The cost of not redistributing money- an interlude on interpersonal welfare comparisons

I’m eventually going to release a second part to my essay on the costs of not redistributing money, considering the extra costs caused by relative income effects. However I wanted to take some space here to respond to a family of objections to any project of measuring the welfare costs of inequality- objections to the comparison of utility between persons.

I.

This post is part of a series on the utility losses created by economic inequality, you can read part 1 here.

One of the common objections to any argument from increasing aggregate utility to redistribution is that interpersonal utility comparisons are illegitimate.

This can come in a variety of forms. The first form is what I call the appeal to the sensitivity of the rich- the notion that rich people care more about money than the non-rich, and this must be factored into the utilitarian calculus. It’s a common sense objection which I will concede carries some weight, but it turns out that weight is extremely small, even if the rich care about money a lot more than the poor.

Let us take an elasticity of marginal utility of income of 1.5 as we did in part I of this series. Assuming all households care the same about money, someone who earns just enough to place them in the top 1% of household incomes (430,600 dollars) has a marginal utility of income equal to 5% of the marginal utility of income as the median households (59,039 dollars). Even if their utility of money is twice as much at any given level of income than the median household, they will only value money their marginal dollar 10% as much as a person at the median income values their marginal dollar. This implies that some redistribution from the richest household to the median would be a utilitarian improvement unless ninety percent or more was lost in the transfer (again, assuming no relative income effects).

II.

The other objection I find both more and less weighty. This objection is that the notion of comparing the utilities of persons is conceptually ill-founded. You can compare, using various techniques, how much a single person wants one thing versus another thing. What would such a comparison even look like between persons- and what does it even mean?

My response to this is that it is a fascinating and important question, but to allow it to delay research or practical efforts on the topic of redistribution would be an inappropriate interpolation of philosophy into social science and public policy. We take it as given in everyday life that such comparisons are possible and deploy them in making decisions regarding the best allocation of resources. To see why being overly concerned about these arguments in the context of wealth redistribution doesn’t fit with our behaviour in other spheres, consider a thought experiment:

Suppose a hospital was running short on pain medication, such that not all beneficial prescriptions of pain alleviating medication could be carried out. It therefore uses various measures of pain- such as pain scales, clinical knowledge, assessment of facial and auditory cues of pain etc. to work out who could benefit the most from pain relief medication, and to prioritise their needs, rationing the limited supplies accordingly.

It might be interesting question to ask ‘but what does it mean to compare the pain of Jack to Jill?’. But almost no one is going to take this as a serious argument not to use the aforementioned information in determining the best allocation of (now scarce) medicines between them. In a practical setting we’re not too worried by the possibility that Jack might be ‘feel’ a billion times worse when experiencing a similar level of pain than Jill, or that Jill’s pain may be of a wholly incomparable character to Jack’s. We just get on with it, on the assumption that your agonised screech implies disutility worse than my painful grimace, and so on. Failure to do so would be both absurd and inhumane.

To put it simply- to the extent that our current conceptual tools have difficulty making sense of comparisons of utility between persons, it is our job to create better conceptual tools, not to abandon such comparisons.

There are lots of unresolved philosophical problems in this world- Hume for example documents many reasons to be sceptical about the empirical sciences in his Enquiry Concerning Human Understanding but he never suggested stopping empirical science until we had it all sorted it out. As a matter of common sense we compare magnitudes of need and desire all the time, and do so almost reflexively when the situation calls for it.