LESSWRONG
LW

Comment Permalink

Thanks for the links. They seem to mostly be saying: the "pay off" for being the swing vote is gigantic, changing everyone's life, so even though the chance of being that vote is infinitesimal it's rational to go for the tiny chance of making a huge difference.

I'm sure this is valid reasoning, but it's disappointing to me if this is the whole story. It's like voting as lottery, that your vote essentially never matters except when it has this giant impact.

I think there is mapping problem here as well. Just as you can't map your vote onto one of the excess votes in a normal election, you can't map your vote onto that one winning vote in a close election. In each case it's a game of probabilities and fractional contributions only. But I can't sort it all out.

See in context

3 Why should you vote?

by kmeme

29th Oct 2010

1 min read

3

For many years I've been interested in the "paradox" that your vote tends to never alter the outcome of an election, yet the outcome is in fact determined by the votes. I wrote a blog post about this and tried to explain it in terms of emergence, we as voters are just feeling what it's like to be just a tiny part of a much bigger system.

Then I tried to explain that "voter turnout" is in fact one of the most important metrics for an election, it determines the legitimacy and stability of the process. So therefore even though your vote won't determine the winner, it will contribute to voter turnout and thus is productive and useful.

http://www.kmeme.com/2010/10/why-you-should-vote.html

However I don't find my argument all that compelling, because even voter turnout is going to be approximately the same whether you vote or not.

In the post I bring up littering as something else where your tiny contribution adds up to be bigger result. I personally would never litter on purpose, yet I often skip voting because it seems like it doesn't make a difference. Is voting rational? How do you justify voting or not voting? My post was non-partisan so I'm soliciting non-partisan comments, trying to focus on the theory behind voting in general.

Personal Blog

3

New Comment

24 comments, sorted by

top scoring

Click to highlight new comments since: Today at 11:17 AM

[-]NihilCredo14y80

I get a strong feeling that you wanted from the start to find a result in favour of voting, and have simply been looking for a justification. To employ one of the LW mantras - if (under a given value system) it is rational to stay at home, would you truly want to believe that it is rational to stay at home?

Anyway, voting can be modellised fairly well as a tragedy of the commons; the resource being shared is "degree to which politicians are and feel controlled by the electorate", and each voter can eat a little of it in exchange for some free time. Most of the arguments about TotC also apply here; however, the nature voting does also create some more sophisticated scenarios, such as an organised minority of exceptionally disciplined voters.

[-]knb14y50

I don't vote. Part of the reason is that I have little confidence in my ability to judge which candidate will be better (because I don't have traditional left-right views. ) I also have little reason to put effort into determining who is more likely to be better since voting has so little benefit.

I don't think voting is immoral, though, and I can see that there are some situations where voting is the smart thing to do. For example, if you want to demonstrate your affiliation with a group, or if you just like the act of civic participation.

[-]Manfred14y30

Humans are bad at small numbers. Your vote does indeed matter. But it only matters 10^-8 of a presidency. With littering you're probably anchored to 1 piece of litter, so 1 matters. But in voting you're probably anchored to 1 presidency, rather than, say, 1 dollar. Noting, of course, that on the order of 10^10 dollars get spent per presidential race, so your vote is about $100.

[-]kmeme14y00

In a typical election with a greater-than-one vote differential, it doesn't seem like your vote matters at all to the outcome. In the specific sense that if you had not voted, the outcome would have been identical. So I guess that is something to resolve, do you deserve fractional "credit" for the win, even if the win differential was by millions of votes?

[-]Manfred14y40

Assuming that in a perfect election, some set of people do "deserve fractional credit," it's easy to show that, since the labels on different voters can be changed without affecting anything, everyone deserves fractional credit. Similarly, if you assume that some set of people "don't deserve credit," nobody deserves credit. So there really is a dichotomy: either everyone (who voted identically) deserves credit or nobody does.

You seem to be approaching it the second way - starting with the thought that there is some set of people whose votes were definitely extra.

