Great arguments aren’t always right, but they should be bold, persuasive, and force the scholarly community to respond by testing the arguments’ logic and limitations. In recent years, there are few arguments that have been more generative of thoughtful scholarship than Kaplow and Shavell’s claim that income redistribution should be done solely through the system of taxes and transfers and that legal rules should be chosen solely for their efficiency properties.1 This conclusion is instinctively repugnant to many scholars outside of the law and economics tradition, and surprising to many within it. Yet, first rank economists that they are, Kaplow and Shavell’s logic, at least under the assumptions of the model they use to make their argument, is unassailable.
But, what Kaplow and Shavell’s logic proves and what it has often been taken to prove are two very different things. Although many excellent scholars have offered incisive critiques of the Kaplow and Shavell result, Zach Liscow’s recent note in the Yale Law Journal does as fine a job as I’ve seen of both identifying the reason for this difference and arguing from within a welfarist framework that equitable considerations should apply to legal rules too. The note is admirable in its accessibility, clarity, and rigor. I would include it on the reading list for any law and economics or tax policy seminar that addressed the merits of redistribution through the tax and transfer system.
The slip between what Kaplow and Shavell prove and the more general claim that legal rules should not be used to redistribute income, arises from the fact that Kapow and Shavell consider only legal rules that redistribute in in the same way as the income tax system, by making the amount of damages conditional on the incomes of the affected parties. Liscow asks, instead: what if we choose our rules such that liability varies with a different factor, such as the incomes of the parties that are generally affected (not necessarily the incomes of the parties involved in any particular dispute)?
Liscow first asks us to consider the choice between a strict liability rule for pollution and a negligence rule, where both rules induce the efficient level of care by the polluter. The difference is that under a strict liability rule the polluter bears the cost of the harm and under a negligence rule the party that is harmed bears the cost. If polluters tend to have higher pre-tax incomes than the people being harmed, then by choosing strict liability we induce the efficient level of pollution and redistribute income to the people harmed. This allows us to reduce the amount of distortionary redistribution that would need to take place through the tax and transfer system if we had a negligence rule and thereby increases both efficiency and welfare.
Choosing among equally efficient rules, when redistribution comes for free, is the easy case. But Liscow goes further to specify the circumstances in which even legal rules that are inefficient can increase overall efficiency by redistributing income at a lower distortion cost than the tax system. He then argues that the legal system has certain institutional advantages over the tax system at identifying persons to whom we want to redistribute resources and can, for that reason, play a role in maximizing social welfare. Whereas the income tax conditions tax liability on income, which is only an imperfect proxy for ability-to-pay and other traits we actually care about from a social welfare perspective, legal rules may be able to incorporate additional information that is correlated with these traits and help target redistribution.
More generally, Liscow’s contribution can be seen as reminding us of the complex and sometimes unpredictable interconnectedness of policies in a second-best world. Once we move away from a fantastical world of head taxes or endowment taxes, and conditions of perfect information and perfect competition, the arithmetic of distortions changes such that adding one distortion to another could equal two distortions, or something smaller, or something bigger. In some sense, the genealogy of this insight goes back at least to Lipsey and Lancaster,2 but the importance of this category of challenges to myopic welfare analysis is often overlooked and Liscow provides a fresh and compelling application. The note is essential, and enjoyable, reading for anyone who seriously engages with Kaplow and Shavell’s argument.
- Louis Kaplow and Steven Shavell, Why The Legal System Is Less Efficient Than The Income Tax In Redistributing Income, 23 J. Legal Stud. 667 (1994).
- Richard G. Lipsey and Kelvin Lancaster, The General Theory of Second Best, 24 The Review of Economic Studies 11 (1956).