FREAK-TV: Why Economists Don’t Vote


Today is Election Day, albeit a quiet one. There isn’t much at stake in New York. There’s more action in New Jersey, though voter turnout is expected to be low, as it is in California. The irony is that the typical voter is more likely to have an impact in a smaller election than in a larger one, but it’s the bigger elections that draw far more voters.

In this new installment of FREAK-TV, Levitt explains why many economists don’t bother to vote — but why you should, if you really feel like it. If you find this viewpoint depressing, and want to become even further depressed, you should read this article we wrote a couple years ago about the utility of voting.

See you at the polls.

Or not.


If more economists decided to vote, suddenly the 'economist vote' means something and can be a deciding factor

If you were the ONLY person to say "well maybe my vote won't do much" then it would be true

Dan Hirschman

I love this video post because it shows a clear example of how economists think about issues and that's beautiful.

The economist's viewpoint also pretty spectacularly worthless in this particular case, at least if your goal is to predict or explain behavior.

You can always invent a reason why people get utility out of a behavior. But once you go down that path economics ceases to be a predictive science.

The problem is that economists want it both ways - they want to assume everyone is utility-maximizing all the time (which thus requires inventing a utility to voting, or other seemingly useless behaviors) and to assume that people tend to maximize certain obvious things (like income or health outcomes). Once you open the floodgates to any sort of outcome giving utility, economics itself loses a lot of utility (in the old fashioned sense).

That's why I find economists maddening when they say things like "Voting isn't rational". Any action can be made rational - rationality is an axiom not an observation. When economists tell people something isn't rational, they really mean that it doesn't make sense given the kind of utility functions they are comfortable working with (and that do a reasonably good job in some economic realms).

I mention this problem because of a known result - those who study economics behave more like economists (see for example many fabulous game theory experiments involving games of cooperation). Economics is not just a descriptive science but a prescriptive set of norms and values, no better or worse than any other. So when economists say, "voting isn't rational", that may actually convince a few people not to vote. But you haven't convinced them by 'enlightening' them to their true interests but rather by actually changing their utility function (in the language of economics). And I think that might be bad for all of us.



You talk about the extremes. You mention the lack of importance of a single vote and the problem of everyone recognizing this.

What what about the middle ground?

What if all the economists, political scientists, statisticians and everyone who studied these three fields decided not to vote? People in these fields accept that it's rarely worth their time to vote, and there becomes a social pressure NOT to vote. This changes the dynamic from the case of single person not voting to that of a particular non-random group not voting.

I convinced a political scientist friend of mine that the correct position for him to take is not to not vote and tell others that he doesn't vote. Rather, he should press other people to vote, and not let them know that he doesn't vote. If he tells everyone that he doesn't vote, and why he doesn't vote, then he might convince them not to vote either. As a professor, he might convince 50-100 people a year not to vote. Along with the other professors in his department, they might convince thousands of people not to vote every election cycle.

Unless the non-voters are duplicitous about their lack of voting, the it is not merely a question how often elections are decided by a single vote.

And given Levitt's breadth of influence these days, this role model effect is quite large for him.


Therefore, I would suggest that Levitt and others be clear to separate the first order effects of not voting (i.e. the rarity of elections decided by a single vote) and the second order effects (i.e. how many other people might be convinced not to vote and the frequency of elections being within that margin).



This utility analysis ignores an important component of democratic elections: margin of victory. A politician who wins election with 60% of the vote is in a much, much stronger position than if he wins with 50.1%. A victor who wins a landslide has more "political capital," he has more latitude to take positions that invite attack, he is less likely to be seriously challenged for reelection, and so on. Margins of victory are an enormous part of the effect of elections in a democracy--elections are not simple "winner take all" affairs, despite appearances. (One wonders if FDR, for example, could have passed the New Deal if he had won election with only 50% of the vote in 1932, rather than 57%.)

There is a strong utility to be considered in casting votes for a candidate you support, regardless of whether your vote will be the deciding one, because it will affect the ability of the victor to push his platform (whether by strengthening the candidate you support or weakening the one you oppose). It will also have a big effect on whether that candidate is challenged in the next election cycle (if a no-name candidate grabs 48% of the vote against an incumbent Congressman, for example, you can be sure that the opposing party will be trying to field a strong candidate two years later).



Granted, nothing is happening in New York this year, and 3 out of my 4 local elections are uncontested. But the 4th election is for county legislator, and the challenger is a one-issue candidate (I hate those on principle) who says income taxes are too high, yet has never been to a budget meeting. Yeah, I'll go vote tonight just for the pleasure of voting against this guy.


"When economists tell people something isn't rational, they really mean that it doesn't make sense given the kind of utility functions they are comfortable working with..."

True, but people who do vote want to argue that it DOES make sense given those same utility functions.

If you want to tell me that you vote because you feel warm and fuzzy for participating in the process, that's one thing. That's what you're trying to "maximize".

