Wednesday, October 3, 2012

A quick thought on voting

I saw this article this morning on how "your vote doesn't count", and also heard her talk on C-Span about it.

I really, truly despise arguments like this. It's like listening to a die-hard socialist argue against the market. Can you imagine someone making an argument like this about the market? Your choices can't possibly affect prices. OK. Do we just deny the information aggregating properties of hundreds of millions of (individually insignificant) people making choices in a market because "your purchase doesn't count"? Of course not.

This hipster/chic non-chalance about democracy is not a good thing, in my opinion.

22 comments:

  1. Could you explain to a non-American about the relevance of the voting part of a Presidential election?

    As I understand, it is the Electoral College that votes for the American President, yes? And technically, the President of United States is a constitutionally limited office that can only act as far as the Congress and Senate allow it?

    So how can a vote for the US President count when he is actually elected by members of the Electoral College and when he responds to the legislature and not to the general public?

    ReplyDelete
  2. "I really, truly despise arguments like this. It's like listening to a die-hard socialist argue against the market."

    Now you know how I feel when you argue that the choices of 538 congresspeople will lead to a better outcome than the choices of millions and millions of taxpayers.

    "Can you imagine someone making an argument like this about the market?"

    Very easily...you make pretty much the same argument against allowing taxpayers to choose which government organizations they give their taxes to.

    "Your choices can't possibly affect prices."

    Hmmm...is the problem that there are no prices in the public sector? Well...yeah...and there aren't any prices in the non-profit sector either. The driving concept behind markets is that everybody wants more for less. With that in mind...all things being equal...which non-profit would a donor support? A non-profit that saves 1 starving child for a $1 or a non-profit that saves 2 starving children for $1?

    "Do we just deny the information aggregating properties of hundreds of millions of (individually insignificant) people making choices in a market because "your purchase doesn't count"? Of course not."

    Errr...well...you sure seem to deny it when confronted with pragmatarianism.

    "This hipster/chic non-chalance about democracy is not a good thing, in my opinion."

    In my opinion, it's not nearly as bad as nonchalance about the market.

    ReplyDelete
    Replies
    1. There are prices in both the nonprofit sector and in government. Many of which are reached by mechanisms that are similar (or exactly the same) as market prices.

      Delete
  3. Prateek:

    short answers:

    1. The 538 members of the electoral college are elected directly by the voters in each state, according to it's number of representatives(apportioned by population and a minimum of 1) and senators (2 per state). Customarily they always vote for the candidate who won the popular vote in their own state. It only becomes significant in it's own right in the case of a discrepancy between the national popular vote and the result of the state elections this happened in 2000. Hypothetically it could also be decisive in the case of a 3 way split, in which no candidate has the support of 279 electoral college members required to win... the most likely scenario would be that the smallest of the 3 parties represented in the colleg would then choose between the larger two. This has never happened.

    2. Not quite as limited as you would think. The president has tremendous influence through appointing the leadership of the executive branch, and can veto any legislation that passes congress and therefore has a say in all legislation. He is also commander in chief of the armed forces, and as we have seen in Libya, has tremendous discretion in practice. Not to mention that if the same party controls the presidency and the congress, then the president's effective power is almost unlimited (because the president and senate together can appoint supreme court justices who will do as they please).

    ReplyDelete
  4. Come on, Daniel, marginalism! First of all, buying in the market we, of course, get the thing we want. But what's more, our dollar does have *some* effect on the price, even if miniscule. But no single vote has any influence on an election unless it is tied or within one. If my candidate is losing 154,000 to 78,000, my adding the 78,001st vote does not even infinitesmally effect the outcome, because the outcome is binary. The argument you critique is sound: unless you cast the deciding vote, your vote does not matter at all.

    Don't vote! It won't matter anyway.

    ReplyDelete
    Replies
    1. "unless you cast the deciding vote, your vote does not matter at all."

      1. (minor point) Whether you will cast the deciding vote is not known before the fact.

      2 Question: Is it also your position that votes of each of the 154,000 who voted for the winner also do not influence the outcome since their votes were not decisive individually? It is hard for me to understand how this could be so.

      Delete
    2. 2. It's simply because for every vote, the test is the same: compare the two worlds: the one where you vote and the one where you don't. If the two worlds have no change in election outcome, your vote did not influence the outcome.

