Polluting the Polls: When Citizens Should Not Vote

Simon,

Thanks for your questions.

Regarding the second one (about demandingness), I’m concerned that voting well requires significant personal effort and time, and so can be costly for the individual voter, and, indirectly for society. The reason it can be costly for society is that the time a person spends becoming a good voter is not spent developing other excellences that benefit the common good. (If it turns out that it takes quite a bit of care and study for Phyllis the Physician to vote well, that might come at the expense of her creating medical breakthroughs. It could be better for us if she ignores politics and just focuses on medicine. This might sound like an extreme case, but I think the point can generalize in a weaker form to other people. I’ll say more about this when I respond to Christian later. ) I do think there are right answers to a large range of political questions, so I’m not worried about that sort of problem in this paper.

Regarding the first question: I say bad voting occurs when a citizen votes without sufficient reason for harmful or unjust policies or for candidates that are likely to enact harmful or unjust policies. The ‘without sufficient reason’ clause is important because one might vote for a harmful policy but not be negligent in doing so. (Perhaps one had excellent justification for thinking a policy would benefit the common good, but it turned out one was wrong. In that case, I don’t want to say that the person voted badly. Or, alternatively, you can say she did vote badly but isn’t culpable for doing so.) The most common ways of bad voting are voting on the basis of 1) immoral beliefs and attitudes, 2) ignorance, and 3) epistemic irrationality and bias. (Regarding 3, the kind of thing I have in mind here is what Bryan Caplan complains about in his book The Myth of the Rational Voter, where he claims that people are systematically incorrect about how economies work because they are “rationally irrational”. But I am not committed to Caplan’s particular claims.)

One objection is that voters might tend to vote on the basis of character or skill rather than for the policies their favored politicians would enact, and perhaps voters are quite good at selecting virtuous or skillful candidates. There’s a grain of truth to this objection. But I think character-based voting might be the most common form of bad voting. Just as having lots of moral virtue doesn’t necessarily make one a good surgeon, neither does it make one a good Senator or President. (Though I don’t want to overstate this.) A candidate might be a really great person but due to misinformation, ignorance, or whatever, have terrible ideas about how to run the country. Right values, wrong means. The same goes with political skills such as the ability to work across party lines and get bills passed. A politician might excel at such things, but still have a detrimental effect on the country because she advocates bad policies.

Christian,

If someone could show me (empirically) that heavy participation is needed to keep democracy working well enough to keep us free and equal, I’d agree that (regrettably) we have an obligation to participate heavily. I’ve seen the claim that heavy participation is needed or good asserted quite frequently, but haven’t seen particularly good evidence for it. [As an aside, I wonder what people think about, e.g., Diana Mutz’s work on how empirically it looks like the things that lead one to participate make one a bad deliberator and the things that make one a good deliberative democrat tend to make one a bad participator.] I think it’s more plausible to claim there’s a level of participation that’s good enough, all things considered, and Western democracies tend more or less get that. It seems that reasonably just constitutional democracies survive despite less than full participation and despite serious shortcomings in citizens’ political virtues. Given the extent of bad voting and other bad forms of participation and their effects on policy, some of these democracies might function better with even less participation than is now seen. What contemporary democracies need most to preserve equality and liberty is not full, informed participation, but an electorate that retains a constitutional culture and remains vigilant enough that it will rise against any leader that tries to abuse their liberties.

So, “weakening democracy” sounds scary. But I don’t think the question here is about whether we want to dispense with democracy in favor of something else. Rather, it’s just about what the optimal amount of voting is, given that we want to keep a democracy. If good voting were cost-free, we’d want to ask people to vote well more than they do. Once we recognize getting more good voting has costs, both for individual citizens and for the common good, we have to be more cautious about recommending more participation. Instead, we’re forced to balance lots of competing considerations.

It might seem at first glance unfair if I benefit from a public good of good governance but don’t myself provide for it. However, there’s another way to look at it. We each receive a bundle of goods from society: cultural, social, economic, political, etc. (You can parse this in different ways. The details won’t matter. The point is just that you get different kinds of goods in virtue of living in a flourishing society.) Suppose we grant that it’s unfair for you to receive these goods without paying back something in return (though the unfairness of it won’t necessarily imply you should be forced to pay your debt). Does it follow that you have to pay for each kind of good with a similar kind of good? Not necessarily. Rather, it might be that we get to specialize. I pay for the bundle of goods by providing economic opportunities. You provide good culture. Someone else provides good governance. A fourth person provides a mix of the three. Etc. We can pay our debts with different currencies, and so there’s no unfairness. [In a paper I’m working on now, I’m playing with the idea that by providing one type of good, I am indirectly providing all the others, because I facilitate other people’s specialization.] I think liberals should favor this view of how we pay our debts.

So, if non-participation means contributing to the downfall of a just society, yeah, that’s bad. But if non-participation means contributing to the common good in others ways, then it’s not bad, but possibly commendable.

