Voter turnout's already too high - leave it alone!
Mark Thoma links (but disagrees with) to a an article by some Australian do-gooder who thinks that people should be forced to vote. Gabriel also comments. I've already left some comments there but here I wanted to say a bit more about the argument that voting is a public good subject to the free-rider problem. So first, let us just note that the author of the article doesn't really know what a free-rider problem is and equates with "when other people don't do what I want them to do".
But anyway. Is higher voter turnout better? Aside from some amorphous and ill defined appeals to 'good citizenship' I really don't see why. Voting is supposed to be a mechanism which takes individual preferences and somehow aggregates them into a social outcome. If the outcome of the voting process is the same whether 20% of those eligible vote, or 60% do so, what's the difference? Most of these arguments are just a smoke screen for the self-interested motives that knows that a different % of voters (this goes both ways) would deliver a different outcome. But this means that what people care is that the outcome resemble their preferences, not how many people vote per se.
Suppose it's gonna be Obama vs. Giuliani. And suppose that 20 million people prefer Giuliani and 20 million + 1 prefer Obama. If 40 million people stayed at home on election day and only that one guy who prefers Obama shows up, what's the problem? The system worked even with .00000256 % voter turn out. And if there is some, even very small cost, to people of voting then there are potentially huge savings from having those folks stay at home.
Or think about it another way. Suppose that choice of our glorious leader is not made in the voting booth, but instead, the government sends out pollsters (who for the sake of argument, are 100% reliable and trustworthy) to ask people who they prefer. "You want Barack or Rudy?" This is exactly the same as voting except from you having to go to the government, the government comes to you. And suppose that there's some small cost of sending the pollster to each individual citizen.
How many people should be polled? Obviously sending a pollster to every single citizen would be very super expensive (but somehow the voting-fetishists don't mind if that cost would be born by individuals). Welp, those ol' buddies the Central Limit Theorem and the Law of Large Numbers tell us you don't need a large sample (*) to be able to estimate the relevant voting shares and preferences. Something like 10% would work, 15% would be golden (and of course that's how the stations are able to fairly accurately predict winners. If they had bigger budgets and could actually get 10% to 15% they'd get it spot on almost every time). So in this case, polling only 10% or 15% of the eligible citizens would work just fine and would save the government oodles of money relative to asking, say 60% or 70% of the population, and accuracy would increase only a tad (I'm gonna keep this a post without any serious maths).
In fact polling 64% of the eligible voters seem ridiculous from that perspective. And that's what the voter turnout was in the 2004 election.
At this point the counter argument is usually something about how we're all members of the polity and voting gives us a sense of belonging. Ok, fine. That's probably why you get that crazy 64% number rather than something much more reasonable and lower. But there's no externality here. If it gives you a sense of belonging you go vote. If it doesn't you stay at home. End of story. And it's not like you can force people to 'feel' like they belong just by making them do something they don't want to.
And like I said before. Only people who own at least two heads of cattle should have the right to vote anyway.
(*) Of course this requires the sample be random. In a sense the sample of those who actually vote from the universe of eligible voters is not random. It's self selected. It's people who have strong feelings (and sometimes even, yes, better information) about the outcome. But that's the way it should be. Do you want people flipping coins in the voting booth?
(So I tend to think that having a small cost of voting is a good thing as it discourages those who only care marginally (in the non-economic sense of the word). Part of the problem with voting as a social-choice mechanism is exactly that it's one person one vote - it doesn't account for the intensity of preferences. If I only care a little I have the same weight in the process as someone who cares a lot. Having a small cost (which could be just the opportunity cost of time it takes to vote) dampens this problem as it eliminates some borderline indifferent voters.
But there's another problem. 'Caring a lot' about an a outcome, or having very intense preferences about it are not always good either. And some of the people with very intense preferences should also be somehow taxed on their voting. For an obvious example take that crazy guy in Iowa who asked Edwards about OJ. Obviously he feels very strongly about something. He's also a total nut case who probably won't be dissuaded from voting his nut case ideas in the booth by a small cost. This is of course a typical problem in a democracy but here overall I think it just falls into that Churchill quote about democracy being the worst possible system except for all the others, and I don't really see an objective way of solving this problem)
But anyway. Is higher voter turnout better? Aside from some amorphous and ill defined appeals to 'good citizenship' I really don't see why. Voting is supposed to be a mechanism which takes individual preferences and somehow aggregates them into a social outcome. If the outcome of the voting process is the same whether 20% of those eligible vote, or 60% do so, what's the difference? Most of these arguments are just a smoke screen for the self-interested motives that knows that a different % of voters (this goes both ways) would deliver a different outcome. But this means that what people care is that the outcome resemble their preferences, not how many people vote per se.
