Why Do People Vote?

Gary Becker and Richard Posner had posts the other day on the “paradox of voting.” Even though these guys may be the world’s two smartest bloggers, I think they both really missed the boat.

The fact that millions of people vote on election day is evidence that the private sector can produce public goods. More than that, it is evidence that private production may be superior to public production — a prediction that runs counter to everything you will read in almost any public finance textbook.

How do I get from here to there? Bear with me.

Here is the paradox of voting:

Bp — C < 0,

where B is the benefit the voter receives from having his preferred candidate defeat an opponent; p is the probability that the voter’s vote will determine the election; and C is the cost of voting. By everyone’s estimation, it really doesn’t matter how large B is. The voter could be willing give up his entire wealth if it would ensure the election of his preferred candidate. Even so, the probability that the voter’s vote will be decisive in the election is so infinitesimally small that the expected benefit (the product of the two) will be close to zero. Since the cost of voting is definitely not zero, the cost must always exceed the benefit.

Put colloquially, voting must always be more trouble than it’s worth. And yet millions of people do vote. Why?

Now I’ll interrupt for a personal story. After discovering this paradox as a graduate student, I decided to do the rational thing and for many years I didn’t vote. Then on election day one year our soon-to-be Texas senator, Kay Bailey Hutchison, said to me, “I assume you have already voted?” I couldn’t find the courage to tell her the truth.

As time passed, I began to reflect more on what was going on. If candidate A is running against candidate B, the election of A is a “public good” for all who prefer her candidacy. Because it’s a public good, we all benefit from her election whether or not we do anything to help make it possible. Hence we all have an incentive to be free riders on the voting efforts of others.

Economics textbooks teach that we act in our self interest. So if it is in our interest to be a free rider, that’s how we will act. Here’s what’s missing from the textbooks. Free riders are not admired. They are more likely looked down upon. Free ridership is not viewed as an attractive feature — and everyone knows it. So few people will want to admit they are free riders or will want to be seen as free riders. It’s certainly not anything that anyone wants to boast about.

There are undoubtedly thousands of ways people can take advantage of this natural human trait. They can persuade other people not to be free riders — to make them want to contribute to an effort (even if it’s not in their narrow economic self-interest to do so) and to make them feel guilty if they do not. An election is only one such example.

In my home town, Dallas, Texas, there are some spectacular public goods erected with private money. There is a magnificent arboretum, a new park that was built over an eight lane highway that connects two parts of down town, a state-of-the-art symphony hall, a new performing arts center, the world’s most expensive outdoor art museum, a unique Asian art museum, a bridge that spans the Trinity River, a one-of-its-kind Ross Perot science museum, an entire downtown arts district — to say nothing of the new George Bush Presidential Library and public policy institute. Also, there is my own National Center for Policy Analysis. I’m just hitting the highlights here.

These “public goods” benefit the entire community. They were mainly funded by contributions of thousands of private citizens (who were of course appropriately honored for their gifts). I suspect that part of the willingness to voluntarily cooperate on projects like these is in our genes. I can certainly see why it would have evolutionary survival value.

Economic theory teaches that the private sector (for which overt coercion is not an option) will tend to under-produce public goods and over-produce private goods. But as I look around Dallas, I see no evidence that would confirm that prediction. To the contrary, I see the opposite. The projects to which I refer are not obviously too large or too small. I cannot assert that they are of optimal size. But there is no obvious evidence of under-production or over-production. Surprisingly, traditional economic theory seems incapable of explaining what is happening before our very eyes.

Public sector production of public goods is a different matter. For two decades Dallas has operated a bus and light rail transit system that has a pitiful cost/benefit ratio. (At one point I think we had the lowest bus ridership in the nation.) The Dallas public school system is a disaster — just like inner-city schools in other large cities. Welfare in Dallas encourages dependency, invites fraud and wastes money in all manner of ways. While Medicaid patients wait all day for routine care at Parkland Hospital’s emergency room, we refuse to allow those same patients to pay far less for flu shots and other basic care at private, walk-in clinics.

Why does the private sector do as well as it does in producing public goods, while the public sector does so poorly? Part of the answer is that in the private sector the beneficiaries of the public good have no direct say in the matter. Strange as it may seem, this is actually a good thing. Another good thing: There are no unwilling (taxpayer) contributors. Private sector public goods are designed and funded by private givers. Although many things may motivate them, maximizing the social value of the project is likely to satisfy them more than suboptimal production.

In the public sector, things are very different. Political rivalry here typically pits taxpayers (whose goal is lower taxes) against beneficiaries of the good (who will always prefer more of it if they don’t have to pay for it). Phil Porter and I have shown that there is no robust political system that will produce optimal quantities of public goods. Government may over-produce them or under-produce them, depending on the relative strength of these competing interests. If the special interests that produce the good can exert their influence, there will be other distortions as well.