      @Gene:

      As much as I love marginalism, I think it's fair to say that there is some granularity in prices, so your purchase actually won't influence the price. In theory it would make the true price go from 8.99 to 8.99000000001, but since we don't deal in fractions of cents, there won't be an effect. Actually, you could even call that a nominal rigidity! I'm off to write a paper on recessions being caused by the fact that prices cannot adjust in fractions of cents! ;-)

      Delete
    3. "It's simply because for every vote, the test is the same: compare the two worlds: the one where you vote and the one where you don't. If the two worlds have no change in election outcome, your vote did not influence the outcome."

      Thanks, prometheeFeu, I understand the argument. I just think it's gibberish to say that each of the individual votes count for nothing when we know that collectively they decide the election.

      I would also add as a side issue that the outcome is *not* binary, except in the cases of elections for "dictator for life". The size of a majority (or minority) position will influence the party platforms and candidate selections in the next round of elections.

      Delete
    4. @ArgosyJones:

      I think part of the issue is that people have different ideas on what it means for a vote to "count". I prefer to think of it as calculating individual incentives. (but that's hardly pithy)

      On the non-binary issue, it's true. But the more simplistic binary model can be extended to that. Your 1 vote still won't change the magnitude of the victory in any influential way.

      Delete
    5. PrometheeFeu:
      "It's simply because for every vote, the test is the same: compare the two worlds: the one where you vote and the one where you don't. If the two worlds have no change in election outcome, your vote did not influence the outcome."

      The logical problem there is that that presupposes that you are the only one with free will, while everyone else's actions are determined. As Gott's Copernican Principle says, you are not special.

      Delete
  5. from the article:

    "...they offer a methodology for calculating the value of a vote. On their account, the expected utility of a vote is a function of the probability that the vote will be decisive.."

    I think this is a fallacious idea. Under this methodology, we would have to say that in a system that routinely delivered unanimous votes, that nobody's vote had any value, even though all voters got their preference 100% of the time.

    ReplyDelete
  6. She isn't saying that elections don't aggregate information. She is saying that your vote doesn't count and that therefore the information that is aggregated is low-quality information.

    If a socialist came to me and argued that my purchases don't count because my single purchase can't possibly affect the price level, I would not respond that I have a duty to keep markets working. I would point out that I wanted chicken and asparagus for dinner and that's why I went to buy them and that's why my true preferences were aggregated. In other words, my purchase does count: I had a chicken and asparagus dinner tonight and it was quite enjoyable.

    But honestly, I don't know why this needs saying here Daniel. Surely you already know this. The voting process doesn't provide any incentive to reveal your true policy preferences which makes it about as similar to markets as apples are to oranges.

    Now, surely, saying those things is bad. It makes it less likely people will vote well. But I honestly can't bring myself to pretending that your vote counts in public while knowing the whole thing aggregates policy preferences about as well as a coin flip.

    ReplyDelete
    Replies
    1. Your true preferences counted because you were able to accurately convey to producers what you wanted more of? You told them with your hard earned money..."hey! I really love me some asparagus and chicken! Please keep supplying them! It would really suck if I went to the super market and discovered that they were all out asparagus and chicken."

      Yet, how come the same concept doesn't apply to the public sector? Why should your true preferences count for private good but not for public goods? If it's no big deal that we disregard your true preferences in the public sector...then why would it be a big deal if your true preferences were to be disregarded in the private sector?

      Your true private preferences are disregarded...therefore...you wouldn't be able to enjoy a chicken and asparagus dinner? Your true public preferences are disregarded...therefore...???

      Delete
    2. "Why should your true preferences count for private good but not for public goods?"

      In so far as sending a message to producers, my decision to purchase is equivalent to a vote. It does not count for much. The aggregation of so many votes/purchases makes my individual action largely meaningless.

      Here is the difference:

      If I am rational and I want chicken, I will buy chicken. This will accurately report my preferences to producers. That is because buying chicken gets me chicken. So the aggregated reports (purchases here) will be high quality information.

      If I am rational and I want Romney to be president, I could very well choose to vote against him or not at all. That is because me voting for Romney does not get me a Romney president. So the aggregated reports (the votes here) are low quality information.

