Do people vote in their self-interest?
Photo by author
In Against Democracy, political philosopher Jason Brennan argues that we should try epistocracy, rule of the enlightened, and maintain this system if it yields better outcomes than the widespread system of democracy. In practice, this would mean restricting the number of people who can vote from every adult to those who can demonstrate, in some way or another, that they are adequately politically-informed. Brennan does not concern himself with the specifics, focussing on rebutting pro-democracy arguments over speculating over the structure of a possible epistocratic experiment. Generally, Brennan envisions a political system where fewer people concern themselves with political matters and where those who do make better decisions on behalf of everyone.
One type of criticism which Against Democracy preempts is the notion that politically-informed people tend to be concentrated in certain communities (e.g. higher wealth and education level) which may cause them, on average, to have interests which differ from the overall societal interest.
In response, Brennan contends that people do not vote in their own self-interest. Beyond the empirical evidence he cites, it is suggested that a purely self-interested person would not vote in the first place. Since the probability of impacting the outcome of an election is practically zero, self-interested voters would be better off doing anything else to maximise their wellbeing.
The obvious response, which goes unchallenged in the book, is that people are not perfectly-informed and may overestimate their impact on electoral outcomes. This is especially true considering the ‘every vote counts’ mantra (which Brennan rigorously criticises) frequently flouted during election times. In addition, voting may actually provide utility to the individual independent of the outcome of an election. Though Brennan portrays politics as a whole as a harmful force in the lives of most citizens, this does not necessarily apply to the specific act of voting, where people may derive satisfaction from voting for their favourite politician. Even a fully self-interested individual behaving entirely rationally who understands the insignificance of their vote may therefore decide to vote anyway (as long as they don’t have something better to do which would bring them greater happiness than voting).
If voters don’t act in their self-interest, it also becomes difficult to explain the discrepancies between party preferences between different demographic groups. For example, why does a greater proportion of the white American population vote Republican than the African American population? It could be the case that white voters are largely unaware of the issues facing African Americans, shifting their perception of where the common interest lies. This is corroborated by the reduced gap between white and black voters when both groups are college-educated and more likely to understand the issues facing different parts of their country.
Tellingly, the gap still does not completely close. This may be because being college-educated is not synonymous with being politically-informed. There are obviously college graduates who are politically-uninformed (or worse, misinformed) and vice versa. However, even if there were a perfect way to measure relevant political knowledge for representative samples of voters from different demographic groups, I suspect that different preferences would persist at the highest levels of political knowledge. At the moment, just as with the epistocratic experiment as a whole, the best anyone can do is speculate.
On the other hand, if people voted in their self-interest, the world would almost certainly be a very different place. Taking the single example of wealth taxation, a tax on ultra-high-net-worth individuals would be beneficial to the vast majority of voters, assuming that government revenues from the 0.1% (or even the 0.01%) funded services for the remainder of the population. Yet, there is inevitable pushback from the general population whenever such a policy is discussed, due to the perceived unfairness in punishing success and hard work. Regardless of the validity of such sentiments, it may very well be that voters consider more than their immediate wellbeing at the ballot box.
It is thus unclear to me as to whether an epistocracy would have to deal with a body of wealthy educated voters benefitting themselves at the expense of others. Though some forms of epistocracy could have this problem, such as granting votes based on performance in a political knowledge test, it should be noted that other set-ups would not have this issue. For example, we could adjust the politically-informed voters based on a set of demographics which we deem to be important such that the vote is representative of the general population. For example, since there are proportionally more low-information female voters than male voters according to Brennan, a politically-informed woman’s vote could be weighed more heavily than that of a politically-informed man. This would ensure that the voting body has the same demographics as the general population without compromising on the average level of political knowledge held by voters.
Another way to deal with the problem of self-interest is to randomly select a sample of the general population and compel them to become politically-informed for the sole purpose of voting in an upcoming election. No one else would be permitted to vote. This would overcome voter self-interest as it exists as a objection to epistocracy over democracy.
Self-motivated or not, the effects of an exclusive informed body of voters are unclear. For better or worse, the epistocratic experiment will probably never be conducted. The answer to the question of self-interest in this context is therefore anyone’s guess.