More later…

Aaron:

Could you elaborate on how my definition of bad voting might set the bar too high?

You bring up a good point about reform. First, keep in mind, this is a piece of moral philosophy, the task of which is to identify a true moral claim. It’s not a piece of civic education or a pamphlet for civic reform. Its goal is to say there is a problem, but it takes much more work to show how best to fix the problem. I agree that ceteris paribus we’d want to lower the impediments to good voting. (I don’t know if the media conspires to impose impediments, though they may in fact impede good voting.) You say, “It may be that being a good voter is not all that difficult within the right kind of political institutions and culture.” I hope so. But that doesn’t imply that people shouldn’t vote badly.

It might well be the case that if this paper were widely read by the public, it would induce undesirable results. Maybe it would induce good voters not to vote out of self-doubt and would induce bad voters to vote more. That would make the paper self-effacing, but it wouldn’t make its claims false.

For instance, suppose certain critics of utilitarianism are correct when they claim that if people accepted utilitarianism, this would make the world worse by utilitarian standards, simply because most people are not good at employing such standards. If so, this does not show that utilitarian standards are false. Rather, it just shows that we should not advertise them. As David Brink notes, there is a difference between a criterion of right and a method for making decisions. The former is about what makes actions right or wrong, but the latter is about figuring out how to do what’s right or wrong. A good method for A might different for B because they have different cognitive abilities. A is good at making calculations while B isn’t. But the standard of right action is the same for both. The point of the decision-making method is to help them get to the right action.

So, I’m not giving up on institutional reform, nor am I necessarily recommending that anyone go around telling laypeople they ought not vote badly. Rather, I am just arguing that people ought not vote badly. Whether we should tell them that depends on what will happen if they hear it.

Hi Aaron,

Thanks for the comments. On Monday I’ll probably post more about bad voting.

Anyways, I agree with quite a few of your claims. We should (like Macedo argues in his recent Brookings Institute book) lower the costs and impediments of good voting and civic engagement. Also, it is a bad thing if people of lower SE-status are consistently placed in positions where they are unable to vote well due to a lack of education or some other good. These things should be changed. However, it doesn’t follow that people should not vote badly. I agree that we should educate people better, and I also hold that they should not vote badly. We should probably also have more economics courses in high schools, but that doesn’t mean that people shouldn’t vote badly. (There may be cases where a lack of education, etc., excuses one from voting badly, if one genuinely cannot figure out that one has terrible political views and it’s not one’s fault for having these views.)

If we see people are voting badly, this calls, all things equal, for institutional reform (if we can get it with a net positive effect). But the fact that calls for institutional reform doesn’t automatically let individuals off the hook.

Steven,

As it happens, I reject deliberative democracy, though a rejection of deliberative democracy is not required for my thesis. (See Gerry Gaus’s critiques for some the main reasons why I don’t accept deliberative democracy.) Do you think I need to argue against deliberative democracy to make the argument secure? What if I specify this is a non-ideal theory, a theory about what to do under non-ideal circumstances? Does that make a difference?

Ryan,

Good article suggestion. In fact, I cited that article and their earlier book as a source of an objection (individual bad votes have vanishingly small expected utility, so why say morality requires you not to vote badly?) Geoff Brennan even kindly provided extensive feedback on this article. Their work is some of what inspired this in the first place.

Steve,

The account of the “bad voter” given by the deliberative democrat is different from mine. I’m worried about people who vote for harmful policies or candidates likely to support harmful policies. I think you have an obligation to try to avoid imposing bad governance on people. (I agree with the deliberative democrat that you should offer public reasons, you should accept your own fallibility, and you should typically play fair and go along with the outcomes of elections even when your side loses. But that’s more about being a good democrat than about voting well.) What do I mean by “harmful”? Feel free to plug in any plausible, minimal theory of harmful policies here. For instance, some things I think are wrong are voting (without a good excuse) for racist, sexist or war-mongering policies, voting for policies that retard growth or destroy wealth or other basic goods without at least compensating by producing some other goods, voting for policies that exploit the few for the majority’s sake or the majority for the sake of the few, etc.

I am skeptical not about ideal theory itself, but whether you need to do ideal theory in order to do non-ideal theory. I think the relationship between the two is highly contingent. Rawls says you need ideal theory to give you a guidepost to make reforms in non-ideal worlds, but I’m not convinced.

Theorizing about institutions is, in a way, like engineering. Ideal theory is like designing cars on the assumption that they’ll never encounter slippery pavement, will never overheat, and will never be driven by bad drivers. If we did not have to worry about such things, we might be able to build cars that would cruise at hundreds of miles per hour. We might find it theoretically lamentable that our cars are not ideal in that sense, but we still have good practical reason here and now to not build cars like that. And note that we don’t need to design the first kind of car in order to design our real life cars.

So, a little more:

Note first that defining ‘bad voting’ as I have doesn’t make it tautologous that one shouldn’t vote badly. You can feel free to treat ‘voting badly’ as a technical term. Even on this definition (where voting badly is voting without sufficient justification for candidates likely to support harmful policies, etc.), it remains an open question whether one shouldn’t vote badly. It needs an argument.

