for invaluable help in assembling school quality data. NBER working papers are circulated for discussion and comment purposes. They have not been peerreviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

School quality affects upward mobility in educational attainment. This conclusion comes from an analysis of families with coresident teenage children in the 1940 census. We study parents in the bottom quartile of the education distribution and define "upward mobility" as a generational move up the educational ladder to the top three quartiles of the child's cohort. At the state level, upward mobility is strongly tied to teacher wages. This relationship holds when we narrow our focus to families on adjacent sides of state borders in the South, where state minimum salary laws created sharp teacher-wage differences between otherwise similar counties.

It is widely believed that female students benefit from being taught by female teachers, particularly when those teachers serve as counter-stereotypical role models. We study education in rural areas of the US circa 1940--a setting in which there were few professional female exemplars other than teachers--and find that female students were more successful when their primary-school teachers were disproportionately female. Impacts are lifelong: female students taught by female teachers were more likely to move up the educational ladder by completing high school and attending college, and had higher lifetime family income and increased longevity.

This paper aims to reconcile diverging results in the literature on the effects of compulsory schooling reforms on earnings. I point out, through a simple model of human capital accumulation, the importance of identifying parental education information to better target the set of potential compliers. Using parental background data, the empirical analysis uncovers the large and positive effects of a French school leaving age reform previously shown to have produced zero and statistically insignificant effects on the earnings of impacted cohorts. The analysis suggests that identifying parental background information is likely a crucial effort in analysing contemporary compulsory schooling policies.

No abstract