Presidential Decision Markets

We have had many prediction markets on who will be elected, but almost none on who should be elected.  So far, the only exceptions I know are decision markets on which nominees would most help their party gain the U.S. presidency.   

Today, I’m pleased to announce that Peter McCluskey has funded the creation of six now-live InTrade markets that should tell us how a Democratic versus Republican U.S. President will differently effect oil prices, long term interest rates, US government debt, and US troops in Iraq!  If enough people trade these assets, then for voters who know which direction they want these parameters to move, InTrade market prices could advise them on how to vote!

Two different approaches will be tried on these four parameters.  Two parameters will use a binary shock response futures approach, with one asset each paying based on whether, over the course of election day, oil prices or T-bond interest rates move in the same direction as the probability of a Democrat winner. For example, if you think (speculators think) that a Democrat is more likely to raise oil prices than a non-Democrat, you should be willing to pay more than 50 for the asset that pays 100 if these two move in the same direction.  You can cash in these assets a few days after the election. 

For the two other parameters, four assets will support simple conditional estimates.  Each parameter will have Democrat, Non-Democrat asset pairs.  The Democrat asset in the pair pays only if a Democrat is the next president, while the other asset only pays otherwise.  Bundling each pair together gives assets whose prices estimate:

  • The probability US Govt debt will rise from FY 2010 to 2011 (assets: D, ND).
  • The number of US troops in Iraq mid 2010 (assets: D, ND).

Dividing the price of each individual asset by market chances of a Democrat president (or not) produces estimates of these parameters conditional on a Democrat president (or not).  The difference between the Democrat and the other conditional estimates say how much of a difference speculators expect a Democrat to make on that parameter.  You’ll have to wait up to four years to cash in these assets. 

I see four key open questions:

  1. Can we get precise enough prices to see a Democrat versus not difference?
  2. Can we convince voters that such prices give reliable non-partisan estimates? 
  3. Can we convince voters these prices show causal effects of the party in power?
  4. Can enough voters tell which way they’d like these parameters to move for prices to be useful?  (I’m not sure which ways I want.)

Added: Peter has been told InTrade will waive all trading fees in these markets!

GD Star Rating
loading...
Tagged as: ,
Trackback URL:
  • Silas

    Robin_Hanson: I read the “shock response futures” and I’ll have to echo and extend the commenter’s response. That method of determining impact of a candidate is highly flawed. The market will likely guess the winner before election day happens, meaning the prices have already incorporated the impact of that candidate, unless the election is very close. So the change in one day won’t tell you anything.

    A better way might be to compare, over the n months before the election,

    a) a given party/candidate’s chance of winning (measure by polls or other prediction markets)

    to

    b) the movement of those critical prices.

    As an example, that might mean: for the 3-months before the election, plot 10yr treasuries against John Edwards’s polling numbers, and run a statistical regression, and repeat for all candidates. Then you can bet on, say, which candidate do financial markets *most* believe will negatively impact 10-yr treasury yields. Assuming they can understand the math there.

    Just an idea.

  • http://profile.typekey.com/bayesian/ Peter McCluskey

    Silas, if the markets were 100 percent certain of the outcome before election day, the shock response futures would be useless. If they are 95 percent certain, then I would expect the change in confidence over election day to affect oil and bond prices (but the signal to noise ratio might be a problem).
    Even if this is a problem near election day, the shock response futures won’t be able to incorporate that information until markets become fairly certain about the election outcome. Since I care more about what information the prices convey when key voters can still be influenced by news, I’m not too concerned by the possibility that prices will become less informative after the election is effectively decided.
    I hope that scholars do the alternative analysis that you suggest, but it won’t be easy to determine how reliable the results of that analysis are.

  • http://gondwanaland.com/mlog/ Mike Linksvayer

    This is great, thank you Peter!

    Minor Intrade complaint: I wish they had grouped the two Dem/NonDem pairs into “events” as they do with other related contracts, so that there would be 4 rather than 6 presidential decision events to navigate.