In the face of such “government failure,” the private sector can respond in one of two ways: (1) it can produce the good privately and make it available to the whole community or (2) it can privatize the good and provide it to those willing to pay. This latter approach is called “internalizing the externalities.” Writing in The New York Times, Nicholas Kristof notes that this second approach is more common than is generally believed:

  • Hurricane Sandy revealed that the East Coast power grid can be unreliable. A private solution: by one estimate, 3% of stand-alone homes in the country worth more than $100,000 now have their own backup generators at a cost of more than $10,000.
  • In response to the inadequacy of public policing, 1% of employees in such cities as New York and Los Angeles work as private security guards.
  • In place of dilapidated public parks and playgrounds, there are gated communities with private parks and tennis courts.

What determines when the private sector will produce public goods or privatize them? There is no general theory that provides us with an answer.

Phil and I made a major step toward a political theory of public goods. It’s now time to reinvent the economic theory of private production.

Comments (29)

Trackback URL | Comments RSS Feed

  1. Vicki says:

    Very interesting post.

  2. Devon Herrick says:

    We were taught about the Paradox of Voting back in graduate school. There’s an old joke about two economists who run into each other at the polling place. They are both embarrassed to be seen by the other economist since every good economist knows it doesn’t pay to vote. They each (respectively) claim their wives forced them to vote.

    I do not believe the Paradox of Voting equation holds. Where B is the benefit, the benefit does not have to be tangible. An enhanced sense of well-being or an institutionalized believe voting is your duty is probably enough benefit to compel many people to vote. The value for p does not necessarily detract much from B. In battleground states, people may be motivated to vote by the sense of urgency. But why did people vote in Texas? Everyone knew which presidential candidate would win Texas. If the probability of changing the outcome was a major factor in whether people vote, nobody would vote unless they over-estimated their impact on the election. This limits voting to naive people or stupid people. This isn’t the case since people who play the lottery are the least likely to vote.

    Knowing my vote would make no difference one way or another, I still voted. Knowing it’s not particularly rational to vote, I still voted. My polling place is 30 miles away close to a house I own but no longer live in, but I still voted. Knowing my party would probably lose, I still voted. It was an exercise in duty and responsibility. However, I probably would not have been willing to wait an hour if the lines had been long. I needed to check on my old house so I voted while I was in the neighborhood. I probably would not have made a special trip.

  3. Louise says:

    I really enjoyed this post — it’s interesting to see how these incentives play out.

    I definitely think Dallas is a great test case — but also the DFW metroplex. There are so many small communities that have a park, a school, etc — and it’s all private. I can see why families would opt into these because the scale is smaller and they have more leverage for quality control.

  4. Cindy says:

    What an interesting read! I wonder about the voting paradox — I voted. It wasn’t terribly hard to do — I think Devon’s right that it’d be interesting to consider how my behavior would have changed if I had had to wait in a very long line or had significant difficulty getting to the polls.

    I think if I’d have to miss a significant amount of work to get to the polling place and back or would have to drive far out of my way, I might change my mind. Does this assume early voting is available?

  5. Greg Scandlen says:

    Economics too often fails to account for human emotion. One of the reasons I vote is the ritual of it — being part of a community that comes together every two years, just as my parents did and their parents did. This connection is fulfilling to me. (I wonder, btw, if early voting actually increases the number of voters. I suspect it may do the opposite by removing the ritual).

    Same with philanthropy. There is great emotional satisfaction in creating something that would never exist without your personal involvement — “I built that!”

  6. Robert A. Hall says:

    On November 7, 1972, 40 years ago, I defeated an incumbent Democrat and was elected to the Massachusetts state senate in a district that last elected a Republican in 1938. The last town to come in, Shirley, at 6:30 am, put me ahead by nine, out of 60,000+ cast—I had trailed all night. It was my first full time job after graduating from U-Mass in June. I held the seat for five terms and retired undefeated. I always vote. I will link to this from my Old Jarhead blog. (www.tartanmarine.blogspot.com)

    Robert A. Hall
    USMC 1964-68
    USMCR, 1977-83
    Massachusetts Senate, 1973-83
    Author: The Coming Collapse of the American Republic
    All royalties go to help wounded veterans
    For a free PDF of my 80-page book, write tartanmarine(at)gmail.com

  7. Jordan says:

    The Paradox of voting doesn’t have anything to do with the actual cost. “p” is always so small that B or C means little.

    Private production of public goods in Dallas should tell us that “p” is misleading. People don’t vote thinking that their vote will be the deciding factor, but rather that contributing a vote is an affirmation of democratic ideals they value. Much like donating to help build a new art museum or public park is an expression of other values.

