Behavioural utilitarianism and distributive justice

What are the distributive implications of utilitarianism? Is it compatible with a concern for equality, as many utilitarians have argued? We analyse these questions in the context of a pure allocation problem. We consider an infinitely-lived economy and, drawing on the behavioural literature, assume that individuals have reference-dependent preferences: agents’ utility is a function of current consumption and a reference point which captures consumption habits, or the agents’ upbringing. Assuming a history of inequalities in consumption, we show that the utilitarian allocation is equalising: starting from an unequal distribution, inequalities decrease over time at the utilitarian optimum. However, even though agents are in a relevant sense identical, equality does not obtain at any finite time.