  • http://dmorr.livejournal.com Dave Orr

    It turns out that the regular markets are pretty often wrong when it comes to elections, so prediction markets might be wrong too.

    In particular, the stock market generally drops if a democrat wins and goes up as a republican wins. Yet the markets have almost always done better in the first year and in the whole terms of democrats than republicans since at least WWII.

    In other words, the stock markets very consistently react in the wrong direction to presidential election results. If you had watched which way the market went on election news and bet the other way, you would have made quite a lot of money. Is there any reason to think that prediction markets would behave better?

  • http://entitledtoanopinion.wordpress.com/ TGGP

    Looks like we’re part of the way to Futarchy.

  • http://hanson.gmu.edu Robin Hanson

    Dave, we aren’t asking for perfection, just something at least as good as the alternatives – do you know of a better source?

    Silas, Peter is right, as long as some info remains to be revealed in the last day, any expected correlation with party should show up in these assets.

  • Silas

    Robin_Hanson and Peter_McCluskey: Wouldn’t you still agree that my metric would have much greater statistical significance and would be far more robust against interference from the regular noise of the market? (noise that could exceed the size of the correlation on election day…)

  • http://www.nber.org/~jwolfers Justin Wolfers

    Congratulations Robin and Peter – I think that this is a fantastic innovation, and I look forward to tracking these markets.

  • Rick

    An error occurred…

    We’re sorry, your comment has not been published
    because TypePad’s antispam filter has flagged it as potential comment
    spam. It has been held for review by the blog’s
    author.

    Wow. Just wow.

    Okay, let’s try again.

    Robin, you have no fascistic tendencies in your censorship fetish.
    You’ve overcome all bias and truly desire a “tighter” comment system
    because of the masses of threatening, spam and unintelligible comments
    that so clearly flood this site every day. And of course you are
    always right and never wrong. And to prove that point let me show you
    what an ignorant fool would write if he were trying to comment on this
    post. Of course my own opinions are diametrically opposed to
    those of this terribly biased idiot. I’m with you. We’re all
    and always with you. (But I may as well sign on from a
    different computer with a different login name anyway.)

    Okay, I’m a big fan of futures markets. They fascinate me
    philosophically, psychologically and mathematically and I therefore
    appreciate that you bring your learned knowledge of the matter to our
    attention. But, as usual, you surreptitiously slip your own biases
    into the mix – and what’s worse – you do so in the push-polling manner
    that makes it look as though your biases were actually our own ideas
    in the first place.

    Now I know that this will only prompt you to more vigorous calls for
    comment censorship (something quite extraordinary in a blog actually
    named “Overcoming Bias”) but Robin, I need to point out that
    your piece opened with the line: “We have had many prediction markets
    on who will be elected, but almost none on who should be
    elected.”

    You then went on to inform us of the good news to the extent that
    this problem is now being rectified on account of our being newly able
    to bet on oil prices, government debt, etc. But Robin, if our goal, is
    overcoming bias can’t you see that the creation of these
    particular markets contribute to rather than lessen our biases
    as to what’s important?

    Look, I realize that a blog post can’t cover every single aspect of
    the subject under scrutiny but the fact that you consistently
    try to pass off economic factors that favor the wealthy as “objective”
    economic preferences that we should all advocate for and as the
    barometers for success that policy-makers should focus their
    attentions on draws me to the required stand to point out those
    biases.

    As I’ve referenced
    here
    before, you
    know
    good and well that the majority of the citizens of this
    country would live happier, longer lives if our national economic
    policies (as well as moral attitudes) were more New-Dealish than
    Milton-Friedmanish. That being the case, the barometers that might
    best inform us as to “who should be elected” are more along the
    lines of looking at rates of inequality, rates of homelessness, rates
    of uninsured citizens, wages for blue-collar jobs, purchasing power
    etc. – unless
    of course
    you “look
    down on patriotism and piety of every kind”,
    in which case the
    hunger of billions is but a hill of beans as compared with the
    immorality of demanding that the wealthy be bereft of a few more
    dollars of their “hard earned” cash.

