The Measurement of Reperations

With the recent unveiling of a list of demands of the Movement for Black Lives, many questions have been answered about the policy goals of the social movement. I agree with many of these policy proposals, and would encourage any readers to look through them, in all their detail and nuance. I’ll post a few articles going through some of the policy proposals and attempt to give a normative analysis of the effects of these policies, but the ultimate rationale of these policies is given by the movement itself, and anything I can show is simply an alternative framing of a very well done, and well thought out process by many talented activists. I would like to use this piece to briefly overview the proposal to from a committee to study the idea of reparations, and some of the challenges to doing an effective analysis here. This is in no way an argument against reparations, or the idea of studying them, and if anything, is intended to reinforce the case that a strongly supported, and well funded, study is necessary.

I first want to present the case of what in my mind would be a plausible way of studying this issue, and present the idea of why it would be a bad methodology. The framework of the study I will be discussing here is this, though it can be applied to many similar situations. Before I really dig in to this, I need to give a definition of what exactly reparations are. For that definition, I will use the first paragraph listed here.

Typically, in studying the potential damage caused by a policy(in this case, slavery and subsequent discrimination) the approach is to take what is called an equivalent variation. This essentially asks the question, “How much money would it take to compensate someone so they would have the same level of happiness under the new policy as the old?”. At it’s face value, this type of analysis would seem ideal for addressing the question of how to value reparations, and it may well be the type of analysis needed. The primary difficulty however, comes from choosing what the base points of comparison are. One of the most obvious routes of analysis for this is to take the well being of those who are currently living in Africa, and were not a part of the slave trade, and comparing that to the well being of descendants of slaves living in America today. Given the relative poverty of Africa today, it would not be surprising if doing the study in this manner showed an overall net increase in well being for those living in America today, diminishing the case for reparations. This however, is not a good way to study the issue, for two reasons I can readily identify.

The first flaw in this methodology, is that to take an equivalent variation, one must assume one of two things. The easiest assumption to make, is that of identical utility functions across people(i.e. how money maps to happiness). This is a dubious assumption at best. There is little evidence that two individuals will have identical utility functions, so to assume that utility functions are identical across cultures is almost absurd. This assumption does not need to hold, as long as one can come up with two utility functions that accurately map the difference in preference across cultures, but at our current level of knowledge and ability, anyone claiming to have this ability is probably wrong.

The second flaw invalidates the methodology, even if one could accurately approximate the utility functions necessary for the computations. If the development of Africa happened in a vacuum, away from international influences, one could imagine the study to have more merit. This however, is not the case. The imperial influence of America, and western society as a whole reaches far and wide. The theft of natural resources, overthrowing of democratically elected leaders, and outright war with other nations cannot be discounted when determining the relative well being of one group of people to another. If one wanted to study the case of reparations based on the well being of African Americans descendant of slaves, relative to the Africans, one would have to begin with estimating the effects of colonialism. This is a very difficult thing to study as well; how do you even begin to value the destruction of a culture?

Another way to study this issue would be to view it as an immigration study. Note that this is not to treat slavery as immigration, but to value the difference between them. The way of looking at this would be to assume identical people, that immigrated in two different ways, through slavery, and through conventional means. The analysis here would be to determine how much money we would have to give to people descended through conventional immigration, for them to prefer to have been descended from slaves. I would argue that this is a better method of study, as it removes the problem of colonial rule, though the problem measuring utility across cultures remains. It also discounts the discrimination that immigrants of all nationalities have historically faced in the United States. Given that one of these proposals is a universal basic income for all groups, with an increased value for those of African descent, this might help answer the question of what that amount should be, relative to other marginalized groups, but would not answer the question of how to value this with respect to white Americans.

The problem of how to value these with respect to white Americans would seem to have a similar solutions. Simply ask a similar question, of how much would white Americans have to be compensated to be descended from slaves. This however, has plenty of pitfalls as well. It assumes that white Americans can even imagine the additional costs they would bear(a more relatable version of the cross cultural utility problem). It also doesn’t take into account the idea of relative time of immigration. Those with families who immigrated more recently will face more costs than those who immigrated less recently, and that is a difficult issue to reconcile. It also leaves a more theoretical question of how much the costs of immigration, and slavery, are due to cultural biases, and how much they are due to the time needed to learn about the cultural norms and taboos of a new society, and how, if at all, those should be valued.

Overall, this is a difficult issue to even contemplate how to study, and the answers need to come from people more knowledgeable, and smarter than I. This is the case for a very well funded study of the issue, with collaboration across many fields, and strong input and leadership from marginalized groups to ensure their interests are fairly accounted for.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s