Jared Bernstein and I had a piece earlier this week discussing the problems that a government job guarantee would address along with some of the problems which make us reluctant to endorse one. I thought it would be useful to summarize the four areas in which we have serious concerns about the economic response:

1) The number of people currently employed who would opt for a guaranteed job.

Proponents of a guarantee argue that most private sector employers would improve their wage and benefit package to match the terms offered by a guaranteed job. Given the large portion of the workforce who stand to gain from a job with the terms being proposed ($15 an hour wage, plus genenefits), it isrous health care and other be possible that tens of millions of workers may see a guaranteed job as better than those on offer in the private sector.

2) The ability of the government to effectively manage a jump in the size of its workforce by at least 10 million and quite possibly two or three times this size. 

The issue here is whether the people doing jobs provided through the guarantee are actually doing useful work. This is not just a moral concern that people work for their pay. It will be pretty much impossible to maintain political support for a job guarantee if people employed under the program routinely come to work late, leave early, or don't show up at all, or alternatively sit around and do nothing when they are ostensibly employed. Since by definition many of these people will have little work experience, the task will be even harder.

3) The ability to find employment that does not displace existing public or private sector employment.

If the guaranteed jobs duplicate work already being done by workers in either the public or private sector, it will not, on net, create jobs. Rather, it will displace workers currently being employed. This is likely to be an especially big problem in the public sector, where state and local governments may see workers employed through the program as a way to replace higher cost public sector employees. For example, if workers in the program provide child care, they may displace workers already employed by the government doing child care.

It is inevitable that some amount of work in a jobs program be duplicative, but the question is whether this amount is closer to 5–10 percent, in which case it should be manageable, or more like 70–80 percent, in which case it would be a very serious problem. The larger the program, the more serious a problem this becomes, both because it matters less with a small program and it is more difficult to find useful work for 30 million people than 10 million people.

4) The impact of the program on inflation and cost structures in the private economy.

The point of the program is to guarantee all workers a decent standard of living. This will put a much higher floor on wages than is now the case. That will mean that it costs much more to go to restaurants or have your hair cut. If people want to hire someone to clean their house, mow their lawn, or look after their kids, they will have to pay considerably more than they do today. That is not a bad thing. But if the people who purchase a large amount of goods and services that depend on low-wage labor are able to push up their own pay to cover these higher costs, then we could see serious problems with inflation.

These are the four areas where I see serious issues created by a guaranteed jobs program. As Jared and I argued in our piece, various experimental programs now being proposed should provide us more insights on these issues.