Election Day

A lot of people today are voting, and a lot of people are not voting, and a lot of other people are wondering if their vote matters in the least. Here is what we wrote on the subject last year. If it is indeed irrational to vote, as many economists believe, then instead of mourning low voter turnout, perhaps we should be lauding low voter turnout as a sign of great rationality on the part of the American public.

But if you really do want higher voter turnout, there are at least three things that could help: pay people to vote (even if the payment is in the form of a lottery ticket); make voting mandatory, as it is in Australia; or — and this would be an interesting, albeit impossible, experiment — forbid anyone from conducting and publicizing pre-election polls. If you don’t know the likely outcome of an election, would you be more willing to vote? I am guessing the answer is a big fat yes. When we get bombarded with poll numbers, supporters of the leading candidates are disinclined to vote because the victory seems in the bag; and supporters of the trailing candidate are disinclined because defeat seems inevitable.

And if none of these measures work, you might just try handing out live pigs or jugs of whiskey, as they used to in the old days.


Why are more votes necessarily better? I'd rather see fewer, well-informed voters casting their ballots than many more who are motivated solely by that lottery ticket (or jug of whiskey) and informed by little more than a negative ad or two on TV.


The way I see it is that we don't exclusively want low voter turn out, and we don't exclusively want high voter turn out. Ideally, we want only well-informed, rational human beings to vote.

Under the current system, we do get a lot of well-informed voters, because one must know enough about a candidates positions to have some opinion, otherwise they wouldn't bother voting without compensation. However, these voters may not be the most rational, because they are incurring the cost (time, effort) involved without any compensation.

Per ballot incentives would add a great deal of rational, but uninformed voters, becuase they have no opinion and only want to receive the subsidy. However, some sort of universal subsidy, to be spent however the consumer--er, voter--chooses would add fewer voters, but these voters are more likely to be both rational and well-informed. One such subsidy might be making election day a holiday. Those who deem themselves to apathetic to decide wouldn't be compelled to vote anyway, however this should create a "time effect" like an income effect. Demand will rise for every activity to which this time may be allocated, including voting. So informed citizens who wanted to vote but otherwise could not would be able to use this extra time to vote, and would probably feel morally obligated to do so now that it was made easier. The very modest increase in total voters would be made up solely if informed citizens for whom voting was never considered because there were too few hours in the day.



I'm not certain motivating people to vote should be our first priority. It seems to me that voting should be done by people who understand the issues and understand what the candidates really stand for (their TV ads certainly make things clear as mud, most of them don't even say who the candidate is, just that the other guy's a [derogatory term]).
But if our goal is to represent the people (even the stupid ones) I think a lottery system might actually work to get people to the polls, a million bucks makes people do crazy things (see: Reality TV). I also like the idea of making Election Day a national holiday. That would really help to underscore the importance (debatable, I know) of voting. It's pretty inconvenient to vote if you've got a standard 9-5.
However, I think that if you're going to really increase the number of people that vote you absolutely need to do something about the voting system. Add the "None of the above" option on the ballot and implement a weighted voting system (I really like A, but B is my second choice, and I hate C) which might allow third parties back in. Hopefully that would break the left/right partisan-ism and get representatives elected to actually represent their constituents.
At any rate I'll be really glad when the elections are over. Politicians just make me weep when I see their manipulative ads and I know they're working on some poor person who wasn't taught critical thinking by our failing school systems.



I think making election day a holiday might lower the turnout of interested voters. If today were a holiday, half my office would have taken off tomorrow as well, and they'd all be out of town somewhere. They'd have kids home from school and needing attention, and they wouldn't want to drag them to the polls. Further, you can't make it a mandatory holiday for private businesses, so many people would have to work anyway. In an admittedly irrational way, it would be easy to say, "Well, I don't get Election Day off, so I guess I won't have time to vote, and that's not my fault, so there goes my civic duty."

Maybe if elections were on Wednesdays it would decrease the vacation effect, but still...


Isn't the reward system already built in with a person's vote? I voted for x because they represent the values I believe in, or I voted for z because I can't stand for what y believes in.