In most elections, individual bad votes are unlikely to have significant expected disutility. Suppose electing candidate P over candidate Q will cost the economy $33 billion dollars next year, and this comparative loss will not be offset by any other value P provides. At the time of the election, P commands an anticipated proportional majority of 50.5% of the voters (i.e., there is a 50.5% chance a random voter will vote for P), and there is a turn out of 122,293,332 voters (the number of voters in the last U.S. presidential election). In this case, if I also vote for P, the objectively worse candidate, my individual vote has an expected disutility of a mere $4.77 x 10-2650, many thousands of orders of magnitude below a penny. (This uses Geoff Brennan and Loren Lomasky’s formulae for calculating the expected utility of individual votes.)

Bad voting is a collective action problem of sorts. The problem isn’t so much that I vote badly but that lots of us vote badly. So what does morality require us to do in collective action problems? I think a fairly plausible claim is this: you should not contribute to collective action problems when you can avoid doing so at low personal cost. This claim is plausible in part because it is relatively undemanding and can be shown to fit within a number of plausible background theory. (This rule would fit inside a plausible rule consequentialism, for example, or could be derived from the categorical imperative.)

In some collective action problems, like the prisoner’s dilemma or tragic commons, I can’t solve the problem on my own. Attempting to avoid contributing to the problem doesn’t solve the problem. Instead, it opens me to exploitation as a sucker without doing any good.

If everyone in my small village litters, maybe I could solve the problem by cleaning litter for 90 hours a week. But many think it’s implausible that morality would require you to do that. That’s too demanding and unfair. (I realize some think morality is often highly demanding, and I won’t argue this point here. But here I’m arguing for a weak thesis about avoiding collective harms, so those with a demanding conceptions of ethics will accept the weak thesis as a corollary of a stronger claim.)

Bad voting is not like that. I do suffer a few small personal costs if I stay home (since people feel good about themselves for voting, even when they shouldn’t, because they just polluted democracy with a vote for a racist or regressive candidate). But it’s generally not very onerous if I abstain.

Compare this to polluting the air. My individual contribution to air pollution is negligible, perhaps. The same goes for you. But collectively we’re causing a problem. So, if we need to solve the problem, it’s plausible that everyone contributing to the problem should bear some burden in solving it. (Perhaps, pro tanto, they should bear it equally, or perhaps in proportion to their contribution.) It would be unfair if, for example, half the population stops polluting while the other goes on as usual. We bad voters should not vote because it is harmful to everyone, but I, the individual bad voter, should not vote because it is unfair that I benefit from polluting democracy as I please while others suffer the burden of polluting democracy less. Ceteris paribus, we should share the burdens of not polluting the polls.

There are also collective costs from bad voters staying home. Widespread voting helps produce more social cohesion. It’s at least empirically possible that when bad voters vote, this tends to make them care about voting more, and this may inspire them to reform and become better voters. I think these costs are likely to be outweighed by the benefits of reducing bad voting, but it’s hard to say without something like an empirical study of the indirect positive effects of bad voting. (The literature searches I’ve done haven’t turned up much. Lots of people assert things like this, and then cite others asserting it, but I haven’t found much evidence.) Another complaint is that it’s hard to take democracy seriously when most voters abstain from voting. I agree, but in response, it’s also hard to take democracy seriously when a large percentage of bad voters vote. Regardless, democracy performs better, even with low voter participation, than its competitors (oligarchy, etc.) do. So, at worst, low voter participation means we are not able to take democracy as seriously as some people would like to, but this doesn’t mean we must replace democracy with something else.

Hi Ryan,

First, I do think character has some role to play when it comes to voting. You don’t want to hand the reigns of power to a terribly vicious person.

Still, even if a voter simply intends to vote for the person with the best character, I don’t think this lets a voter off the hook.

Individual voters are (on my view) obligated to refrain from helping to impose bad governance on the polity. Voting for people of good character can violate this duty. Having good moral character does not reliably track having good ideas about policies. Many incredibly virtuous people advocate harmful policies (because they mistakenly believe them to be helpful, due to misinformation about economics, etc.). So, on my view, you in general shouldn’t vote for a virtuous person unless you have good reason to think the policies that person will try to implement will be good rather than harmful.

The same goes for political skill as opposed to character. Having the ability to get things done, get bills passed, make compromises, etc., is great if it leads to good policies, and awful if it leads to bad policies. So, you shouldn’t vote for someone on the basis of their skill unless you have good reason to think the policies she will try to implement will be good rather than harmful.

In summary, character or skill-based voting is acceptable only to the degree that it tracks the tendency to produce good rather than bad governance.

Cheers,
J

I did a short interview with the Guardian a few days ago about this paper for their daily new broadcast/podcast. Here’s a standalone version: http://www.guardian.co.uk/world/audio/2008/oct/23/political-philosophy