Posted Content•
The causal effect of education on earnings
TL;DR: This paper surveys the recent literature on the causal relationship between education and earnings and concludes that the average (or average marginal) return to education is not much below the estimate that emerges from a standard human capital earnings function fit by OLS.
Abstract: This paper surveys the recent literature on the causal relationship between education and earnings. I focus on four areas of work: theoretical and econometric advances in modelling the causal effect of education in the presence of heterogeneous returns to schooling; recent studies that use institutional aspects of the education system to form instrumental variables estimates of the return to schooling; recent studies of the earnings and schooling of twins; and recent attempts to explicitly model sources of heterogeneity in the returns to education. Consistent with earlier surveys of the literature, I conclude that the average (or average marginal) return to education is not much below the estimate that emerges from a standard human capital earnings function fit by OLS. Evidence from the latest studies of identical twins suggests a small upward "ability" bias -- on the order of 10%. A consistent finding among studies using instrumental variables based on institutional changes in the education system is that the estimated returns to schooling are 20-40% above the corresponding OLS estimates. Part of the explanation for this finding may be that marginal returns to schooling for certain subgroups -- particularly relatively disadvantaged groups with low education outcomes -- are higher than the average marginal returns to education in the population as a whole.
Citations
More filters
Posted Content•
TL;DR: In this paper, the authors used law changes in the unemployment compensation system enacted in Germany during the 1980s and 1990s to demonstrate that "incentives" are more important than "capabilities" in determining variation in IT usage.
Abstract: Two main hypotheses can be found in literature on why elderly workers have a lower probability of using information technology than their younger peers: lower learning capabilities and reduced incentives to invest in human capital. I use law changes in the unemployment compensation system enacted in Germany during the 1980s and 1990s to demonstrate that "incentives" are more important than "capabilities" in determining variation in IT usage. Elderly workers only fell behind the IT usage rates of their younger peers during the 1980s and 1990s, when unemployment benefits got increasingly generous, thereby reducing their incentives to invest in human capital.
01 Jan 2012
TL;DR: In this paper, the authors use Swedish registry data to estimate lifetime returns to college and to what extent they vary with respect to observed and unobserved characteristics, including cognitive and non-cognitive ability.
Abstract: A vast amount of research has been devoted to estimating the return to education. Recent work has however clarified that if returns both vary across people and influence their schooling decisions, standard methods typically fail to identify relevant treatment effect parameters. In this paper, I use Swedish registry data to estimate lifetime returns to college and to what extent they vary with respect to observed and unobserved characteristics. The data set includes nearly complete measures of lifetime earnings and high-quality ability measures, which allow me to examine how returns vary with respect to cognitive and noncognitive ability. I implement the local instrumental variables (LIV) procedure (Heckman and Vytlacil, 1999, 2001, 2005) to recover marginal and average treatment effects in the presence of self selection on idiosyncratic gains using the distance to nearest university and local labor market conditions in late adolescence as conditionally exogenous shifters of the selection probability. The large data set allows me to implement the method with more flexibility than in previous studies. My findings lend support to the notion of self selection, but mainly on observed characteristics. In particular, I find that returns vary substantially with respect to both cognitive and noncognitive ability.
01 Jan 2014
TL;DR: The authors analyzed changes in schooling attainment following a reform-driven middle school building program which took place in Denmark during the 1930's through 1950's, and found lifetime returns are much larger than previously thought.
Abstract: Ben-Porath (1967) shows how the returns to human capital depend on the number of years left in the labour market, yet few empirical studies relate schooling to length of working life. No studies go on to consider post-retirement income. We analyze changes in schooling attainment following a reform-driven middle school building program which took place in Denmark during the 1930's through 1950's. Compliers eventually attained on average one more year of schooling, yielding a 12 percent earnings return, a retirement delay of 6 months, and a 6 percent pension income return. Non-pension savings and longevity are unaffected. We find lifetime returns are much larger than previously thought.
Posted Content•
TL;DR: This paper used a credible regression discontinuity design to estimate causal education effects and found that large and significant effects on earnings, political interest, and attitudes toward immigrants were found in the Swiss education system.
Abstract: We use a credible regression discontinuity design to estimate causal education effects. Pupils in the Swiss education system had to pass a centrally organized exam that classified them into different levels of secondary school, and that ultimately determined their educational degree. A major feature of this exam was the local randomization around the classification threshold due to the impossibility of strategic sorting. Our preliminary results suggest large and significant effects on earnings, political interest, and attitudes toward immigrants. The extension to a wider set of data is part of ongoing research.