Meanwhile my train of thought is "the election was won, therefore it's wrong to say that nobody deserves any credit, so it must be the other half of the dichotomy and everyone deserves credit."

Perhaps my extra assumption that nobody deserving credit is bad would be more appealing if instead of "everybody/nobody deserves credit" we use "everybody's/nobody's votes counted." If nobody's votes counted, the election wouldn't have had an outcome. So everyone's votes counted. Then I assume that if your vote counts, you "deserve credit," and bada bing.

So yes, I think that your vote can count even if, in retrospect, it changed nothing, so long as votes are interchangeable and anonymous.

[-]kmeme14y00

I think that your vote can count even if, in retrospect, it changed nothing, so long as votes are interchangeable and anonymous.

I think that's the crux of the issue. My take was to assign a mapping between people and votes such that your vote was in the "excess" portion and thus didn't matter. But just because such a mapping exists doesn't mean it is fair or valid to assign it. Instead I imagine it's a statistics problem where all mappings are possible, which leaves you with a non-zero but tiny "contribution" to electing the winner.

And if you voted for the loser? Then I think the contribution to voter turnout mentioned in the post comes into play. Again a only very tiny amount, but non-zero.

Then finally social issues likes signaling status and desire to belong to a group probably are pretty big factors, maybe bigger than the above "real" factors.

In the end I think it's possible to justify voting or not voting depending on your values, particularly how you value your time relative to these fuzzier benefits.

[-]NancyLebovitz14y10

To a large extent, I vote to avoid regret-- if someone I detest gets in, I want to feel that at least I did a little something to not let that happen.

[-][anonymous]14y20

If you basically like your country's political system as it is, then voting helps keep it that way. Having large numbers of people at polling places at election time is a show of trust and civic participation. It means that we're still invested in our institutions and political traditions. Even if your vote doesn't count much to sway outcomes, your presence at your neighborhood polling place is visible to your neighbors and does increase the sense of community participation.

Of course there are compelling arguments that "civic religion" is a bad thing, but if you think it's a good thing, then showing up to vote does help.

[-]knb14y00

Having large numbers of people at polling places at election time is a show of trust and civic participation.

I suspect that the opposite is true. It seems to me that people would be less likely to vote if they trust their fellow citizens to make the right choice and believe that the political system is fine. Turnout was at its highest in 40 years in 2008, in spite of the fact that most people agree there was a profound amount of political distrust and acrimony in that election.

[-]humpolec14y20

Isn't some form of Twin Prisoner's Dilemma here? Not in the payoffs, but in the fact you can assume your decision (to vote or not) is correlated to some degree with others' decision (which it should be if you, and some of them, make that decision rationally).

[-]steven046114y20

To make things worse, if there's a multiverse, if you're correlated with some people in the same universe as you then you're also correlated with a humongous number of people in different universes (far more humongous than the number of actual copies of you).

[-]ata14y30

I would love to see a get-out-the-vote campaign based on telling potential voters that the fate of the multiverse depends on them.

[-]Will_Newsome14y20

When you're not being awesome, your lack of awesome is correlated with all of the lack of awesome in the entire multiverse.

[-]nshepperd14y10

To my mind this is the most compelling reason to vote. If you're rational and you want more rational people to vote, then you should vote, because then they will too (assuming a reasonable number of rational people have similar relevant information to your own).

[-]Jayson_Virissimo14y10

Gordon Tullock explains why he doesn't vote in this video.

[-]Matt_Simpson14y10

See this and this for a discussion of when it's rational to vote.

edit: and this

[-]SeventhNadir14y00

I think one of the arguments declaring that voting is rational is a bit suspect.

But here's the good news. If your vote is decisive, it will make a difference for 300 million people!

In the rather unlikely event that your vote decisive, this is true enough (for US voters anyway). The error he makes though is the assumption that your decisive vote will always create a positive change. If you're going to take the credit for the right decisive vote, you have to take the blame for the wrong decisive vote.

