AME Labour Economics-seminarium i arbetsmarknadsekonomi vid Institutet för social forskning (SOFI).

Jan Stuhler från UC3M presenterar "Intergenerational Mobility and Assortative Mating in the US" (skriven tillsammans med Lola Collado och Ignacio Ortuño-Ortín).

Detta seminarium hålls på plats, det går inte att delta via länk.

Läs mer om våra seminarieserier, workshops och konferenser här:

Konferenser och seminarier

Abstract

We provide novel evidence on intergenerational mobility and assortative mating in the US, based on longitudinally linked records from the 1900-1940 US Census. In contrast to earlier work, we use an extended-kin design that compares the correlation in years of schooling across 55 different kinship moments. By abstracting from the influence of measurement error in socio-economic status, this method provides more credible estimates of the level of mobility and sorting than conventional parent-child measures. Moreover, the method allows us to understand the “anatomy” of intergenerational persistence, decomposing overall persistence into assortative and other margins. First, we show that mobility is lower than conventional estimates suggest. Particularly striking is the degree of assortative matching, with a spousal correlation in latent advantages of 0.87 (i.e., 40% larger than the spousal correlation in schooling). Second, we study differences in the level and anatomy of intergenerational mobility across regions. As earlier studies, we find much lower mobility in the South than other regions, and illustrate that this result holds across all kinship types. Despite greater spatial segregation, assortative matching was only slightly stronger in the South as compared to other regions. Differences in mobility are instead due to (i) schooling being a better proxy for status, and (ii) stronger intergenerational transmission in the South as compared to other regions. Finally, we show that early 20th century US has only slightly less mobility, and moderately more sorting, than modern-day Sweden. While the observed kinship correlations are much larger in the US, most of that gap is due to schooling being a worse proxy for latent status in Sweden compared to the US.