The Limits And Possibilities of Public Works

Kevin Drum says I need to get real about the idea of direct government employment for the long-term unemployed:

Pundits like us don’t really have an obligation to produce detailed white papers as the price of entry for criticizing public policy. Still, I can’t help but feel that a case like this requires something. Sure, there’s probably a job somewhere around Salina for Paoletti. But what we’d really need are public jobs for all 5 million of the people who have been unemployed for over a year. I’m just not sure what that would look like, and I’d like to hear at least a remotely plausible scenario for creating quick, useful public jobs on that kind of scale in a sluggish economy before I insist that there must be a way to do it.

Ryan and I looked into this a bit and there seems to be something what Drum is saying. I figured the Economic Policy Institute would have an ambitious jobs program in mind and indeed they do but it’s a five-point plan only one of which involves the sort of public service employment I had in mind:

With a goal of putting one million people back to work, the program should be funded at $40 billion per year for three years, with funding allocated to local governments and states according to the Community Development Block Grant (CDBG) formula. The CDBG formula targets communities based on levels of short-term and long-term economic distress, as reflected in measures such as poverty, population decline, and age of housing stock. The formula could be improved with measures of unemployment and long-term unemployment, but to speed startup the first year could use the existing formula.

The U.S. Department of Labor should allocate funds and oversee the program at the federal level. Projects would be selected for funding by the highest local elected official based on the ability of the project to provide immediate employment to community residents, its benefit to the community, and the management capacity of the applicant. Local governments would design public sector programs or select projects proposed by nonprofit organizations and public-private partnerships that can quickly employ residents of the targeted communities while delivering a needed service.

A million people isn’t five million people, but it’s a lot more than zero.

But the biggest problem, in many ways, is something I was reminded of by Peter Orszag’s CAP talk yesterday where he bragged of ARRA “what has been most striking is that for an initiative this large, we have not seen any substantial incidences of fraud and abuse.” Fraud is bad. And abuse is bad. But mass unemployment is also bad. It seems to me that it would be better to have 1,000,000 currently unemployed people at work doing public service even if that meant 77 instances of fraud. But Barack Obama obviously doesn’t want to see headlines about “dozens of instances of fraud and abuse in DOL slush fund boondoggle.” And I think it’s in the nature of the sort of thing EPI is proposing that in the interests of speed and flexibility, you’d be creating a system that’s open to some measure of abuse and fraud. From a macroeconomic point of view, that’s not so bad — a little fraud is a small price to pay for a feasible means of taking a bite out of a giant problem. But I have no idea how you sell a skeptical public on that message.


The biggest takeaway here is about the importance of acting swiftly and at the necessary scale in a timely manner. Once people enter the “long-term unemployed” category, it gets much harder to design appropriate policy remedies.