  8. “While Medicaid patients wait all day for routine care at Parkland Hospital’s emergency room, we refuse to allow those same patients to pay far less for flu shots and other basic care at private, walk-in clinics.”

    We don’t really refuse to allow them to pay do we? (Like Canada does in some cases.) We just refuse to reimburse them for their payments.

  9. Uwe Reinhardt says:

    The paradox of voting merely illustrates how limited the cerebral processes of economists are relative to those of full-fledged human beings.

    As Greg points out, citizens reap psychic benefits beyond those economists can visualize in their limited imagination. These psychic benefits come in many different forms.

    I do not even think of the paradox-of-voting equation when I vote, which I do regularly. Rather, I think back to the US military cemeteries in Europe to which we took our children when they were little. These warriors rest there because they tried to give enslaved peoples the right to vote and to live in a democracy. I vote to honor them.

    Other voters will have different motives; but somehow this works.

  10. Uwe Reinhardt says:

    On another theme, John, I take it you would rather have Halliburton produce the output now produced by the U.S. Marine Corps — whose members are public employees using publicly owned equipment.

    As Yogi Berra or Voltaire or whoever put it, “Chacun a son gout”.

  11. Buster says:

    I think Uwe and Greg are correct. There’s more to voting than a simple cost/benefit calculation. That said, I find it to be rather amusing that once U.S. foreign policy rids a country of a despot, people in the State Department naively assume people in Serbia, Bosnia, Iraq and Afghanistan suddenly have this urge to go out and vote. Americans in general assume the whole world wants to be free (and vote). In reality, most people around the world are far more concerned with economic well-being than whether they can vote for the politician of their choice. I vote not because I actually care who governs me. I vote because I would like to preserve what is left of the economic freedom we still have in the U.S.

  12. Uwe Reinhardt says:

    There is something to Buster’s theory, which can explain why quite a few dictators have remained popular among all but a smallish elite, simply by catering to what the Roman poet Juvenal lamented as the public’s yearning for “panem et circense” (bread and circus games.

    But many originally benevolent dictators eventually morph into tyrants, and then there is yearning for freedom all around, as we have seen in the Arab uprising, or the uprisings that led to the fall of the Berlin Wall.

    The interesting thing is that American voter turnout is so consistently lower than that one sees in the European countries. As that great American Bill Maher has observed in his classic rant on Americans and the French (see Youtube.com): “The French have a strange attitude towards voting: they vote.”

  13. Walt Davis says:

    Why would anyone hold up the French voter as an acceptable model? Look at the results.

    John, I read the Kristoff article that you referenced. I was alarmed, and somewhat shaken, at the content of the comments that were submitted for it. The task ahead seems unachievable.

  14. Uwe Reinhardt says:

    To Walt Davis:

    LOL. Your remark reminds me of this study:

    Volume 367:1588-1605 June 5, 2003 Number 18


    Alfred E. Neuman, Department of Cross-Cultural Studies, Harvard University


    Background: Cross-cultural studies on happiness can help identify opportunities for enhancing human welfare. We compared responses to powerful external sensory stimuli in representative random samples of Europeans and Americans.

    Methods: We placed nationwide random sample of 2,253 Americans and of 2,138 Europeans on piles of cow manure. The manure was gathered from random samples of in-country farms. Sampled individuals were left on these piles of manure for 45 minutes. A survey team of anthropologists and psychologists then retrieved from them their reactions to this experience.

    Results: A sizeable majority (78.3%) of the European respondents remarked that their manure stank. By contrast the majority of American respondents (93.2%) opined that American manure is the best in the world. A plurality of American respondents (76.3%), however, agreed that European manure stank. Close to 17% of the American respondents had visited Europe in the past 10 year, and as many as 23% had visited Europe at one point in their lives.

    For a look into the mirror, see


  15. seyyed says:

    i wonder if election analysis and predictions discourage anyone from voting after seeing something like a 99 percent chance that their candidate was going to win a certain district or state.