If people aren't voting in the numbers that should be voting, then having a poor substitute (lottery ticket, a paid day off - with proof of voting, etc) will not change the turnout. The problem with our broken representative government isn't the voters, it's the candidates. Negative reinforcement (not voting leads to a fine) might actually get the most voters to the polls the quickest, but good luck finding a politician to really take on that stance.

We vote for candidates who say one thing to get elected and then vote against their constituents' platform and their own supposed ideology. This practice happens on both sides of the aisle.

Run off voting, and election reform do offer some glimmer of hope. Breaking the Electoral College would lead to some changes as well. Whatever can end this cycle of money, ads and lies should be implemented the fastest.



This is one of those things, like health care, publicly funded elections, etc., that drives me nuts, because there are very easy solutions out there, well tested by other Western industrialized countries. You don't have to make election day a national holiday, you just hold it on a Sunday, as they do in most European countries. Is that the only reason these countries have voting rates of around 70-80%? No, but it certainly can't hurt.

But, of course, we'll never see it here.


I like the pigs and whiskey plan


So your idea is to reduce the flow of market information to encourage voting? Really? Has Levitt called to skewer you on that yet?

Isn't it efficient for lower turnout to occur in blowout races. At the same time, polls and prediction markets help find the couple dead heats where voter turnout spikes. These responses make a certain amount of sense to me.

Granted, the odds of 1 vote making a difference are still astronomical (as your article last year points out), but at least polls can help focus the public on voting in a "closer to rational" manner.

In the end, voter turnout rates are a non-issue. In the same way that we all don't need to be Nielsen Rating Families, the sample size is already big enough.


Have you seen the way the district of the house are drawn out, no one wants more voters because they would just have to redraw the lines.

See Delay 2003 Texas


I take it that my post regarding one man/woman one vote is too controversial. Hence its removal.

I pose my questions again. Whose idea of well-informed should we use? What would the acid test be? Land ownership, trivia knowledge, color-blindness, mental health, college education? Those who propose any test are really just hiding behind prejudice and ignorance and ultimately subverting the purpose of democracy, which is active participation in our future.

My vote should not be more or less valued than any other citizen's vote.


There's reason to believe that people have a systematic bias towards xenophobic authoritarianism (see, for example, Bryan Caplan's analysis of the Survey of Americans and Economists on the Economy: "Mises and Bastiat on How Democracy Goes Wrong").
Which raises the question: how did we get our relatively high levels of liberty & wealth?
A reasonable answer is that relatively liberal (in the European sense) institutions arose due to historical accidents & then out-performed societies guided by less liberal institutions.
This analysis suggests that the question of what percentage of people vote is less important than the question of what limitations are imposed on the positions candidates will fill, & how voter turnout influences political institutions. I'm aware that I'm not providing any answers here.


Novo Mestro

I've often wondered (and the Slate article gives the same questions) about the actual punishments for not voting under the Australian regime. If it was a negligible chance of a fine (which it is), then rationally it should have no effect. If people are voting, they'd apparently vote anyway, and not care about whether it is compulsory or not.

I think that the support for the law may be close to vice laws, or growing public support for carbon taxes. People could simply not smoke, or stop driving as much, or not go to the brothel. But they don't stop, and they want a form of public reinforcement, even if it is almost purely symbolic, and the apparent representation of a society's expectations is the law.

So people think they should vote, but without community reinforcement through a symbolic law to make them, they perhaps will not.

Tom Dunstan

As an Aussie, I can answer the question about our system. The punishment for not voting is a non-negligible fine, and the chance of being fined is extremely high. That is, if you're registered to vote (and by law you must be if over 18), then you WILL be fined if you don't show up. It's enough of a fine that just turning up and voting is cheaper.

