The debate over whether legal rules should be used to redistribute resources in society or whether redistribution should be left exclusively to the tax-and-transfer system has long occupied philosophers, political theorists, economists, and legal academicians. For many years, the conventional wisdom on this question among legal scholars seemed to be that blanket generalizations were inappropriate. All systems of redistribution distort individuals' choices and entail administrative costs. Therefore, the argument went, a universal preference for using the tax-and-transfer system to redistribute is not justified. Rather, the choice among institutions to accomplish society's redistributive goals was considered to be "an empirical one which must be resolved on a case-by-case basis, in the light of detailed information about the circumstances likely to influence the effectiveness of each method of redistribution." At some point, however, a contrary view developed among economists, a view that has become the new conventional wisdom: that income (or wealth) redistribution is always better accomplished through the tax-andtransfer system than through the legal system. One of the arguments traditionally offered to support this view is that redistribution through the legal system is by nature more haphazard (in the sense of less comprehensive and less precise) than redistribution through the tax system. That is, the tax system can redistribute from all better-off to all worse-off individuals, whereas the legal system can redistribute only from those better off to those worse off who are subject to the legal rule in question. In addition, it was often argued that the legal system's capacity for income redistribution is, in any event, small because redistributive legal rules will be effective only in settings in which the parties are not able to "contract around" the redistributive effect of those rules. Thus, the argument goes, legal rules can perform a meaningful redistributive function only in situations involving non-contract-based rules, such as tort rules governing the conduct of strangers ( that is, individuals not in a contractual relationship with one another). In general, with some significant qualifications, we agree that the haphazardness and contracting-around arguments provide a basis for generally preferring the tax-and-transfer system as the primary means of reducing income inequality.
Logue, Kyle D. "Revisiting the Roles of Legal Rules and Tax Rules in Income Redistribution: A Response to Kaplow and Shavell." R. Avraham and D. Fortus, co-authors. Iowa L. Rev. 89, no. 4 (2004): 1125-58.