The Economics of Voting

This is the first part of a series that analyzes our behavior from an outsider's perspective, without any of the assumptions or dogma that we as a society have picked up over the generations.

Is voting an economically logical act?
In this post I'll attempt to quantify the cost and benefit of voting, as well as hypothesize where the stigma for the act of not voting came from.  If you are rusty on expected value, you may want to do a quick brush up so that it's easier to follow along.. I'll be waiting here until you're done.

Many people feel strongly about the necessity to vote.  Not voting is often looked down upon as the irresponsible act of neglecting your civic duty.  Surely, since many people have such strong feelings about voting, we have all asked the simple question of whether it makes sense to vote or not.  The argument against voting is simple: the cost outweighs the benefit.  The time that it requires for an individual to do her research on each politician and policy and then go to a website and fill out a form... wait what!? you can't vote online? You mean that all my money, proprietary documents, and dick pics are accessible on the internet, but it's impossible to secure a voting system online?  As I was saying, the individual must go to the store and buy stamps and scribble little marks on a piece of paper and place that in a mailbox for someone to physically carry to a center where they can efficiently aggregate.. er.. painstakingly feed each ballot into machine for digitization.  Arguably, some people get their jollies from politics, making this process less of a chore and more of pleasure.  But for the rest of us who find it a chore, should it be considered irresponsible to not vote?

It's difficult for people to admit, but for many, the reason of voting is that it makes people feel like they're a better person.  This is the reason that when you vote, just like when donating blood, you get a little sticker.  This sticker is your badge to tell others how good of a person you are.  In 1971 Richard Titmuss hypothesized that providing a monetary incentive to donate blood would decrease the likelihood of people to donate.  Since then, hundreds of studies have been performed that support this hypothesis.  The explanation of course being that people donate blood not to help others (the benefit to others remained constant), but so that they feel good about themselves.  Giving people money takes away that warm, fuzzy feeling and makes the act feel less altruistic.  Putting the pretentious fuzzies aside, let's see if there's any tangible, economic benefit to voting.

With a vote you have a small say in the outcome of the future; a future that will impact you.  It's a really cool idea.  But how small is this "small"?  In an effort to quantify this answer, let's look at the presidential election in a vacuum where we hold everything constant but your vote (after all, you only have control over your choice to vote).  In the current election process of the U.S. each candidate gets all or nothing; even if you win by a single vote you get to be the one and only president.  In an alternate universe we could imagine that many candidates can win an election and the percentage of citizens that voted for a given candidate gives them that percentage of power; this would encourage voter turnout since every vote counts.  This would also prevent the pesky split vote effect; a paradoxical scenario where the least popular candidate can win an election.  In the current system, for your vote to make any difference there must be an equal vote between two parties in absence of your vote.  What's the probability of this occurring?  Each voting distribution is not equally likely to occur (ex. it is much more likely that there 49 vs 51 outcome over 99 vs 1).  From historical presidential elections I've estimated that 42% of elections are a close call: voting differences fall within +/-5% of the 50% mark.  Assuming that each percentage difference within this close call range is equally likely (roughly true), and a standard voter turnout is over 120M, the chance of having a dead even vote is:
.42 * 1 / 12M = 0.0000035%
(likelihood to land in close call territory) * (number of possibilities we're interested in) / (number of possibilities in close call territory)
Hey, don't despair, the odds that your vote matters are still better than winning the lottery... well barely.

This gives us the probability of having impact but not the magnitude that the impact will have on our life.  How can you possibly quantify how much impact each presidential candidate will have?  There are too many subjective and indirect relationships that it would be impossible to measure.  Instead, I would rather ask an individual a simpler question: How much would you pay to have your preferred candidate win over his/her opposition?  If our gut (the intelligence organ) tells us that we would pay $100 for our candidate to win over the other, then it implies that we think we would have slightly more than $100 worth of grievances to suffer if the opponent candidate won.  Keep in mind that the dollar here is being used as a universal value unit; it doesn't literally translate into a dollar in your bank account.  On the other hand, an economist would quantify our cost of voting by measuring the time taken to vote and the amount of money we could have made during same time; this also varies by individual.  So whether you should vote is a function of both your hourly income and how much you would be willing to pay for your candidate to win.  This calculator takes the amount you would be willing to pay for your candidate to win (opposed to the opposition) and calculates how little you would have to make per hour for it to make sense to take the time to vote.  It's a simple estimated value problem.

How much would you be willing to pay for your preferred presidential candidate to win?

In other words, if it cost $1 more than you would be willing to pay, then the opponent would be guaranteed to win.

$ Population Size:

How much do you make per hour (after tax)?

$ Population Size:

Using this strategy, sadly only the unemployed can logically justify taking the time to vote, given the current voting population.  If you were following this economic model for voting, the voting population size would have to shrink to an incredibly small range (<10,000) for it to make sense again.

Some people look at voting as a community service, where those that don't vote are freeloading off of the time and effort of those that do.  If one is optimizing for the greater good of the community, the question then becomes, how much impact can I have on the community by voting compared to spending those 2 hours elsewhere.  Estimating the impact of 2 hours spent working in the community is easy to quantify; the community has already told us what our time is worth in the form of a paycheck.  More difficult to quantify is the impact that a certain candidate winning the election compared to his or her opposition will have on the community.  If we conclude that every citizen will suffer an average of $2 worth of grievances if the opposing candidate won, this implies a total impact of $319M with a probability of (.42*1/12M) which is an estimated value gain of $22.  If it took 2 hours of your time to vote (it should if you're doing your research), then you should make less than $11/hour for voting to have more impact on the community than donating your time.  Again, we see that it's dependent upon your hourly income and your estimation of per citizen grievances.

So does it make economic sense to vote?  The answer is as it is to most questions in life: it depends.   If we're optimizing for our own economic gain, the answer for any sane citizen is no.  If we're optimizing for the greater good of the community, then the answer is maybe; but we should seriously consider donating that time directly to the community for greater impact.  No motivation is any more correct than the other.  What matters more is that you can logically justify your actions, given your stated motivation.


While writing this post, it occurred to me that I was doing myself a disservice by broadcasting this idea.  Arguably those who are likely to agree with this post and be discouraged to vote, are likely to have a similar belief system as me.  The fewer like-minded people to me that are voting, the less likely I am to have an outcome that is optimized for myself.  Therefore, I should logically try to both encourage like-minded people to vote and discourage people with antithetical views from voting.  Both have equal effect and interestingly, both are actively practiced.  Campaigns to encourage voting is usually performed by targeting populaces known to support a given candidate; see canvassing and GOTV.  Perhaps we in general feel so strongly about voting because we are influenced by these pro-voting campaigns.  While non-partisan pro-voting campaigns exist, they still target a specific sub-populace (women, youth, etc).  On the other hand, voter suppression is the (often deceitful) act of discouraging voter turnout from supporters of the opposition.  In the case of the 2011 Canadian election, automated telephone messages were used to trick voters into thinking that their poll location had changed to a location often hours away from their actual voting booth.

Why do you vote?  Let us know in the comments.

Do you think it makes sense to vote?

Yes

No

I refuse to vote in this poll

Poll Maker