Labor Market Polarization and the Great Divergence: Theory and Evidence
- Date Posted:
- Is Database:
- Database
Large cities in France btw 1994-2015 lost middle wage jobs, but they were replaced two-to-one with high paid jobs, a reverse pattern in smaller cities @DonaldDavis @nberpub.
Evidence from France documenting labor market polarization btw large and small cities; study finds a tight link btw large cities becoming even more skilled relative to lower skilled smaller cities, which have become less skilled. They don’t address housing costs.
“….We examine….labor market polarization and the great divergence using detailed data over the period 1994-2015 for a sample of 117 French cities. The data employed includes inter alia population, skills, employment by occupation, and measures of routinizability and offshorability of jobs. The data validate baseline predictions from the theory, including a log-supermodular distribution of skills and differential patterns of employment, whereby at baseline larger cities have less employment in middle-paid jobs and relatively more employment in high paid jobs. Consistent with prior work,we find labor market polarization in the aggregate for France in the study period. We go beyond this to show that labor market polarization is very close to ubiquitous at the local level, with 115 of 117 cities sharing this feature. In spite of this, we also show that these patterns differ importantly in cities of different sizes. Large cities have a sharper loss of middle-paid jobs. Yet in the large cities these jobs are replaced two-to-one with high paid jobs, and the reverse pattern in small cities.All of these observations are consistent with our theoretical framework. In the prior work on labor market polarization, the primary spatial concept of concern was exposure. The literature showed that the regions most exposed to middle-paid routinizable and offshorable jobs at baseline were also the regions that had the sharpest decline in these jobs in the periods of study. While very insightful, these studies also have limitations. First, these are only a subset of the lost middle-paid jobs, and so the empirics are unclear to what extent this exposure is also indicative of the loss of other middle-paid jobs or of the entire class of middle-paid jobs. This would require thinking about dimensions of heterogeneity within the class of middle-paid jobs that the prior frameworks are not designed to address, but which is central to what we do. Our data confirms the prior result that cities with a large initial exposure to the most routinizable and offshorable jobs indeed have the sharpest contraction of these jobs in the study period. Importantly, though, we find that the loss of other middle-paid jobs and indeed of middle-paid jobs taken as a whole have the reverse pattern that might be expected from this. It is the large cities that have the greatest loss of the middle-paid jobs, even though the initial exposure to these jobs is smaller in large cities. Moreover, consistent with our framework, the primary difference between large and small cities occurs in the upper tail of middle-paid jobs, which decline sharply in large cities even as high skill jobs there expand greatly. In sum, we find that the period of study identifies two Frances. In the France of large cities, there is a dramatic change, as there is a sharp contraction of middle-paid jobs, particularly at the top end of these. However in the France of large cities, these middle-paid jobs are largely replaced by high-paid jobs, with a more modest expansion of low-paid jobs. Still there is very sharp polarization within these cities. In the France of small cities, there is a strong, yet more moderated, loss of middle-paid jobs. Some high paid jobs are gained, but the lost middle-paid jobs are primarily replaced by low paid jobs. Polarization of jobs in the aggregate and within cities is accompanied by a great divergence between the Frances of large and small cities. …”

Donald Davis, Eric Mengus and Tomasz Michalski, "Labor Market Polarization and the Great Divergence: Theory and Evidence," National Bureau Of Economic Research, April 2020, https://www.nber.org/papers/w26955


















Ed Comment:Add how the pandemic broke the valley’s hold on tech jobs. I think this will accelerate something I fear—the offshoring of tech jobs abroad, especially the most technical jobs. That’s going to lead to the more entrepreneurialism abroad or the most valuable kind. The more our tech giants can zoom in the talent and breakthrough, the less they will care about US working. There will be tech hubs in every country attracting their best talent.