    So Robin, the creation of these markets increases bias.
    They inform the masses who come to be aware of them that these
    are the important factors to take into account when choosing from
    among presidential candidates. Which, as implied earlier, is
    what the repugnant push-poll accomplishes as well.

    I could go on responding in advance to potential rejoinders such as
    “so go make your own markets” and other such zingers but you’ve made
    it quite clear that critical comments are less than appreciated here
    so I’ll tiptoe my way out.

    Cheers,

    mnu ez
    mnu ez . blogspot . com

  • Rick

    The formatting was perfect the first time I attempted publishing. Thanks for giving me the run around Robin.

  • http://hanson.gmu.edu Robin Hanson

    Silas, we want a measure available well before the election.

    Mnu, surely higher and lower oil prices, or troops in Iraq, can’t both benefit the wealthy.

  • Silas

    Robin_Hanson: There seems to be a miscommunication here.

    Peter_McCluskey’s idea involves people betting on the eventual value of a metric that takes as input the events on election day.

    My idea involves people betting on the eventual value of a metric that takes as input the events on election day and n months preceeding.

    In what sense does Peter_McCluskey’s idea involve a measure that is available before mine?

  • Caledonian

    Mnu, surely higher and lower oil prices, or troops in Iraq, can’t both benefit the wealthy.

    Law of Acquisition #35: Peace is good for business.
    Law of Acquisition #36: War is good for business.

    Either condition can open up new and exploitable niches for the wealthy. They probably won’t be the same ones, but they’ll be there all the same. It is entirely possible that both higher oil prices and lower oil prices would benefit the wealthy.

  • http://arundelo.com/ Aaron Brown

    “effect oil prices” –> “affect oil prices”

  • http://hanson.gmu.edu Robin Hanson

    Silas, I mispoke. The point is to disentangle the direction of causation. On election day most of the causation goes from election to other parameters. Over the preceding months it is much less clear.

  • Silas

    Robin_Hanson: That would be a valid point if I were proposing simply comparing the initial and final values. But my method involves comparing the correlation over a succession of days. To win on my metric, the economic variable would have consistently, over a longer time, move in the same direction as the candidate’s chances.

    In any case, financial markets on election day close before the results are in, or close to certain.

  • http://hanson.gmu.edu Robin Hanson

    Silas, a consistent correlation would not show the direction of causation.

  • Silas

    Robin_Hanson: Neither would an election-day-only correlation, as Peter_McCluskey’s is. Was your point that the months-long, day-to-day correlation is necessarily unrelated to the causation direction, while the election-day-only correlation is strongly related to the causation direction? Or was it a more general point about “correlation != causation”?

    If the former, I disagree. Any countervailing event on election day (like the errant $100 oil trader) larger than the magnitude of the candidate’s effect, flips the sign. In contrast, on my metric, those events must also operate for n days, consistently in the same direction as changes in the candidate’s changes. That why I bring up that my metric is more statistically significant.

    And any chance of taming the italic tag?

  • http://profile.typekey.com/bayesian/ Peter McCluskey

    I don’t want to establish these particular indicators as standard measures for voters to use. But mnuez’s suggestions that people use a larger number of more special purpose indicators would be a step in the wrong direction, as the more indicators that people attempt to use, the easier it will be to emphasize those that support one’s biases.
    I want future decision markets to try to use a few broader measures of wellbeing that capture a larger portion of what it is that people want to maximize.
    One such measure that I have some desire to use is life expectancy.
    One reason I haven’t used that is that I suspect fewer people have useful knowledge of how politicians affect life expectancy, so it’s harder to attract informed traders with the kind of subsidies I’m currently willing to provide.
    Another reason is that the effects of politicians on life expectancy will take longer to become measurable, and Intrade’s business model does not encourage them to design their system to make long-term forecasts work well.

