ABSTRACT
This chapter explains the relationship between on-the-job mobility rates of wage earners belonging to the two-earner household. A natural method is then to estimate this relationship using a bivariate duration model that allows for unobserved variables. By allowing for dependency between the unobserved variables, the efficiency of the estimation method is improved. The chapter analyses job mobility of workers belonging to a two-earner household while allowing for a mutual dependence between the unobserved variables. It investigates the consequences of the inclusion of unobserved variables that affect a worker's job mobility jointly with the unobserved variables that affect the job mobility of the worker's spouse. Interesting enough, the correlation between the unobserved variables appears to be highly insignificant and is equal to 0.01. The finding of no correlation seems a puzzle to us in the light of the finding that model III is statistically superior to model II.
