In this paper an empirical model is developed where the collective household model is used as a basic framework to describe the time allocation problem. The collective model views household behavior as the outcome of maximizing a household utility function which is a weighted sum of the utility functions of the male and the female. The empirical research that has been done is mainly focused on testing and refuting the unitary model. Moreover, in the bulk of time allocation literature the main accent still lies on the development of theory. The novelty of this paper is that we empirically estimate the two individual utility functions and the household power weight distribution, which is parameterized per household. The model is estimated on a sub-sample of the British Household Panel Survey, consisting of two-earner households. The empirical results suggest that: (1) Given that the weight distribution is wage dependent, preferences of males and females differ, which rejects the unitary model; (2) The power differences are mainly explained by differences in the ratio of the partners' hourly wages; (3) Although there are significant individual variations on average the power distribution in two-earner families is about even; (4) The male tends to be marginally more productive in performing household tasks than the female (5) The preference for total household production is influenced by family size for the female but not for the male (6) Both males and females have a backward bending labor supply curve.