      In other words, policy preferences and voting preferences are not the same thing. I may prefer universal healthcare but vote against it because my family is against it. Of course, if I was in charge, I would implement it. I actually want the thing to happen. But I can't make it happen. All I can do is vote. And voting for universal healthcare will make my family angry without getting me universal healthcare. Why would I want that?

      Delete
    3. But what if you just purchase asparagus because you like the way it makes your pee smell? Would that make you irrational?

      "Of course, if I was in charge, I would implement it. I actually want the thing to happen. But I can't make it happen. All I can do is vote."

      Heh, you're like Goldilocks and the Three Bears. Except you only considered the two extremes and skipped over the one that was just right. That's my fault for being too vague. Let me try again...

      Why not just allow taxpayers to choose which government organizations they give their taxes to? For example...at any time throughout the year...you could go to the EPA website and submit a tax payment. They'd give you a receipt and you'd submit your receipts to the IRS by April 15. Of course, if you didn't feel like directly allocating your taxes then you could just give them to congress.

      Delete
    4. The reason why I want asparagus is irrelevant. Think of it this way: in markets, satisfying my desire for asparagus is what tells the market that I want asparagus. In elections, I cannot satisfy my desire for particular policies by voting for those policies. Therefore, the signal (voting) is divorced from my policy desires.

      I would surely support that. But it suffers from the same problem. Since I am not Warren Buffet, my contribution could not make any significant difference to whether the EPA gets funded or not. So once again, the signal (the contribution) is divorced from my policy desires.

      Delete
    5. "Therefore, the signal (voting) is divorced from my policy desires."

      So I take this to mean that you agree that kids of any age should be allowed to vote?

      "I would surely support that."

      It would be awesome to hear you put it in your own words why exactly you would support giving taxpayers the freedom to choose which government organizations they give their taxes to.

      "Since I am not Warren Buffet, my contribution could not make any significant difference to whether the EPA gets funded or not. So once again, the signal (the contribution) is divorced from my policy desires."

      You would be giving your own taxes to the EPA...so how would your signal be divorced from your policy desires? "No no no...I don't really value protecting the environment...but I just give the EPA my taxes because it makes my gf happy...and I value her happiness more than I do my own." The reason why you want protection for the environment is just as irrelevant as why you want asparagus. The basic fact of economics is...however you spend your money speaks a lot louder than your mouth/vote does. We either incorporate your "true" preferences into how scarce resources are distributed...or we don't.

      Delete
    6. "The basic fact of economics is...however you spend your money speaks a lot louder than your mouth/vote does."

      No. You're confused the basic fact of economic you're referring to is that people respond to incentives and that when their incentives are properly aligned, revealed preferences are more accurate than self-reports. So under your scheme, let's assume I want the EPA to be funded. Scenario 1, I send money to the EPA. Scenario 2, I do not send money to the EPA. What is different between the two scenarios? Nothing of consequence. My whole tax bill could not pay an extra secretary at the EPA, much less significant regulatory activity. So your scheme suffers from the same weakness as voting: your incentive is not to reveal your policy preferences.

      I support giving tax-payers having the freedom to choose which government organizations they give their taxes to because it is more morally correct and it would make me feel good to not send money to certain programs. For instance, I would be much happier if I did not send any money to the USCIS. Not that it would impact policy, but hey, it would feel nice.

      Delete
  7. Gene FTW (surprisingly, since I would have thought he'd like voting).

    ReplyDelete
  8. The comparison to the market leaves me cold. I go to a fixed-price market to get something worth more to me than I am required to pay for it. The benefit of my vote is (in purely consequentialist terms) only probabilistic.

    On that issue, here's John Quiggin: "I’ve pointed out, many times that the analysis behind this claim is false in general (it may be valid for particular voters in systems like the US presidential where some states are safe for one candidate or the other). The simple intuition is this – absent any info about location etc, the likelihood of casting a decisive vote is (roughly) inversely proportional to the number of voters, say 10^-8 in the US context. If it costs you $10 to vote, and you think the benefit to the world of having Obama rather than Romney exceeds $1 billion, then as a consequentialist you should vote."

    ReplyDelete
  9. "If I am rational and I want Romney to be president"

    Heh.

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.