On the other hand, if the question is whether there are enough jobs for everyone who might like to work (a question more about aggregate demand than about labor market institutions, but never mind) then it makes no sense to exclude early retirees, housewives, etc. Unless you think that women’s decisions to stay out of the labor force are completely unrelated their job prospects?

I don’t have to think that decisions on work are completely unrelated to prospects, merely that if we use E/P ratios as a measure of good labor market performance, we are assuming that being employed is an outcome preferable to the likely alternative. This is generally true for prime-age males in our society, but mostly not for school/uni aged people, those who would like to retire but can’t afford to, mothers of young children who don’t have adequate child care and so on.

Well, it’s distinguishing the largest five EU countries—Germany, Italy, France, Spain and the UK —with about two-thirds of the total EU population between them, from the other smaller ones. What else should it be doing? :-) And four of those five do indeed have significantly lower employment-population ratios than the US; the other, the UK, of course is the EU country with the most US-like institutions.

If you focus on prime-age males, the only group for whom an E/P ratio is a really good measure of labour market performance

This presumes we’ve agreed on what aspect of labor market performance we’re interested in. Central banks, who I think we can agree are the arbiters of unemployment most of the time, are interested in inflation (i.e. wage) pressure from tight labor markets. Since prisoners aren’t competing for jobs, they don’t factor in here. Hence, if you buy the argument—made on the center and the right as well as on the left—that inflexible labor markets mean a higher U is needed to keep the same level of wage stability, it makes no sense to add prisoners to U—if they could work, they would replace the current unemployed, not add to them.

On the other hand, if the question is whether there are enough jobs for everyone who might like to work (a question more about aggregate demand than about labor market institutions, but never mind) then it makes no sense to exclude early retirees, housewives, etc. Unless you think that women’s decisions to stay out of the labor force are completely unrelated their job prospects?

My riposte, again, would be that it’s not a question of erring on opposite sides of the same goal, but that the authorities in the US can sustain a more expansionary posture in the long run (and risk errors in that direction more easily) because the working class is weaker here. So going forward we can expect a continued willingness to experiment with massive fiscal stimulus, ultra-low interest rates and unorthodox monetary policy on this side of the Atlantic, not so much over there. This assumes, of course, that the objective is to maximize profits; whereas your neoclassical tradition assumes it’s to maximize some social welfare function.

You promised at one point you were going to write something about why Marxists don’t have a real alternative to the neoclassical paradigm. Well, this is one area where maybe you guys could learn a bit from us.

“Major” is doing an awful lot of work here. The US is middle-ranking on E/P ratios
http://lanekenworthy.net/

If you focus on prime-age males, the only group for whom an E/P ratio is a really good measure of labour market performance, the gap between the US and EU-15 is actually smaller than on unemployment rates, because the US has quite a high rate of disability.

We’ve been over this quite a few times in the comments thread. To spell out a point I haven’t made in detail so far, differences in the E/P ratio for the population as a whole are mostly driven by social institutions that affect female LF participation rates.

As you say, different measures for different purposes, but I think it’s fairly clear from the original post that the question I’m interested in answering at present is whether more “flexible” labor markets will produce better labor market outcomes for (employed, unemployed and potential) workers. Since incarceration is, for most prisoners in the US, a labor market outcome, I’m including it.

The gap between the US and the major EU countries in employment-population ratios is way bigger than the gap in employment rates. Without an explicit explanation of what unemployment means, it looks a little ad-hoc to say that U should include the one major population outside of the labor force that’s larger in the US, but not the various out-of-the-labor-force populations that are larger in Europe.

Before posting this, I carefully read the three posts on the subject—this one, the previous Bryan Caplan one, and refuted doctrines #8. Between them, they mention exactly one factor you thought had changed between the 1990s and today: increased geographic mobility in the EU. So my “not one word” isn’t correct; I apologize for getting carried away by my rhetoric. But unless you think lack of geographic mobility was the entire explanation for the really substantial gap in unemployment rates over the past 20 years (which seems implausible, and which you certainly don’t say), I think my broader criticism stands: You claim the gap will disappear without explaining why it existed in the first place.