Suppose it's gonna be Obama vs. Giuliani. And suppose that 20 million people prefer Giuliani and 20 million + 1 prefer Obama. If 40 million people stayed at home on election day and only that one guy who prefers Obama shows up, what's the problem? The system worked even with .00000256 % voter turn out. And if there is some, even very small cost, to people of voting then there are potentially huge savings from having those folks stay at home.
Or think about it another way. Suppose that choice of our glorious leader is not made in the voting booth, but instead, the government sends out pollsters (who for the sake of argument, are 100% reliable and trustworthy) to ask people who they prefer. "You want Barack or Rudy?" This is exactly the same as voting except from you having to go to the government, the government comes to you. And suppose that there's some small cost of sending the pollster to each individual citizen.
How many people should be polled? Obviously sending a pollster to every single citizen would be very super expensive (but somehow the voting-fetishists don't mind if that cost would be born by individuals). Welp, those ol' buddies the Central Limit Theorem and the Law of Large Numbers tell us you don't need a large sample (*) to be able to estimate the relevant voting shares and preferences. Something like 10% would work, 15% would be golden (and of course that's how the stations are able to fairly accurately predict winners. If they had bigger budgets and could actually get 10% to 15% they'd get it spot on almost every time). So in this case, polling only 10% or 15% of the eligible citizens would work just fine and would save the government oodles of money relative to asking, say 60% or 70% of the population, and accuracy would increase only a tad (I'm gonna keep this a post without any serious maths).
In fact polling 64% of the eligible voters seem ridiculous from that perspective. And that's what the voter turnout was in the 2004 election.
At this point the counter argument is usually something about how we're all members of the polity and voting gives us a sense of belonging. Ok, fine. That's probably why you get that crazy 64% number rather than something much more reasonable and lower. But there's no externality here. If it gives you a sense of belonging you go vote. If it doesn't you stay at home. End of story. And it's not like you can force people to 'feel' like they belong just by making them do something they don't want to.
And like I said before. Only people who own at least two heads of cattle should have the right to vote anyway.
(*) Of course this requires the sample be random. In a sense the sample of those who actually vote from the universe of eligible voters is not random. It's self selected. It's people who have strong feelings (and sometimes even, yes, better information) about the outcome. But that's the way it should be. Do you want people flipping coins in the voting booth?
(So I tend to think that having a small cost of voting is a good thing as it discourages those who only care marginally (in the non-economic sense of the word). Part of the problem with voting as a social-choice mechanism is exactly that it's one person one vote - it doesn't account for the intensity of preferences. If I only care a little I have the same weight in the process as someone who cares a lot. Having a small cost (which could be just the opportunity cost of time it takes to vote) dampens this problem as it eliminates some borderline indifferent voters.
But there's another problem. 'Caring a lot' about an a outcome, or having very intense preferences about it are not always good either. And some of the people with very intense preferences should also be somehow taxed on their voting. For an obvious example take that crazy guy in Iowa who asked Edwards about OJ. Obviously he feels very strongly about something. He's also a total nut case who probably won't be dissuaded from voting his nut case ideas in the booth by a small cost. This is of course a typical problem in a democracy but here overall I think it just falls into that Churchill quote about democracy being the worst possible system except for all the others, and I don't really see an objective way of solving this problem)
posted by YouNotSneaky! at 2:06 PM
21 Comments:
In a perfect world, you would be absolutely right, but... the costs of being able to cast a vote have, historically, never been random. Your 'two heads of cattle' comment was funny precisely because it sums up part of that history.
The historical patterns that determined where and when the right to vote (or where balloting was guaranteed to be secret) followed a very determinate pattern and logic.
On this topic see for instance recent work by Engerman and Sokoloff, Acemoglu and Robinson, Baland and Robinson, or -- on US voting rights -- Besley, Persson and Strum.)
Most of the countries of the world that adopted compulsory balloting did so for the same reason that groups of people in those countries earlier felt that it was necessary to pass measures (like the US voting rights act) to take active measures to guarantee that all citizens would have an equal chance to make it to the polling booths and vote their conscience in countries that nominally guaranteed that right but where in practice their was abuse.
