To explore firms' claim that rogue managers act against company policy in perpetuating sex discrimination, this study develops a principal/agent model of manager discrimination, which predicts that discrimination reduces managerial pay. It tests this hypothesis with 2000 Census data to find the effect of wage discrimination against female non-managerial workers on the wages of their managers and male coworkers. This study uses the unexplained portion of the decomposed wage gap to approximate wage discrimination against female non-managerial workers. Under a reasonable set of assumptions, for every one percent discriminatory decrease in the wages of female non-managerial workers, the wages of their male coworkers in the same geographic area fall by at least 0.227% and the wages of managers fall by at least 0.531%, which is consistent with the proposed model
Tobit models are useful for running linear regressions where some observations are cut off by the observation process. Bivariate Tobit models allow the study of two censored variables. To install the bivariate tobit routine in Stata, type . ssc install bitobit in Stata.
In cases where one variable cannot be observed it its value is below the other, the standard bivariate Tobit does not yield the correct results. The Ordered Bivariate Tobit corrects for this case. This paper proposes a log-likelihood function to consistently estimate an ordered SUR Tobit model. This model is useful to analyze time elapsed between two events only observable in a particular order and if they have taken place by observation time.
Retirement age is often used as a proxy for retirement leisure, but if retirement is correlated with mortality this may be misleading. Using data from the Health and Retirement Study and an ordered SUR Tobit model, we analyze the determinants of retirement and death age to see who consumes retirement leisure. We find that men, Hispanics, white collar workers, people in good health, with Defined Contribution pensions, or who have high Defined Benefit accruals consume less retirement leisure. We also find a variety of factors that significantly influence retirement independently but do not affect retirement leisure, resulting in misleading predictions.