  16. David R. Henderson says:

    Nice post, John. It motivated me to write my own post in which I talk about how I went through a similar evolution about voting. Here’s the link:

  17. Charlotte Spencer says:

    My first thought when reading this is how many people I heard during the past election day (people at the grocery store, gas station, on the radio, and even at work) saying how they felt about voting. It was to my surprise that most of them did not even try to go vote because they felt like their vote wasn’t really going to make a difference, and going out of their way, missing work, and spending time driving and waiting in line to go vote was not worth it. My question is, what if not 10 or 20 people feel that way, but actually hundreds and hundreds of people? That would certainly make a difference, and the reality is that not 10 or 20 people feel that way about voting. The reality is that hundreds of them across the nation think that their vote doesn’t matter, and the truth is that when you add all of these numbers up..then that’s when numbers do matter, and every single vote matters.
    I’m new to this country, and I was very impressed by the way that the election system works here. Coming from a country where there is only 1 day you are allowed to vote (there is no prior voting or anything like that), and the process is extremely exhausting for everyone involved (waiting lines, traffic every where you go because everyone is trying to go vote, chaos on the streets and at the voting centers, etc), you really appreciate the American system. They really make it convenient and pleasant for everyone that actually wants to vote to go ahead and do it at their convenience to avoid any hassle or altercation. I think some people just really need to experience the voting experience in foreign countries where you REALLY don’t have a choice and only get to vote whenever it is determined by the authorities. I think it is only then that those with that mindset of “my vote doesn’t matter” will realize how important it really is to do it if you can. Great post!

  18. Uwe Reinhardt says:

    I wonder why we don’t vote on Sundays, so that most people do not have to take time off from work or stand in 3 hour-long queues on Tuesday evenings, often in the cold?

  19. Linda Gorman says:

    In well run states there is absolutely no reason why a reasonably well organized individual would have to wait 3 hours to vote, ever.

    A lot of local elections are decided by a few votes. So are some statewide ones. Like the 312 votes in Minnesota that elected Al Franken and provided the 60th vote for ObamaCare. John Fund says that Minnesota Majority has identified 1100 felons who voted illegally in that election and that 191 have so far been convicted. More cases are in the pipeline.

    It’s more than just being the deciding vote. As Hugh Hewitt says, “if its not close, they can’t cheat.” That applies to every party and individual in the process.

  20. Uwe Reinhardt says:

    Ms. Gorman, I recommend you read former Republican White House staffer David Frum’s column on our voting system at


  21. Studebaker says:

    I’m interested in exploring the possibility of a poll tax of sorts, where citizens are allocated votes as a function of taxes paid minus government benefits received. The more taxes you pay, the more weight your vote counts. People who work for the county, for instance, would be prohibited from voting in special elections where bonds or county tax are being considered.

  22. Uwe Reinhardt says:

    To Studebaker:

    Cool idea. Because defense spending is always implicitly on the ballot, soldiers, sailors and Marines presumably should not vote either. They may be biased in favor of the party that wants to up defense spending.

    Maybe that is why in older democracies of yore, only the landed gentry could vote.

    Questions: should bankers be allowed to vote?

  23. Linda Gorman says:

    Re Frum: Not impressed. He’s just arguing that a decentralized system is inefficient without providing any evidence for the claim.

  24. Uwe Reinhardt says:

    Well, Ms. Gorman, you writer:

    “In well run states there is absolutely no reason why a reasonably well organized individual would have to wait 3 hours to vote, ever.”

    It’s hard to disagree with this. It’s like saying that something that is perfect should be perfect.

    The empirical question, on which you shed no light, is how many well run system there are.

    If you Google something like “long lines for voting in November elections,” you will find many articles complaining about long lines, poll stations running out of ballots, etc.

  25. William Hallman says:

    You voted because a cheerleader asked you to?

  26. Linda Gorman says:

    Those of us in the states that have updated voting systems have a broad choice in how we vote. We don’t have to go to the poll on election day. People can sign up for permanent mail ballots, request mail ballots or vote early at select polling places. This is why I said that a reasonably organized person in a well run state didn’t have to stand in the lines.

    That Google includes reports on lines and so forth doesn’t invalidate my point. In my county there was one polling station at which lines were deliberately created in order to have a media circus. The rest had no waits, wait times were posted on the internet, and voters could go to any of them.

  27. Uwe Reinhardt says:

    To William Hallman:

    I am not sure to whom you are addressing the question: “You voted because a cheerleader asked you to?”

    But if that cheerleader cheered for the right team and looked intelligent enough, she could (a) move me to vote, at her request and (b) even vote for her preferred candidate (although I would probably lie about that).

  28. Al says:

    I guess I might disagree with many. I think voting is a privilege that does not require everyone to vote. I don’t think voting should be made easier just to increase the numbers that vote. In fact I disagree with absentee ballots unless necessary and a prolongation of the voting process though more than one day in a row would be OK. I want to see all voters registered with a picture ID that is inserted into a machine that checks the appropriateness of the vote so that we are assured that each vote is legal. Then I want a hard copy of the actual vote made that was done in secret. If someone is not interested enough to vote then their vote need not count.

  29. Uwe Reinhardt says:

    I agree with Al on the picture ID. I have never understood the opposition to this idea. What is so hard about getting such a card. Voting is a privilege.