Distributive Justice for Behavioural Welfare Economics
研究了行为偏好不完整时，功利主义方法仍能给出精确的政策建议，并发现功利最优的维度远低于帕累托最优。
Abstract The incompleteness of behavioural preferences can lead many or even all allocations to qualify as Pareto optimal. But the incompleteness does not undercut the precision of utilitarian policy recommendations. Utilitarian methods can be applied to groups of goods or to the multiple social welfare functions that arise when individual preferences are incomplete, and policymakers do not need to provide the preference comparisons that individuals are unable to make for themselves. The utilitarian orderings that result, although also incomplete, can generate a unique optimum. Non-separabilities in consumption reduce this precision but in all cases the dimension of the utilitarian optima drops substantially relative to the Pareto optima.