But, a post like Abhinav's is exactly what they're trying to address. A non-voter does care about his taxes and wars. He just realizes that his vote is so unlikely to make a difference in his taxes, and our wars, that it doesn't make sense to travel to the polls. It's really just "democratic guilt tripping" to tell people "if you don't vote, you can't complain" or "if you don't vote, you don't care about the outcome". And, I'm surprised that people who read this blog miss that point.

That's why the linked article is so interesting (people should actually read it). When you remove the "warm and fuzzy" incentive by introducing mail-in voting, participation actually decreased.



I think voting has gone the same way as buying music: people just don't do it as often anymore because their choices stink. You're not going to pay $15 for a new album or spending time voting if you're indifferent to the available options.


I find it maddening that people are trying to argue that there is no utility to voting, simply because your one vote will never be "the deciding vote". It comes across as incredibly obtuse. Newsflash: The whole point of an election is that MANY people vote, and those MANY votes decide the winner. The more people who vote for the candidate you like, the greater the chances that candidate will win.

Pretty basic stuff.


I'm a economist and I don't vote. However, I encourage people to vote. I donate to and volunteer for political entities. But I don't vote. Why? Simply, my one vote won't affect an outcome, and no amount of warm and fuzzy feeling will change this fact.


I totally agree with Dmitry- there's not alot of wisdom in this post- you should vote "if you really feel like it" sounds like an entreaty to an 8 year old- the fact of the matter is that voting helps bind us to the community- apathy not only destroys the sense of community, but increases the liklihood of bushofascism- that is why these pseudo rationalistic viewpoints are "depressing"- staying connected and striving to make a difference are essential components of fulfillment


Talk about Arrow's Theorem of Impossibility!
And about district lines!

You are the guy who would make a great post about it!

Wilton, CT

How about we make elections like lotteries? Everyone who votes gets a shot at winning an amount of cash?


To Baltimark #26:

Lets assume that these "economists" do care about which party has a more regressive or progressive tax view, which party increases or decreases federal deficits (just to take 2 economic issues amongst which parties have clear distinctions). So lets say these people agree with the policies of party A and disagree with that or party B. Their utility would increase if party A came to power instead of B. So, this would mean they do care about the election, which in a democracy means that whichever party gets most votes wins. Given all this, these people don't vote solely saying that one person can't change the outcome, believing that it is this single person against everybody else and these people themselves can't belong to a group. The reality is completely opposite, democracy doesnt mean every man for himself, instead it encourages people to organise into groups and collectively bring about change. I find the view mentioned by the authors irrational, and a single mail-in experiment in Switzerland insufficient support for the view.

And as can be seen in other posts not all economists agree with the authors, making a blanket statement that all economists consider voting a waste of time also seems like rabble rousing.



you should examine voter turn out in the state of Oregon which went to permanent absentee balloting (everyone votes by mail) in the mid-late 1990's.

an additional caveat in Oregon is that funding / tax measures (except in certain main elections) require a double majority to pass (50% voter turnout of registered voters and 50% voting in favor of measure of votes cast)


...seems like a problem for fallacy of composition man.


There's an interesting book I recently read on democracy. It's called "The Myth of the Rational Voter: Why Democracies Choose Bad Policies" by Bryan Caplan. I highly recommend it to anyone who has even a passing interest in the subject.


I wonder if Paul Krugman will vote in the Presidential elections.


I basically agree with Levitt's utility argument, but with a very large caveat. If you know in advance that you're not going to vote, you have removed a large incentive to stay informed, and that will probably tend to reduce your overall
civic awareness. I think it would be very difficult to maintain an equivalent degree of interest in your civic environment without that motivator.

If you knew with certainty that you'd invest the same effort in studying your political surroundings, then skipping the actual voting part seems relatively trivial.

In addition, there is another effect that comes into play when you plan to vote. Your awareness of the issues and your intent becomes part of your social dialog. The network effect of having strong conviction about an issue and sharing that conviction with others is far more significant than the actual vote. But how do you achieve a similar result if you're "out of the loop" on the current voting issues?



Your embedded video player should buffer when paused, like others most do. I am in a hotel on a very slow WiFi connection and the video is unwatchable because it loads too slowly. On other sites, like youtube, you can pause and wait for it to load, and then watch the video stutter free. Pausing video on your site stops it from loading as well, such that even if you leave it paused for a minute or more it immediately starts buffering again when unpaused.


I love the knee-jerk reactions. "Ridiculous!" "How can you not want to vote!" And so on. For the same reasons you should eat your brussel sprouts, obviously: kids in Pakistan are starving for your voting rights!

Voting is no more effective than warding off the Evil Eye or having a lucky penny. It's not just that your vote won't be the deciding vote. It's that your vote won't matter at all.

If nobody in my city voted, the state's position wouldn't change. If everyone in my whole city voted republican, the outcome wouldn't change.

The outcome will not change.

Even if it could, I could, what, make the other nasty, worthless person win? :P