When voting is 'voluntary' you do tend to want to give up and stay home when the 'poll officer' tells you your papers 'just aren't in order'. But when everyone in the neighborhood is standing in the same line its the crooked poll officer who feels the pressure to change. So, in some ways this is a free-rider problem and raising the cost of exit may help solve it.
Now there may well be more efficient ways to level access to the voting booth, and this kind of argument might apply more to Paraguay or the Phillipines (countries where voting is now compulsory) or Alabama in the 1960s, than it does to most of the US today, the point is just that political economy considerations often make it somewhat easy to predict the outcome of making voting compulsory.
[Search wikipedia for 'compulsory voting' to see the list of countries.]
The best suggestions I saw in the SEIU's "Since Sliced Bread" contest was paying people $35 to vote. That would get the disenfranchised-feeling poor vote out for once. It would also accomplish what mandatory voting does without hitting anyone over the head with a fine.
Next thing you want to try -- (don't recall if it was one of my contest ideas) -- is REVERSE campaign matching funds: matching any contribution to any candidate from taxpayer funds (buy our government back). This will keep incumbent congresspersons from spending half their time raising money and the other half satisfying those who gave -- incumbents will run from contributions.
Suppose it's gonna be Obama vs. Giuliani. And suppose that 20 million people prefer Giuliani and 20 million + 1 prefer Obama. If 40 million people stayed at home on election day and only that one guy who prefers Obama shows up, what's the problem?
representation or leverage or something like that.
it ain't about who wins or looses but also by how much. after all, nobody would be pissed about rigging in Russia, Nigeria, Gabon or Singapore if it was just about the name of the winner(s).
but all in all, yeah, the big deal is the assumptions made about the preference of the non-voters. didn't anybody learn from the last french election ?
OMG, this must be the first time I disagree with you close to 100%!
Let's restrict the topic to 2-alternatives, 50%+1 voting, to avoing any Arrow (Im)Possibility Theorem issues...
I think you're misusing the notion of the median voter in that you want to use ex ante the information you get ex post. Sure, if we know everyone's preferences, no voting is necessary at all. You can just plug into your favorite SWF, and if you don't know, how would you know who is the median voter?
Re: sampling, sure, there's an idea there, but because of self-selection and other strategic weirdness, you don't get a random sample. Plus, when the difference is between 49.9% and 50.1%, you sampling error margin will be much larger.
To know where we stand and given the magnitude of the decisions that will be taken on the account of this choice, I think that having the entire population is worth it.--That is assuming the unassumable, i.e. that this system does what people claim it does.
okbut,
You're right that historically speaking it's another matter. But in this context this is just screwing things up one way just because in the past they were screwed up another. Compulsory voting might've helped in 1960's Alabama, but in the end there was no need for it, and today it would be just a nuisance.
denis drew,
If voting really was a public good, then you're right. In that case the marginal social benefit of voting > marginal private benefit of voting and to get them to line up the proper policy response is to increase mpb, not lower it (through coercion) in which case paying people to vote would make sense.
random african,
"
representation or leverage or something like that.
it ain't about who wins or looses but also by how much"
You're right that often it's about the size of the mandate and not just who wins. But in that case the voters are playing a different 'game', where the payoff they get is some function of the margin of victory. So I think you can tweak the example to capture that fact.
gabriel,
Well, I'm not seriously saying we should be polling people to choose our Capo di tutti capi. But both sampling the pop and having them vote would be trying to overcome the same exact problem you mention - elicitation of preferences. In fact, part of the reason why voting dominates polling is precisely because the sample is biased and self selected. I still say that you need to know only a small sample though (in voting you get a sample out of "those who care", rather than a general pop) in order to know pretty well what the outcome would be. Sure if it's 49.5 vs. 50.5 you'll have some problems (of course problems can arise even with voting) - but note that usually close elections like that have higher voter turnout anyway, so there is some self-adjustment in the system. No need for compulsion.
Sneaky,
You missed the main reason for elections: legitimation. "Vox pop vox dei" is a self-evident proposition. It is not self-evident in any rational sense. It is self-evident in a social sense. Modern people just think that, if they have had a chance to pull the lever, they can live with the results. Even if they never pull the lever, or pull the lever for the loser.
Governments can't work without the acquiescence of the governed. Acquiescence requires legitimacy. At one time, it came from priestly mutterings. Now, it comes from lever-pulling.
Lever-pulling may be a ridiculous means of aggregating preferences. But that doesn't mean it is not a real source of legitimacy. And as long as you want to preserve a concept of "your cow" that does not entail physical control of Bossy conjoined with your own artillery, you had better get behind lever-pulling.
anon, I think I addressed the issue of 'legitimization' in the post above this one.