Arguments around the contructive value of "coerced" votes aside, it seems to me that the single best feature of a compulsory voting system is that it negates a whole class of attacks trying to manipulate the vote. That is, you can change the outcome of an election by doing enough to prevent people (generally voting for your opposition) from voting. Examples: hold the election on a weekday to make it difficult for working people; a boss not giving an employee (or reluctantly giving) time off to go and vote while polls are open (employers must give an employee time off to vote under Australian law, and the fine is expensive enough that everyone does) (this also applies for any power relationship, e.g. abusive husband/wife etc); understaff targeted polling booths (leading to long lines which makes people give up); have overly strict id requirements that voters can't meet; physically prevent people from approaching booths (roadblocks in Florida 2000, OMG); fraudulently ringing up voters and telling them they're not allowed to vote. There are many more. Note that all of the above have definitely occurred in recent US elections, with the possible exception of the power relationship one which doesn't get documented much.

I suspect that the Saturday one (or Sunday of you like) would do a great deal to improve interested voter turnout in the US, probably more than any other factor.



while I am not normally a conspiracy theorist, I have always thought it odd that exit polls in 2000 and 2004 were so far off the actual results of the election. Then this article in "Rolling Stone"
It strikes me that some voters may not be motivated to vote because their votes really are not counted.

That said, the article leads me to two quesions:

1) was the methodology in the "Rolling Stone" article sound?

2) Would there be any economics methodology(s) to prove or disprove the hypothosis that the 2004 election was stolen?



I think Tuesday elections are fine.

Frankly, people can come up with excuses for any day not to vote. The current system assumes that one does not have a 3 hour round-trip commute to work; work locally, vote locally.

I like the idea of banning polling information, but I fear it would be near-impossible in these "wired" times.

The lottery system does not appeal to me. In fact, it might be a source of further voter fraud.

Any possible solutions? I think there are two things that would encourage potential voters to go to the booths. One, discontinue the use of Election College (hopefully, I have the correct term). This would indicate that EVERY vote counts, and not just enough to win the state or locality. The second would be including a "None of the Above" choice (which an ealier commenter had mentioned). Again, this goes to the source of the apathy. If run-offs are needed to determine the will of the people, so be it. Currently, the voters who aren't tied into the "party politics" don't feel that they have a voice.

If we can muzzle the media regarding the speculation of polling results, the combination could recharge the voters. Actually, the "muzzling" could be enforced on the broadcast networks and cable stations.



I've heard two explanations for low voter turnout with which I'd like to enthusiastically agree. Gerrymandering and the electoral college. In the extremes, to paraphrase my senator Barack Obama, constituents don't choose their congressmen, congressmen choose their constituents.

In effect, gerrymandering makes the voters mere commodities with predictable traits to be shipped around allowing a skilled incumbent with helpful colleagues in his state to get re-elected without fail.

Discontinuing the electoral college would greatly increase the value of most votes, making it a more rational choice. Right now, for instance a republican who votes in DC would have to be a masochist or a fool, as would a democrat voting in certain areas of Alabama.

Charles Bronson

Any kind of incentive to get voter turnout to increase would be staunchly opposed by the conservatives. If all the under-represented groups started voting (minorities, the younger generation, the poor) this country would be a very different place, and nothing frightens a Republican more than change.

I know that here in California both the courts and the police tell ex-cons that convicted felons cannot vote, which is an outright lie; convicted felons, so long as they are not on parole, are allowed to vote. This is just one example of those in control going out of their way to maintain the status quo.


regardless of whatever incentive scheme is derived, the elections will still be determined by the average voter of the population.

if you believe yourself to be above average, or below for the honest folks out there, there would remain no reason to vote.

and now to the behavorial economic consequences...


I think we shouldn't concentrate on voter turnout. As many have pointed out, we don't want people to vote just to vote.

What we need to concentrate on is EDUCATING citizens on the issues.

Then they can make an informed decision on whether to vote or not and if they do vote, their vote is actually meaningful.


It's interesting, the first few comments support more informed, intelligent votes and fewer votes from people only turning up to collect a bottle of whisky (or a pig). Having just read THE WISDOM OF CROWDS by James Surowiecki, I personally disagree.

Surowiecki argues (very convincingly) that the larger, more diverse and more independent a group of people, the better a decision they will render. Thus it's important to have uninformed and well informed voters in the pool to get the "best decision".