Should you bet to equalise average utility, or raise marginal utility?

Harry Clarke ponders whether to bet $1000 on Labor:

If Labor did win I would net about $700 which would buy me 4 dozen 2004 Kalimna shiraz and would yield me a-once-weekly decent bottle of plonk for almost a year. … If Labor lost I would lose $1000 but gain the utility of having the party I support electorally in power for another term and being able to avoid several years of comradely triumphalism in the blogosphere.

If you believe that the betting market is efficient, then one reason you might want to bet is as a form of insurance. In that case, you should bet on the party that you want to lose. In the last election, one of my Labor-supporting colleagues followed Harry’s reasoning – placing a bet on Howard so that he could drown his sorrows when the Coalition won.  

But another view is that you should move money to the state of the world where its marginal benefit will be higher. So if you are a die-hard Coalition (Labor) supporter, the wine will taste better in a world where Howard (Rudd) is PM. To see how this might work, imagine that there are two possible worlds – one in which Rudd is PM in 2008, and one in which Howard is PM in 2008. If you’re betting to get more out of your marginal dollar, you should set things up so that you have more money in the possible world that you would prefer to live in. On this basis, you should bet on the party that you want to win.

(Of course, all this assumes that the betting odds are an accurate reflection of the true probabilities. Poll-addicts would presumably behave differently. An easy way to test whether journalists and commentators really believe their own polls is to ask them whether they’re now betting their life savings on Labor.)

Advertisements
This entry was posted in Australian Politics. Bookmark the permalink.

8 Responses to Should you bet to equalise average utility, or raise marginal utility?

  1. The Colonel says:

    I plan to get drunk whoever wins. This appears a failsafe strategy.

  2. steve says:

    Read an interesting theory recently that a futures market that has been running in the US for years most accurately reflects the political winner in the US. Think it was a Boston University economics project initially. Of course, it could not happen here because there is no set date for the election.

  3. reason says:

    Of course, all this assumes that the betting odds are an accurate reflection of the true probabilities.

    There is also the massive assumption that you are right about the effect on the quality of your life on who is in power. Remember the US Presidency/Dow Jones “conundrum”.

    I also don’t understand why the marginal value of money would necessarily be higher in a “better” world. The marginal value of value of money surely increases when you have less of it, so I assume you imagine that Labour will cut your taxes.

  4. reason says:

    I my post above, I was not aware that the HTML blockquote didn’t work. The first paragraph is a quotation for the confused.

  5. EconoMan says:

    Given that the betting sites all take a significant margin, I’d suggest that if you believe the market is currently efficient, don’t bet at all. I put a sizeable (for me) bet on McKew to win Bennelong at $4.75. Not because I expect she will win, but because my opinion of her probability of winning was higher than the 21% implied (ignoring margins). With the change in prices, I could now lay off for a guaranteed win, barring an independent winning. (I still think the price on Howard is too low so I’m waiting.)

    Anyway, on the insurance versus ‘everything tastes better’ argument, I’m not sure I agree that the party in power impacts on the utility of everything else, as opposed to being a separate source of utility. I will gain happiness if the ALP win the election, but I don’t expect that it will change the marginal utility I gain from all other activities in my life. It might increase my marginal utility of commenting on politics related blogs such as this one, but I don’t really need betting winnings to do that!

  6. Matt says:

    You present them as if they’re distinct, but the idea of betting as insurance and betting to maximize expected utility (or as you put it, “move money to the state of the world where its marginal benefit will be higher”) are not.

    The reason people buy insurance is that the return on the policy is high when the marginal utility of consumption is high (i.e., my house burned down, so that benefit check is REALLY useful).

    How we think about your question simply turns on whether the marginal utility of wine consumption is higher when your party is in power (party time!) or is not (drown your sorrows).

  7. Stephen L says:

    Yes to Matt. My marginal utility will be much higher if the Libs win (which is why I have some money on them). If Labor wins I won’t need the money much. This is partly because they promise to expand industries in which I work, but also because if they’re putting more money into things I believe in I need to do less work dipping into my pocket to keep institutions I believe in alive.

    On a more psychological level, if the world feels like a better place why do I need money? Its only if there is no pleasure in the intangibles of life that I need to spend money on luxuries to distract myself.

  8. Jack Lacton says:

    Coalition is $2.20 on Betfair. Put everything you’ve got on them and then lay off when they’re at around $1.90, which is guaranteed to be the case. This will allow you to make a nice profit regardless of the result.

Comments are closed.