Factors affecting the transmission of earnings across generations: A quantile regression approach

In this paper we estimate intergenerational earnings mobility models using quantile regressions. Quantile regression is a less restrictive estimation approach than more commonly used methods such as ordinary least squares and instrumental variables, which only estimate the mean effects of intergenerational earnings correlation. Further, we investigate the role of education as an intergenerational transmission mechanism for earnings. We find that the intergenerational earnings correlation is greater at the bottom of the son's conditional earnings distribution than at the top, and that controlling for son's education reduces the intergenerational earnings correlation. We also find that education is relatively more valuable at the bottom of the conditional earnings distribution and thus tends to compress the distribution.