Equality by lot

A 2021 article by Dimitri Landa and Ryan Pevnick of New York University titled “Is Random Selection a Cure for the Ills of Electoral Representation?” is another indication that sortition may be slowly becoming a political option that needs to be fended off, even in the conservative Anglophone political science academia. The short answer of the paper to its own title is of course a resounding “No!”.

The second paragraph of the paper starts by saying the following:

Our goal in what follows is to develop considerations that have been largely overlooked in conversation regarding the merits of sortition-based proposals, and that should inform our assessment of the viability of those proposals as corrections and alternatives to electoral mechanisms. At the core of those considerations is the analysis of incentives facing citizens and public officials under different institutional schemes.

It turns out that this is a long winded way of saying that sortition is deficient because, unlike elections, it does not provide decision makers with counter-incentives to the inevitable tendency for corruption – i.e., using their political power for their personal benefit. Over and over, in various guises, the article makes the same argument: election provide some mechanism for motivating officials to please the population (namely, their wish to be re-elected), even if in reality this mechanism does not seem to function very well. Sortition on the other hand just lets officials do as they please. The supposed shortcomings of sortition are accentuated by the assertion that empowering an allotted body to make decisions reflects an ideology of “deference” toward that body, which certainly sounds like an anti-democratic, even authoritarian, mindset. In contrast, elections, the authors say, is based on a principle of “accountability” – which, is obviously as democratic as apple pie.

In operational terms, the entire argument relies on the notion that that some, indeed for the argument to work it must be that a significant portion of, decision makers – elected or allotted – would be open to legal bribes – specifically, “lucrative job opportunities” once they have left office – and would be willing to advance narrow interests at the expense of the general interests in return. However, while allotted officials have no reason not to carry out this “kind of implicit, if rarely provable, quid pro quo”, elected officials also have “a strong desire to remain in office” (which, of course, allotted officials do not), so they will tend (again, at least conceivably, even if not in practice) to avoid serving narrow interests, out of fear that doing so would reduce their chance of being reelected.

The claim that this kind of argumentation is “largely overlooked in conversation regarding the merits of sortition-based proposals” is rather extravagant. It would be quite surprising if, in a society that is as economistic as ours, incentives would not figure prominently in any theory of government. In fact, a search on this blog for “accountability” turns up dozens of matches. It is also questionable whether one really needs about 20 pages to make this argument. It may be suspected that the 20 pages are required in order to make the banality of this argument less glaring. Be that as it may, refuting this argument can be done in relatively short space. Most of the points of refutation appear in one of those posts that turn up in the search linked above: a 2014 post of mine titled “Democratic Accountability”.

One more point could be added. While Landa and Pevnick tend to elide this point, the problematic implicit quid-pro-quo scenario, which they use to argue that sortition would likely generate corruption, rests on the claim that while this scenario is plausible, it is “rarely provable”. For if it is provable, then legislation can be made to prohibit it, eliminating the temptation for corruption. But the authors never offer an argument that such a scenario is indeed hard to detect. It would be straightforward, it seems, to set up an oversight panel that would have to approve any lucrative job by a former official, and would examine any case in which a former official becomes unexpectedly rich. Any such case could be treated as being potentially an attempt at a bribe, with all the attendant legal consequences. Presumably, the main reason for believing that making accurate determination of the facts in those cases would be difficult to do, or more difficult than doing so in many other cases of potential conflicts of interests, is that in electoral systems such an examination is never done. But this fact about elections-based systems could very well be no more than another facet of the endemic corruption in those systems. Thus Landa and Pevnick make a generalization from the corruption of the election-based system to the sortition-based system and attribute the failings of the former to the latter.

Filed under: Academia, Elections, Sortition, Theory |