problem. If we could help to overcome the "slow uptake" problem after recessions by (1) providing public employment that bridges the gap until private sector jobs are available, (2) keeping people connected to the labor market and reducing the likelihood they'll drop out, go on disability, or choose some other socially costly alternative, (3) enhancing our long-run growth prospects, (4) saving ourselves money in the long-run, and (5) accomplish this with policies directed at "supply-side" problems that help with demand at the same time, shouldn't we do it?
Infrastructure spending has these features. We can delay basic maintenance for awhile much as a household or business can defer maintenance on a car or delivery vehicles, but there comes a point when the failure to do basic maintenance will cost us even more in the long-run (change your oil now, or change your engine later). We have delayed investment in infrastructure long-enough, and it's time to put people to work rebuilding for the future. These policies don't depend upon whether its cyclical or structural, we have the need -- the benefits exceed the costs (which are unusually low due to the recession) -- and there are millions of people who want to work, but cannot find employment. Why not put them to work doing something productive?