Is it a skills gap or a bias against the long-term unemployed?

I'd like to call your attention to what is essentially more wood for the fire of the seemingly endless 'Skills gap' debate - a summary of some recent research by two Northeastern University academics, William Dickens and Rand Ghayad, titled 'It's not a skills mismatch: Disaggregate evidence on the US-unemployment-vacancy relationship' posted this past weekend on the Vox site.  

You should take a few minutes to read the piece from Dickens and Ghayad - it provides an interesting and somewhat alternative method of examining the alleged skills gap problem - the scenario that so many corporations, government agencies, and educational institutions seem to take as truth. Namely that one significant reason (perhaps the most important reason) for persistently high unemployment, even after the official recession was over, is that a fundamental 'skills gap' is preventing organizations from filling many of the millions of current vacancies that exist in the USA.US Beveridge Curve 2004 - 2010

Dickens and Ghayad analyze the behavior of the so-called 'Beveridge Curve', the relationship between unemployment and job vacancies that is an indicator of the efficiency of the labor market. As you'd expect, the Beveridge Curve suggests that as the vacancy rate increases, the corresponding unemployment rate decreases (in an efficient labor market), as more people find work in an environment of increased opportunity.

But in the months following the end of the recession, the data points that connect the level of job vacancies and the reported unemployment rate continue to lie outside of the range of expectations suggested by the Beveridge Curve, implying some kind of inefficiency in the labor market. The most common explanation for this discrepancy is the 'skills gap', simply that the nature of work is changing so rapidly, that the skills and experience that employers demand simply can't be met efficiently by what skills and experience that the unemployed job seekers present.

Dickens and Ghayad take the Beveridge Curve data a bit further however, by examining how the relationship between increased job vacancies correlates to the unemployment rate for multiple stages of unemployment, from the short term, (less than 5 weeks), to the long-term, (more than 26 weeks). The results of this analysis are startling. Essentially the researchers found that only at the longest unemployment durations, does the Beverdige Curve 'jump' or shift.  At the shorter and medium term unemployment periods the data shows consistency with the historical trends, i.e. as job vacancies increase, unemployment rates decrease.

Said more simply - as the economy continues to recover, and more job vacancies get created, the expected reduction in the unemployment rate implied by the Beveridge Curve has held except for the long-term unemployed cohort, defined as being unemployed for greater than 26 weeks. For this group of unemployed workers, and this group only, job vacancy increases have not resulted in the expected and historical level of unemployment reduction. The authors contend then, that a fundamental 'skills gap' problem can't be the cause of aggregate high unemployment, otherwise we would see this effect across the entire range of unemployment durations, not just the longer term. From the Vox piece:

 It is possible that the long-term unemployed are increasingly made up of workers whose skills are not suited to available jobs. However, if this were the case why wouldn’t we see some outward shift in the short-term relationship as well? Furthermore, the fact that the vacancy-unemployment relationship has shifted in all industries when only the workers who were previously employed in those industries are considered calls the mismatch hypothesis into question as well.

 Another possibility is that the long-term unemployed in this recession may be searching less intensively, either because jobs are much harder to find or because of the availability of unprecedented amounts and durations of unemployment benefits.

This seems like a more likely explanation, though if a drop in search intensity is due only to difficulty finding jobs it again raises the question of why we wouldn’t see that at shorter durations as well.

It is interesting and challenging data for sure, but it has some pretty important implications if the author's conclusions are correct. Mainly, that the recovery is passing right by the long-term unemployed. Whether it is job search fatigue and discouragement, or a more troubling employer bias against these people, either way, if this recovery is effectively creating a large and growing class of unemployable (for myriad and hard-to-determine reasons), then even as the total unemployment rate seems to stabilize, the problems we face as a country and an economy will persist.

What do you think - does this 'skills gap' really only exist, (if at all), once someone has been out of work for quite a long time?