Race is an effective redistributive proxy- it’s efficient, distributively just, and doesn’t distort work incentives.
Logue, UMich Law Professor, 4 [Kyle D., associate dean for faculty and research and the Douglas A. Kahn Collegiate Professor of Law at the University of Michigan Law School, “Reparations as Redistribution,” Boston University Law Review, 84, No. 5, 1319-74. https://repository.law.umich.edu/cgi/viewcontent.cgi?article=2737&context=articles]
This tension - the efficiency-fairness tradeoff - is impossible for a tax-and transfer regime to avoid entirely. On the one hand, when policymakers try to make the system more progressive or more distributively just, they end up sacrificing efficiency. That is what happens, for example, when marginal tax rates are increased on the highest levels of income, or when special deductions are phased out as income rises. On the other hand, when policymakers try to improve the efficiency of a tax, it is often at the expense of distributive justice. The paradigmatic example of the efficient but distributively unjust tax is the head tax, a type of lump-sum tax that imposes the same tax liability on every individual in the society. Such a tax is perfectly efficient, in the sense that it produces no substitution effect (after all, there is no way to avoid the tax other than to leave the country), but it is obviously unjust, as the poorest person in society pays the same tax as the richest person. This injustice, in fact, is why we use an income tax rather than a head tax, because the income tax, for all its flaws, at least allows us to allocate tax burdens roughly on the basis of relative well-being.
In sum, real-world redistributive programs must find reliable proxies for well-being that are relatively observable. Such redistributive regimes typically have a serious problem with distorting work incentives, because most good proxies for well-being are functions of work effort; therefore, it is generally understood that tradeoffs or compromises must be made between distributive justice concerns and efficiency concerns. These are practical facts about redistributive programs that must be taken into account, and alternative systems of redistribution must be compared with one another in terms of how well they deal with these issues.
B. Race as a Redistributive Proxy: The African American Tax Credit (and other design alternatives)
Now let us return to the three properties of race that make it a potentially useful redistributive tool: correlation, observability, and immutability. First, race correlates strongly with overall well-being.122 In study after study, and in decades of Census surveys, differences in race correlate remarkably well with differences in every conceivable measure of human well-being, for adults and especially for children. This fact alone is sufficient to justify including race, along with income and wealth, in the policymaker's redistributive toolbox and thus is sufficient to justify some degree of racial redistribution. 123
Second, race is relatively observable. Researchers frequently ask people to designate their race on Census surveys and other forms, and the answers are generally considered reliable indicators of race. Individuals know what their race is, and they generally answer truthfully when asked about it. Moreover, according to social science research on the question of defining race, it turns out that external observers also are able to identify someone else's race with a high degree of accuracy, at least in cases in which that "someone else" is, by their own self-designations, either black or white - and not, for example, of mixed race.124
Third, redistributive transfers made purely on the basis of race would not distort work incentives, because race, unlike income or wealth, is immutable, or at least relatively so, and therefore - unlike income or wealth - is not a function of how many hours a person works or in what job. Thus, racial redistribution has the quality of a distributively just lump-sum, that is, efficient, transfer. The transfer could even be means-tested - for example, made available only to blacks below a certain income threshold - without losing its lump-sum quality, so long as the racial transfer was in fact set up, and could credibly be characterized as, a one-shot deal. If, however, the program were designed to provide recurring transfers to African Americans on an annual basis (for example, through a special African American tax deduction or refundable credit, along the lines of the earned income tax credit), such transfers could also be means-tested based on annual income. But that sort of annual means-testing, by making the payments a function of individual taxpayer's annual income, would reintroduce the work-leisure distortion. Of course, that tradeoff between fairness and efficiency might well be appropriate and certainly is nothing new in the design of a tax system. 125