The main part of your argument is that “EU unemployment rates should be higher in expansions and lower in contractions.” Well, maybe they should be, but they really haven’t been. What you might respond—I’m guessing—is that unnecessary fiscal and monetary restrictiveness have kept European growth below potential. But why would European governments keep making that mistake for 20 years? And given that they have, why expect they’ll stop now?

So in my mind, the prediction that the gap in U will disappear means either (a) Europe’s working class will soon be as demoralized as America’s, which I see no evidence of, or (b) that the American working class is becoming as aggressive as in Europe. Or© that new tools have been found to maintain the profit share even in the face of low U, which is certainly not true—the opposite, if anything, as centralized wage bargaining and co-determination are breaking down and there’s nothing like the fascist solution even on the horizon. Or (d) demand conditions will shift sufficiently in Europe’s favor to make elites tolerate a somewhat lower profit share in the context of faster growth. This is the most plausible option but it’s still not very plausible, since it basically comes down to exports and there’s every reason to think that the US current account position will get better relative to Europe’s, not worse, in the coming decade.

So I conclude that the same factors which led to higher U in the major European countries than in the US since the early 80s will continue to operate in the decade to come.

Now, you don’t have to accept to accept this analysis. But the striking thing about John Q.’s posts is that he offers no historical analysis at all. He says not one word about what he thinks has changed in recent years that would explain the change in relative unemployment rates. His argument comes down to, he likes Europe’s labor market institutions better than the US’s (so do I!); therefore, Europe ought to have good labor market outcomes. The actual experience to the contrary isn’t even wished away, it’s simply ignored. That’s what I think is funny.

([1]Pollin: “the principal factor generating the concurrent fall in unemployment and inflation in the 1990s was the fact that, as of the 1990s if not sooner, working people had experienced eroding bargaining power and a loss of self-confidence as to their ability to raise wages, even when unemployment is low. For example, Greenspan referred to a ‘traumatized worker’ syndrome in public testimony before Congress. Janet Yellen, who was also a member of the Board of Governors of the Fed in the 1990s, made similar observations in closed meetings of the Fed Open Market Committee. If workers do indeed feel an erosion of their bargaining power, this means that, even at low unemployment, they are less willing to attempt to push for wage increases. Weaker demand for wage increases then means lower costs for businesses at low unemployment. With lower cost increases, businesses will be less inclined to raise the prices they charge to consumers, and this, finally, means less inflation at low unemployment. More generally then, the upward cost pressure that businesses will face will be diminished when unemployment is lower when workers feel ‘traumatized.’ This is the crux of what Greenspan himself has said, and I agree with him.”)

Obviously. I just think it’s funny that someone can make a confident prediction about the future trajectory without giving any though to their past trajectories. As far as one can tell from his posts here, John Q. would have made this exact same bet in 1999 or 1989 or 1979.

And Sebastian, that’s just a longer version of your first post—you’re making these mind-numbingly basic extropolations and acting like they somehow whip the carpet out from under the whole argument. If I’d said “I know that every person in prison would have been unemployed”, just like if I’d earlier said “all percentages are comparable,” you’d have a point. But I didn’t.

Furthermore they talk about those discouraged from the labor market as not counting, but they aren’t counted in Europe either, so that isn’t the difference.

Which btw shouldn’t be considered a defense of the US in its stupid incarceration policies re the drug war.

It is possible that incarceration is some fairly noticeable component of the difference in employment rates. But I seriously doubt it is ALL of it, which is what seems to be implicated here.

Furthermore if you pursue Kaplan’s website you will find that he does an intra-European comparison of job flexibility and the labor market, and finds that in most cases it appears to be correlated to unemployment. He doesn’t correlate that to prison rates, but since the US is such an outlier in that, I would suspect that in an intra-European comparison the prison rate can’t be the deciding factor.