Some people might go on to argue that it's the voters job to make an informed choice, but good luck with that. Even the most informed voter is going to be working with sketchy information, politicians don't always (ever?) deliver everything they promise.

Worse still, you can't even tell what positive/negative real life events were actually dependent on who is in office even in theory, let alone in practice.

I'm not going to say whether voting behaviour is rational or not because I'm not sure everyone is using the same definition as me, but I will say that I think people seriously overestimate the power of their vote.

[-]kmeme14y00

I'm sure this is valid reasoning, but it's disappointing to me if this is the whole story. It's like voting as lottery, that your vote essentially never matters except when it has this giant impact.

[-]DanielLC14y10

Littering makes a difference. It's nearly impossible for one person to notice the effect of one person littering, but it makes a difference to a large number of people.

Voting has a tiny chance of making a huge difference. The probability of the election hinging on one vote is minuscule, but if it happens, you change the president for four years, and make a significant mark on the country. This is orders of magnitude more than the difference you make normally.

Note that that isn't necessarily worth it. I don't know if it is.

[-]kmeme14y00

In my post I suggested there are two separate motivations for voting:

1) Picking the winner. Essentially no one does this in big elections, but yes there is that tiny chance. I didn't go into this motivation. Thinking about it now I suspect using a lottery-like mentality a lot of people do vote for this reason: they just might be the one.

2) Adding to voter turnout, thus making the election legitimate. Everyone who votes does this, but it's only by a tiny amount. This I would equate to "not littering" in that you are unambiguously helping but only incrementally. This I think is in fact the stronger reason for voting, but still your contribution is very tiny.

In the end I'm down to "civic duty" and "it's the right thing to do" and stuff like that as far as the best reason to vote. Maybe the lesson is it's good large chunk of people are NOT rational, because I do maintain as turnout goes down, the results get worse, as far as what people "really" want.

Now I didn't mention social pressure either. There was a swiss study where voter turnout didn't go up when vote-by-mail was made an option. The supposition being people no longer felt the pressure to make an appearance at the polls to be a good citizen, they could just surreptitiously not vote in the privacy of their own home.

http://www.mitpressjournals.org/doi/abs/10.1162/jeea_a_00015?journalCode=jeea

[-]Vladimir_Nesov14y00

For many years I've been interested in the "paradox" that your vote tends to never alter the outcome of an election [...]

However I don't find my argument all that compelling, because even voter turnout is going to be approximately the same whether you vote or not.

I don't think these are true, in the relevant sense of acausal influence of your decision on the outcome (though I don't have a formal understanding of how that works).

[-]Vladimir_M14y-10

The real crux of the issue is that people vote mainly for signaling value. For nearly all people, the primary motivation for their political beliefs is to signal status, respectability, and/or adherence to the groups they identify with. (The latter can mean adherence to some particular faction, sect, ideology, ethnic group, etc., but also to the whole country and its abstract ideals in general.) Accordingly, the motivation for voting is to enable a symbolic expression of such beliefs that reinforces and signals them, much like a religious ritual.

So, to answer your question realistically: if the reward in good feelings and (perhaps) status signaling among some group of people you care about is high enough to justify the effort, then it is rational for you to vote. In contrast, the attempts to demonstrate that one should vote because of some deep moral principles or probabilistic considerations are pure rationalization.

[-]kmeme14y10

I think you are right that good feelings and status are a big part of it. Why do people endure all manor of inconvenience to be a part of any big event, a movie opening or concert or rally? A lot of bragging rights to have been a part of something, rather than just watched it on TV.

Still I wonder if that is the whole explanation. The system needs X% of voters out there to be viable, but it has no real carrot to attract people to vote. So then voters just assign a value to voting, and show up in relatively large numbers on their own. And it all works out? It's a very clever arrangement if that's how it works.

[-]MoreOn14y-20

Not voting is like littering. You can drop that little transparent candy wrapper on the ground, and you increase your utility while not lowering anyone else's. It's only when that behavior accumulates, then it's a problem.

Moderation Log