Explaining European Unemployment

Why do European countries have relatively high unemployment?  It's a question that interests me, and probably many readers.  The explanations offered by many German politicians strike me as, generally, unconvincing and ideological.  It seems the entire issue has become so fiercely polarized that there's no oxygen left for any attribution of reasons that isn't directly political (Lohnnebenkosten, Raubtierkapitalismus, job outsourcing, etc., to name some of the more popular keywords from political talk shows). 

Thus I was delighted when J. Bradford DeLong's blog, written by the Berkely economics professor, belched forth a string of 3 citations to careful studies of the problem.  The consensus of Blanchard and Wolfers' survey, the only one I had time to skim read is that high rates of lasting unemployment in Europe are due to a complex mix of factors, the most important being the interaction between shocks (oil-price spikes, sharp drops in productivity) and institutions.  Roughly put, what happens is that a shock causes a rise in unemployment which would ordinarily be eliminated by normal market mechanisms over the course of time.  However, the labor-market institutions in some European countries prevent the adjustment from taking place, and the labor market never recovers, or never fully recovers.  Therefore, unemployment continues to creep up over time.

Now to more details, from the paper itself:

The general idea is as follows. Take an adverse shock which leads to higher unemployment. The normal adjustment mechanism is then for unemployment to put pressure down on wages until unemployment has returned to normal.  To the extent that some labor market institutions reduce the effect of unemployment on wages, they will increase the persistence of unemployment in response to shocks. Research has identified a number of such channels. Here is a non-exhaustive list:

• [button] A rise in unemployment typically comes with higher unemployment duration (rather than higher flows in and out of unemployment). If some of the unemployed remain unemployed for a long time, they may either stop searching or lose skills...
• Why should institutions matter in this context? Because of their effect on the average duration of unemployment. A well documented fact about European labor markets is that, probably because of institutions such as more generous benefits and employment protection, a given unemployment rate is associated with much longer duration than in the United States...

• Higher unemployment may lead to a change in norms....   As long as unemployment is low, workers may be largely ignorant of the rules governing unemployment insurance, or there may be a stigma attached to being unemployed. After a period of high unemployment, ignorance is likely to disappear; attitudes vis—a—vis unemployment are likely to change. Thus, countries with a more generous welfare system may end up with higher unemployment, even when the shocks are gone.

Timely reading, and mostly comprehensible even to non-economists such as myself.  Until the equations began, that is...

Comments

[Tobias Schmidt]

Oh, and beware that this literature has become very technical (statistically sophisticated) in recent years and many of the results depend on the exact specification of the research model. Factors that are highly significant and important in one study do not even come close to statistical significance in others and adding or removing a few observations here and there often dramatically changes the results (see the linked article for examples of this). Whom you choose to believe in the end is often a matter of judging the appropriateness of various econometric trickery.

[Tobias Schmidt]

Blanchard and Wolfers is definitely a citation classic in the field, but it appears that Prof. DeLong's literature selection has been, well a bit selective. While a lot of studies agree on the importance of some labor market institutions, it is amazing, how shaky the empirical cross-country evidence on the detrimental effects of other institutions really is. There is a great debate going on about the complex interaction of different sets of institutions and with (both demand and supply side) shocks. If you're really interested in the topic, make sure you read some other articles and viewpoints as well. Nickell, Nunziata & Ochel - "Unemployment in the OECD since the 1960s - What do we know?" ,for example, is a response to the Blanchard & Wolfers paper from a very prominent labor economist, who claims that shock interactions add nothing to the understanding of the causes of unemployed. For a nice overview of the existing literature and a slightly heterodox view, check out http://www.newschool.edu/cepa/research/workingpapers/EmploymentProtection_InstitutionsAndUnemploymentCritical.pdf

[Detlef]

Well, an additional reason might be that we´re counting unemployment differently?

http://www.eurotrib.com/?op=displaystory;sid=2005/9/30/103936/067

[orange show]

Hmm, wait a minute. Can it be that the great bureaucratic socialistic 'democracies' of Europe are not the most efficient machines? Never before has that been posted on this website. America may be better in some way than Europe? Bizarre thinking on this blog. However, I'm glad to see some fairness in the reporting.