THE BACHELOR WAGE PENALTY HYPOTHESIS EVIDENCE FROM BRITISH HOUSEHOLD PANEL DATA

An earnings gap has been previously shown between bachelor men and married men. The gap appears higher when estimated using OLS .This gap is examined here by considering simple econometrics and with the use of British Household Panel Survey data. The estimate of the gap found from pooled OLS is higher due to unobserved effects; possibly including individual ability. Thus, married men may have earned more even before marriage. The remaining gap of the Fixed Effect estimation of 3.1% as compared to Pooled OLS explains the productivity hypothesis. Married men might have becoming more productive or they have attributes which suit both to “marriage market” and Labour market. Fixed effects estimation shows the correlation between explanatory variables and unobserved effect reported by Stata to be -0.115, which is very low, and the Hausman test strongly rejects that unobserved effect is not correlated with explanatory variable by 2 X of 367.08 even at P value of .000. Hence the fixed effect estimation is better because it allows arbitrary correlation between unobserved effect and explanatory variables. Further, estimates from a binary choice model show men with higher wages have a greater chance to be selected in a “marriage markets” than those with low wages.