Reconsidering the Use of Nonlinearities in Intergenerational Earnings Mobility as a Test for Credit Constraints

Intergenerational earnings regression among Canadian men is nonlinear; middle-earning families experience slower regression. This pattern appears to confirm economic models of educational choice with credit constraints. This paper reexamines the economic model and finds no connection between credit markets and earnings regression nonlinearities. In particular, credit constraints need not produce concavity and concavity does not imply credit market failure. Despite the invalidity of the test, data availability will likely lead to continued research along this path. The paper proposes an amended test using quantile regressions. Applied to Canadian data, the simple liquid-ity constraint conclusion is rejected. I.