Sneaky,
Is this how you addressed legitimacy:
"In the end though voting is just a clumsy way of society trying to make us feel like we matter and like it cares about us - it's a personalized Oprah show for all of us without the decency to own a cow or two. But every reasonable brain cell in your cerebellum will tell you"
?
What does legitimacy have to do with reasonable brain cells in the cerebellum? (Not that cerebellums have anything to do with reason.) Legitimacy is a social fact. It ain't anything reasonable. No more reasonable than that somebody gives me valuable goods for worthless frog pelts bearing pix of eyeball pyramids.
if legitimacy is an issue, wouldn't mandating voting be cheating ?
after all, a big turn-out has more legitimacy if it's voluntary, hasn't it ?
My view is that compulsory voting reduces the role of money in politics. Much of the fundraising in the U.S. goes to influencing turnout (via advertising, GOTV activities, and in some cases voter supression activities). Since turnout isn't an issue in Australia, political discourse focuses on persuasion rather than turnout. There is less money in Australian politics, and the candidates tend to be centrist problem solvers rather than ideologues. This has an awful lot to do with mandatory voting, I think.
[cross-posted at Mark Thoma's blog]
Voting is compulsory in Australia. There's a thought that if it weren't it would favour the right.
The other to remember is that we only vote for one thing at a time. Federal, state, local. We also don't elect the judiciary.
And we have preferential voting. At federal elections you MUST rank your candidates from most to least appealing.
I have been reading about it and I think that his is really funny!cheap viagraBut anyway. Is higher voter turnout better? great question!22dd
You will have to consider aspect inside an important challenge for one from all of the most beneficial blog sites for all of the net. I actually can suggest that web site!
Vienna Hotels
I needed to write down a brief notice in buy to thanks for all the fantastic suggestions you're placing at this website. My prolonged internet lookup has at the finish of the day been recognized with reliable details and techniques to trade with my co-workers. I would assert that several of us readers really are seriously lucky to live in a important location with quite several excellent individuals with insightful pointers.
Hotels In Bilbao
Hello, I was looking through the website trying to find some data and arrived by your weblog. I'm amazed by the information that you've got on this site. It reveals how nicely you comprehend this subject. Saved this net web page, should come again once more for long term. You, my pal, ROCK!!!
Thanks for sharing this with all of us. Every girl has her most beautiful moment in the life just when you wear your own wedding dresses in your wedding.You can visit our wedding dress shop.Cheap wedding dresses have good quality. You can find ball gown wedding dresses,mother of the bride dresses,mermaid wedding dresses,flower girl dresses,strapless wedding dresses,a line wedding dresses,plus size wedding dresses,column wedding dresses,beach wedding dresses,empire wedding dresses,tea length wedding dresses,bridesmaid dresses,cocktail dresses,evening dresses,prom dresses your dream. Get your wedding dress in our bridal shop. Finally,I hope you have a happy wedding.
At the beginning I thought the post was not interesting, but I have to say I was intrigued with this theme. After read the information you posted, and the way you did it, I change my opinion and I started to feel curiosity for your other entries, so I decided to read them all one by one. Now, I have to say it became in one of the greatest blogs I have read in my entire life.
In a very hone world, you'd be utterly ripe, just... the costs of being able to plaster cast a ballot give birth, over time, ne'er ended up randomly. Your own '2 brains connected with kine' scuttlebutt appeared to be peculiar on the nose because doing so sums up a part of that account.
This diachronic styles that will set when and where the authority to vote (and also in which voting seemed to be bound to represent secret) implemented an exceptionally definitive pattern in addition to logical system.
With this issue learn for example recent employment past Engerman as well as Sokoloff, Acemoglu plus Brown, Baland and Brown, as well as -- upon us ballot rights -- Besley, Persson and Strum.)
Runescape Gold for Sale
Eden Gold
The unique Twine is [url=http://www.mmohome.com/]Buy RS Gold[/url]undoubtedly an excellent demonstrate, it genuinely is simply among my favorites because of this of simple fact 2nd that it absolutely was introduced, [url=http://www.mmohome.com/]WOW Gold[/url]wonderful assaults as well as the toss is excellent.
Thanks for sharing ideas and thought,I like you blog and bookmark this blog for further use.
town car taxi service
I would never have thought of re-creating spaghetti-0's, but I'll bet the grandkids will love them.
Post a Comment
<< Home