    Silas, I can imagine that the correlations you want are caused by, say, increased oil prices causing voters to prefer a Democrat.
    I could imagine that we see something similar with the benefit of hindsight for the shock response future contract (i.e. a sudden surge in oil prices starting on Monday afternoon of election week influencing the election result). But in order for that to affect the shock response future price weeks before the election, the shock response future price would need to incorporate an expected last minute oil price surge better than oil futures prices do. I find that far-fetched. Can you describe a more plausible way in which shock response futures would reflect (a week before the election) a causality other than election results causing change in oil prices?

  • Silas

    Peter_McCluskey: Can you describe a more plausible way in which shock response futures would reflect (a week before the election) a causality other than election results causing change in oil prices?

    I can imagine on election day, an errant trader might eat a 2% trading loss ($2000 on a $100,000 trade) just to make it look like “Democrats cause higher oil prices”, or because he stands to win more than $2000 on shock futures that he’s bought. (Just like someone took a loss with $100 oil because of other goals.) People buying futures, then, have to incorporate the risk of this, as well as the risk of anything with a larger magnitude than the election results and opposite in sign.

    I think it’s a good point that my metric might capture causality in the opposite direction, i.e. people change their votes (which changes the odds) because futures prices change. However, traders more closely monitor conditions that could affect their portfolios (because of the financial interest) than voters monitor conditions that could affect their votes, so traders are much more sensitive to small changes in election odds, than voters are sensitive to small changes in futures prices. Thus, a correlation is more likely to indicate the odds –> futures-prices causation direction.

  • http://profile.typekey.com/bayesian/ Peter McCluskey

    Silas, the December 2011 oil futures are currently traded heavily enough that a trader could not reasonably expect to manipulate the results of the shock response future contract without trading oil futures worth millions of dollars.
    I’m not suggesting that voters are sensitive to daily price changes, only that they’re sensitive to changes over months in prices they pay at the pump, and traders may change their predictions about that effect as fast as oil futures change.

  • Silas

    Peter_McCluskey: The trader doesn’t have to fundamentally shift the market. He just has to make one trade show up right before the cutoff point for your measure, thus making “the” price where he wants it to be.

    I’m not suggesting that voters are sensitive to daily price changes, only that they’re sensitive to changes over months in prices they pay at the pump, and traders may change their predictions about that effect as fast as oil futures change.

    Okay, point made. But then, the reverse correlation problem could show up on election day too. The oil prices change voters’ minds as going to polls, which changes odds, which affect futures on financial markets, which affect oil prices …

  • Pingback: Robin Hanson’s conditional prediction markets | Midas Oracle .ORG

  • http://www.gwern.net/Prediction%20markets gwern
  • http://daedalus2u.blogspot.com/ daedalus2u

    Traders don’t have to shift markets only through differential pricing. They can shift markets the old fashioned way, through marketing and disinformation.

    How much are corporations spending to influence the election? Billions? If corporations and individuals are willing to spend millions, tens of millions and even billions on shifting public opinion, presumably they would be willing do the same thing with oil trading profits which are from pre-tax income.

    What does all the saber rattling about war with Iran do to oil prices? It drives them up. Who is rattling sabers about war with Iran? Mostly the GOP. Who does the GOP blame for high oil prices? Obama.

    The price of oil has two components, its actual value as a commodity used to supply energy and other things. The price also has a component of signaling. This component signals the relative risk that future prices will be higher or lower.

    Oil Users make profits using oil to produce other things. Oil Speculators make profits by capturing differences between the signaling prices at different times. The signaling can be manipulated easily, just rattle the sabers with Iran. Oil Users have to take potential future disruptions seriously because they have more to lose than speculators. Oil Users have the financial risk of not recovering their investment in what ever factory it is that they use to turn oil into products that they sell.

    If you look at the ratio of US oil consumption to GDP, there are large fluctuations.

    http://www.epmag.com/Production-Drilling/Oil-price-America-afford_41209

    with GDP varying between 100x oil consumption (1999) and 15x oil consumption (2008).

    Oil users with a lower GDP production to oil consumption ratio have a reduced ability to tolerate higher oil prices. Oil is not the only component of what goes into their production of GDP. Oil is the only component of an Oil Speculators production of GDP.