“We develop a framework where mismatch between vacancies and job seekers across sectors translates into higher unemployment by lowering the aggregate job-finding rate” write Aysegul Sahin, Joseph Song, Giorgio Topa, and Giovanni L. Violante in Mismatch Unemployment on newyorkfed.org.
How much did mismatch contribute to the dynamics of U.S. unemployment around the Great Recession? To address this question, we developed a framework to coherently deﬁne and measure mismatch unemployment. Plausible parameterizations of the model imply that mismatch can explain at most 1/3 of the recent rise in the U.S. unemployment rate. Our formalization of mismatch, and several choices made in our measurement exercise, mean that this estimate should be considered as an upper bound.
The mismatch hypothesis is qualitatively consistent with three features of the Great Recession. First, over the past three years the U.S. Beveridge curve (i.e., the empirical relationship between aggregate unemployment and aggregate vacancies) has displayed a markedrightward movement indicating that, for a given level of vacancies, the current level of unemployment is higher than that implied by the historical data. Put differently, aggregate matching efﬁciency has declined. Second, around half of the job losses in this downturn were concentrated in construction and manufacturing. To the extent that the unemployed in these battered sectors do not search for (or are not hired in) jobs in the sectors which largely weathered the storm (e.g., health care), mismatch would arise across occupations and industries. Third, house prices experienced a sharp fall, especially in certain regions (see e.g., Mian and Suﬁ, 2011). Homeowners who expect their local housing markets to recover may choose to forego job opportunities in other locations to avoid large capital losses from selling their house. Under this “house-lock”conjecture, mismatch between job opportunities and job seekers would arise mostly across locations.
How much of the observed rise in the unemployment rate can be explained by mismatch? The right panel of Figure 3 shows mismatch unemployment (i.e., the difference between the actual and the counterfactual unemployment rates) at the industry level for the 2001-2011 period, as implied by the homogeneous and heterogenous indexes. Table 2 shows the change in mismatch unemployment between October 2009 and the average of 2006.
The main ﬁnding is that worsening mismatch across these seventeen industries explains between 0.59 and 0.75 percentage points of the rise in U.S. unemployment from 2006 to its peak, i.e., at most 14 percent of the increase. Mismatch unemployment has declined since early 2010, but remains above its pre-recession levels.
Mismatch across industries and occupations explains at most one-third of the total observed increase in the unemployment rate, whereas geographical mismatch plays no apparent role. The share of the rise in unemployment explained by occupational mismatch is increasing in the education level.
Mismatch within education groups
Figure 9 shows mismatch unemployment measured at the 2-digit occupation level for different education groups. Note that unemployment dynamics differ greatly by education: for workers with less than a high school degree, the unemployment rate rose from about 7% in 2006 to about 15.3% in 2010, an increase of about 8.5 points. The increase in the unemployment rate over the same time period for high school graduates and those with some college was, respectively, 6.9 and 5.3 percentage points. For college graduates, the unemployment rate went from 2% to 4.7%, an increase of only 2.7 percentage points over
the same period.
The contribution of occupational mismatch to the rise in unemployment between 2006 and 2010 grows as we move from the lowest to the highest education category. In particular, for the less than high school group, mismatch explains a little less than one percentage point (12%) of the 8.5 percentage point increase in unemployment for that group. For high school graduates, mismatch explains 0.89 (13%) out of the 6.9 percentage point increase in unemployment. For those with some college, mismatch explains about 1.0 (18%) out of a 5.3 percentage point rise in unemployment, and for college graduates 0.65 (24%) out of the 2.7 percentage point observed increase. Thus, the fraction of the rise in unemployment that can be attributed to the rise in occupational mismatch increases monotonically with education from about one eighth to roughly one quarter of the increase for each group.
While, admittedly, our approach does not put us in the best position to separately identify the many potential causes of mismatch, we argued that analyzing different layers of disaggregation (e.g., occupation, industry, education, geography), as we do, is informative nevertheless. The absence of an increase in geographical mismatch casts doubts on the “house lock” hypothesis, a conclusion in line with existing research. The non-negligible role played by occupational mismatch, especially for high-skilled workers, leaves room for explanations based on labor demand shifts combined with human capital specialization or with relative wage rigidity.
If mismatch only accounts for a portion of the persistently high unemployment rate, what are the other economic forces at work? As we explained, both the aggregate vacancy rate and aggregate matching efﬁciency are still well below their pre-recession level of 2006. Weak aggregate demand combined with wage rigidity, uncertainty about future productivity and future economic policy, or selective restructuring by ﬁrms during recessions do, qualitatively, imply a slow recovery in job creation. The disincentive effects on job search effort from prolonged extension of unemployment beneﬁts , and the diminished recruitment intensity on ﬁrm’s side are consistent with the fall in aggregate matching efﬁciency. Going forward, disentangling these channels will be paramount in achieving a comprehensive picture of the Great Recession.