3 ms·

There exists a paradox in social choice theory (game theory applied to elections), the paradox of voting -- this comes out of Anthony Downs around 1950. We can

There exists a paradox in social choice theory (game theory applied to elections), the paradox of voting -- this comes out of Anthony Downs around 1950. We can roughly consider that the utility value of voting is equal to a formula that looks like: E[voting] = pV - c, where c is the cost of voting (time cost, information cost, etc.), V is the value of your preferred candidate winning over other candidates, and p is your probability of being the pivotal voter. By pivotal voter, I mean that if everyone other than you cast their vote first, and then you cast your vote, the question is would your vote have a capability to change the outcome. The typical view is that c is small but non-zero, V is reasonably large and by definition positive, but p is infinitesimal. And so, no one should bother voting. And yet, people do vote, an off-equilibrium behavior. And although it could be the case that someone's subjective belief in p is much higher than the true p is, this seems an unsatisfactory explanation because even stupid idiots know that their individual vote is unlikely to make the difference. There are two major lines of rebuttal to the paradox. The first is that individual equilibrium models are not appropriate for assessing these kinds of questions. The second line of rebuttal is that there must exist other terms in the equation. The most promising of these is that the civic religion we have built around voting (social pressure to vote, personal moral imperative to "do one's part in society", the dumb stickers, how good it feels to stand up for what you believe, etc.) creates a "consumption value" of voting. So maybe the equation is E[voting] = pV - c + d where it is assumed that d >= c for many people, who vote. I mention all of this because based on your post, p = 0, V = Infinity, c = high, and d = 0. If you are looking for a call to action, all you need to do is consider the great many benefits of changing your d term. Finally, I think it weakens the approachability of your third paragraph argument (why bother, states should do it instead) that it comes after your second argument (environmentalists don't even help the environment) or your first argument (vegans annoy me because I can't eat gluten, why aren't people listening to me).

fao_ 7y ago
> Finally, I think it weakens the approachability of your third paragraph argument (why bother, states should do it instead) that it comes after your second argument (environmentalists don't even help the environment) or your first argument (vegans annoy me because I can't eat gluten, why aren't people listening to me). Yes, this is true. I would dispute your distillation of my first paragraph. It is not vegans or vegetarians that annoy me, it is the fact that groups and organizations are choosing to catering first and foremost for people who have fundamentally chosen not to eat something, rather than those who cannot eat something because of a physical disability. It means the interests of rich, abled-bodied people are once again being put above the interests of disabled people (Who are more likely to be poor, because they have invested a significant amount of money in coping with their disability -- as a celiac, for example, bread is 3x the price of normal bread, for a loaf about half the size (this is true for most other grain-based things)).
einpoklum 7y ago
> The utility value of voting is ... E[voting] = pV - c, where c is the cost of voting (time cost, information cost, etc.), V is the value of your preferred candidate winning over other candidates That patently ridiculous. There are varigated and complex motivations for voting in general and for voting for a certain candidate or party in particular. It has to do with your sense of identity; self-fulfillment vs passivity; commitment to peers, family, friends; sense of gratitude for past actions by the candidate or the party/movement; and these are all factors which are almost orthogonal to whether the candidate you vote for wins.
- notafraudster 7y ago
  This is ultimately what goes in the d term in the next paragraph, yes. I agree that in hindsight people getting tripped up in the initial model before the criticisms elaborated on it were ignoring something that is obvious to everyone who isn't a game theory dweeb. My point in introducing the paradox was to introduce the solution to my comment's parent.

fao_ 7y ago

> Finally, I think it weakens the approachability of your third paragraph argument (why bother, states should do it instead) that it comes after your second argument (environmentalists don't even help the environment) or your first argument (vegans annoy me because I can't eat gluten, why aren't people listening to me). Yes, this is true. I would dispute your distillation of my first paragraph. It is not vegans or vegetarians that annoy me, it is the fact that groups and organizations are choosing to catering first and foremost for people who have fundamentally chosen not to eat something, rather than those who cannot eat something because of a physical disability. It means the interests of rich, abled-bodied people are once again being put above the interests of disabled people (Who are more likely to be poor, because they have invested a significant amount of money in coping with their disability -- as a celiac, for example, bread is 3x the price of normal bread, for a loaf about half the size (this is true for most other grain-based things)).

einpoklum 7y ago

> The utility value of voting is ... E[voting] = pV - c, where c is the cost of voting (time cost, information cost, etc.), V is the value of your preferred candidate winning over other candidates That patently ridiculous. There are varigated and complex motivations for voting in general and for voting for a certain candidate or party in particular. It has to do with your sense of identity; self-fulfillment vs passivity; commitment to peers, family, friends; sense of gratitude for past actions by the candidate or the party/movement; and these are all factors which are almost orthogonal to whether the candidate you vote for wins.

notafraudster 7y ago
This is ultimately what goes in the d term in the next paragraph, yes. I agree that in hindsight people getting tripped up in the initial model before the criticisms elaborated on it were ignoring something that is obvious to everyone who isn't a game theory dweeb. My point in introducing the paradox was to introduce the solution to my comment's parent.