My friend Matt Bruenig does consistently great stuff, but I have to object to his recent critique of the idea of a Job Guarantee (often called the government as an Employer of Last Resort, or ‘ELR’). I want to walk through his arguments, point by point.
This explainer by the illustrious Randy Wray provides background on the entire idea, including the macroeconomics of ELR, while Matt and I mostly stick to the mechanics.
He starts well, noting it isn’t the 1930s and you can’t put people to work by just handing them a shovel. Jobs require capital, quite true. I’d go further, jobs require capitalized organizations with plans. Call them public enterprises. Decide what sort of work ought to be done and set up assorted enterprises to tackle different jobs. They would provide public services and construct and maintain public facilities. They would supplement the efforts of state and local governments that in many cases fail to do what ought to be done, sometimes for lack of capability. They would utilize workers with a wide variety of skills. And they would be organized to handle more fluid ebbs and flows of employees.
Matt fears the supply of labor for such purposes as too variable to staff permanent organizations in continual operation. It seems to me that jobs could be provided in six month stretches, with the opportunity to re-up. Some employees would be permanent, and paid commensurately.
Work could focus on a sequence of one-off projects. With higher private employment and fewer ELR applicants, projects could proceed more slowly, or be postponed. Just because a project isn’t essential doesn’t mean it would not be worthwhile. Some applicants will have more skills than others, and some work will require more or less in the way of skill. There is already a lot of ‘churning’ in the labor market (people moving between jobs greatly outnumber those in and out of work). Business firms cope with it.
Permanent ELR-staffed operations would reduce unemployment and oblige the private sector to be more flexible. They could be focused on the most benighted localities that lack the resources to provide ample services and facilities. For instance, if the state of Texas won’t build sewers for people in the Rio Grande Valley, the Feds could set up there, train and employ local labor, and situate such communities to exercise more political power in the future. This bears some resemblance to the original War on Poverty, before the community action angle got smothered in 1967, when state and local politicians gained substantial control of the program.
What kind of work is worth Federal support? Day care centers and public schools could use teachers’ aides (properly vetted). Small-scale infrastructure work. Reclaiming unused urban land. Green retrofits of public facilities, and maybe privately owned structures and homes. There’s nothing to prevent a public enterprise from owning any kind of equipment it needs.
Training someone who subsequently leaves the ELR service doesn’t strike me as a problem. We train people now who never work for public institutions. There is a public value to job training. Nor is it a problem if people leave the service because they find something better. That’s the whole point: to force employers to compete for labor with institutions with better standards of employment. If nobody needed the last resort of public employment, that would be fine.
In general I think Matt’s notion of how an ELR system could work is too narrow. He ends by suggesting a modification in line with my discussion above, one that I don’t think goes against the grain of an ELR system.