1. Labor market polarization. This is a very popular idea among the Progressive Left and rightly so; it seems to be true and increasing in importance. Yet it gets dropped like a hot potato when discussions of stimulus come up. A simple interpretation of the data, consistent with labor market polarization, is that we have a larger sum of money chasing the same set of well-qualified, easily-employable workers. Polarization also means not so much substitutability and there is plenty of evidence in the Jones-Rothschild paper of employers finding labor markets — for what they want — somewhat tight.
2. The Jones-Rothschild paper has an estimate that only 42 percent of the job offers went to the unemployed. A lot of the money also was spent on capital, land, raw materials, and other factors of production. I’ve never seen good estimates here, but labor’s share is about seventy percent of gdp, actually a bit shy of that. Let’s say seventy percent of the stimulus gets spent on labor at all, and only forty-two percent of that gets spent on unemployed labor. (It’s actually worse than that because it is 42 percent of the job offers and may well be less than 42 percent of the revenue, most likely so if you think of the unemployed as bringing lower wage offers.) That’s less than thirty percent of the initial expenditure being spent on unemployed labor and that is before any other problems with the expenditures kick in. It’s hard for me to see that as a triumph of the program (NB: we are only talking about one part of ARRA here); would direct government employment have overhead costs that high? How about monetary stimulus? What’s the new calculation for cost per job saved per year?
3. Cutting nominal wages of workers hurts their morale and firing people hurts the morale of everyone left behind. Employers weigh these morale costs carefully when making personnel decisions. To get to the matter in question, when times are tight employers are often quite relieved when workers leave the firm voluntarily. It eases their cash flow, prevents a firing, and everyone is happy, sort of. Bad times are precisely when replacements of these workers do not happen. (Another version of that argument: If Keynesians are right about labor hoarding, job shifters don’t get replaced very often.) So the claim that an ARRA hire of an already-employed worker led to a replacement for that worker at the original firm is not so strong. This happened during down times and very often replacement is postponed, perhaps indefinitely. The Keynesian view, after all, stresses how AD problems hit virtually the entire economy.
4. Very often when the replacement does happen, the replacement is drawn from the pool of workers who are doing well. Some of those workers will be unemployed. But they are the unemployed who least need the help. Their average search time goes down, and that is somewhat of a social gain, but it is hardly the goal of a fiscal stimulus program. We’ve failed very badly at reemploying the hard-core unemployed and that is borne out by other numbers. So of the forty-two percent of offers going to the unemployed, how many were going to the “don’t so much need the help but were searching” unemployed? That will make the calculation look uglier yet.
Labor market polarization, labor market polarization, labor polarization. Market-oriented economists don’t like to stress this theme, but it’s true and it’s a big reason why the stimulus didn’t work better. It’s not an idea we should suddenly leave behind.