This paper was prepared for the annual meeting of the Southern Economic Association, Tampa, FL, November 25, 2013. This paper is released to inform interested parties of ongoing research and to encourage discussion of work in progress. The views expressed herein are those of the author and not necessarily those of the U.S. Census Bureau.

I employ a mixed proportional hazard model to estimate the effect of occupational tenure on the probability of an occupational change, conditional on spell-level unobserved heterogeneity. The individual-level data come from the Panel Study of Income Dynamics and the Survey of Income and Program Participation. The data include information on occupational spells which began before the sample in order to allow for more precise estimates at high levels of tenure. I apply the method proposed by Wooldridge (2005) to solve the initial conditions problem. After accounting for the endogeneity of initial tenure, I find evidence of an increasing relationship between occupational tenure and the probability of an occupational change at higher tenure levels. This effect works through both pecuniary and non-pecuniary channels in PSID data but only through non-pecuniary channels in SIPP data.