What Is Redistribution, Really?

Last week, the New York Times ran a piece claiming that Obama rejected Rebecca Blank for a position as leader of his Council of Economic Advisors because, 21 years ago, she wrote that a “commitment to economic justice necessarily implies a commitment to the redistribution of economic resources, so that the poor and the dispossessed are more fully included in the economic system.” Reportedly wary of the word “redistribution” and its potential attachment to the Affordable Care Act, Obama decided to skip over Blank to avoid a bruising confirmation process.

For some time, I have had a mild obsession with the word “redistribution” for many of the reasons reportedly motivating Obama’s move here. Obama is absolutely correct to think that the word does an enormous amount of ideological work. It is not a neutral descriptor of some kind of policy; rather, it is a word dripping with normative connotations of unjust theft and the like. Probably for this reason, there is a ton of literature specifically focused on the question of what exactly redistribution is.

In my view, the word “redistribution” is–as we generally use it at least–basically incoherent. To refer to something as “redistribution” and have it pack all of the normative punch we ascribe to that word requires that you first establish a baseline distribution against which redistribution can even be measured. In the literature, this is generally referred to as “the baseline problem.” How do you pick a baseline in an objective, non-normative way? The answer is that you can’t. Unless you just pick a baseline at random, what baseline you use to measure redistribution against is ultimately a function of political ideology, making “redistribution” so conceived an ideological judgment call. The choice of baseline then is either arbitrary, in which case “redistribution” lacks the connotative punch it has in reality, or is informed by normative ideas about what truly belongs to whom, in which case the connotative punch of “redistribution” is actually just derivative of entirely contestable theories of what truly belongs to whom.

So for instance, it is very common for those less schooled on this topic to define “redistribution” in reference to what Murphy and Nagel describe as the Everyday Libertarian baseline. Roughly speaking, on this usage, the baseline distribution against which “redistribution” is measured is pre-tax market income. This income is often conceived of as being independent from government somehow, with all other kinds of incomes that deviate from the pre-tax market baseline as merely the consequence of government economic intervention.

This is nonsense of course. Functionally speaking, the massive apparatus that generates pre-tax market incomes, i.e. the economy, is actually one big government program. The government constructs the economic system through law-making–e.g. by creating property law, contract law, commercial law, corporate law, securities law, labor law, employment law, employee benefits law, tort law, and so on–and then administers this constructed economic system every single day through the use of police, courts, jails, administrative agencies, and various other enforcement mechanisms. This apparatus could be constructed some other way, and if it were, the distribution of “pre-tax” market incomes could be dramatically altered. Moreover, since all of the government institutions that construct and administer the economy require tax revenues to operate, “pre-tax” market incomes aren’t actually prior to taxes. The pre-tax market income is thus a myth and the usage of it as a baseline for measuring something called “redistribution” does not make a great deal of sense. Nonetheless, famous commentators like Ezra Klein use that baseline anyways, with no indication that they’ve ever thought through or considered the extensive issues involved in doing so.

Those a bit more sophisticated tend to shy away from this approach to the word “redistribution,” opting for something a tad bit less ideological and incoherent. So, for instance, Dylan Matthews defines the word “redistribution” in purely relative terms against the counterfactual baseline. Under this usage, a policy like food stamps is considered redistribution relative to the no-food-stamps baseline, but also the policy no-food-stamps (or cutting food stamps) is considered redistribution against the yes-food-stamps baseline. Every single complete set of economic institutions is therefore redistributive relative to every other complete set of economic institutions. The word “redistribution”, then, comes to refer merely to the extent to which various distributive schemes differ, not some deviation from a specific baseline.

Mike Konczal, on the other hand, defines the word “redistribution” in terms of policy intent. He notes that while all policies have distributive effects, only some have distributive intent, which is to say that they are policies that try to achieve a specific distributive result (as opposed to simply promulgating general rules of the game). As with Matthews’ approach, this too gets around a lot of the ideological issues surrounding “the baseline problem.” However, it runs into significant line-drawing problems because things that might look more like general rules could most certainly be motivated by distributive intent (the decision not to tax carbon, for instance, seems almost certainly motivated in part by an effort to make sure oil companies get more money). Additionally, this definition lends itself to the conclusion that both installing and withdrawing “redistributive” programs is redistributive. Is there any doubt, for instance, that the GOP’s recent war on food stamps is not driven by distributive intent, namely the intent to change the distribution of resources provided to food stamp recipients whom they regard as undeserving?

Ultimately, there is obviously no answer to what “redistribution” really is. It is a word and can mean whatever you want it to. However, given its ambiguity and the massive normative punch it packs among real people in real life, folks should probably be careful to actually spell out what they are talking about when they use the word. But to ask this of them requires that they have actually contemplated or even been superficially exposed to the significant philosophical issues involved in the idea of redistribution, specifically those around the ideas of just entitlement and economic justice more generally. But most wonky sorts that dominate policy discussions and reporting do not appear to have any such philosophical background, which creates the conditions in which sloppy, incoherent, and ambiguous usages of the term flourish.