November 19th, 2012, 8:00 am . . | . . 7 comments

Why Did Conservatives Doubt the Prediction Markets?

By now it’s well known that just about every conservative pundit predicted a win for Romney (e.g. here, here, and here).  Many of them even predicted a “landslide.”   Everybody on Romney’s staff got it wrong, too.  Romney himself was “shellshocked,” according to one of his advisers.  Early on election day he’d claimed to have written only one speech — a victory speech — because, as he put it, “Intellectually I’ve felt we’re going to win this and have felt that for some time.”  A lot of people have observed that this doesn’t reflect well on GOP psychology and, more importantly, GOP policy.

Yet I haven’t yet seen anybody answer what should be a glaringly obvious question:  why didn’t Team Romney mark their beliefs to market?  Team Romney made faith in markets over faith in government  a  the central issue of the Romney campaign, yet predicted its own victory even though the the prediction markets (e.g. Intrade, Iowa Electronic Markets, and Betfair) were predicting a victory for Obama the whole time.

For example, here is the lifetime history of Intrade shares in Obama’s victory:

Since January 2012 the markets consistently believed that Obama would probably win (i.e. shares never fell below $5 each).

Mitt Romney, on the other hand, never even hit even odds:

And here I have selected Intrade, the most Romney-bullish of the prediction markets.  Other prediction markets were even more bullish about Obama.

Mitt Romney, his entire campaign staff, and all the GOP pundits who got it so very, very wrong, who made faith in the efficiency of markets the central pillar of their campaign, should have either:

a)  Marked their beliefs to market:  looked at what the all-knowing, all-wise markets were saying about Romney’s chances and agreed that Romney was the underdog, or

b)  Bet a lot of money on their beliefs:  if they were really so confident that the markets were wrong and that Romney would win (if they were the “efficiency experts” and “numbers people” they claimed to be), put their money where their mouth was and bet on a Romney victory in the prediction markets until they moved the price in Romney’s favor (this was totally possible, as none of the markets were very liquid).

Yet they seem to have done neither.

The pundits who predicted a victory for Romney were almost without exception the very same pundits who say they believe the market can do no wrong, or at least that the government can do no better.  Why?  It’s a question I think they should answer.

Until they do, here is my best guess:  the people who made these predictions don’t really believe in free market principles, because they are conservatives first, and libertarians second (if at all).

This is something libertarian luminary Friedrich Hayek observed fifty-two years ago.  Conservatives fear and distrust “change,” Hayek wrote, “uncontrolled social forces,” and “new ideas,” whereas the pursuit of political and economic freedom necessarily requires all three.

Hayek wrote that in 1960, but it’s still true 50 years later.  For instance many conservatives prefer:

Now I don’t mean to suggest there’s anything wrong with that, nor that there aren’t a slew of issues on which conservatives are more pro-market than liberals.  I’m just guessing — guessing – that conservatives aren’t nearly as pro-liberty as some of them (including Romney) like to say.

Once you accept that, it seems plausible that right-wing pundits, Romney staffers, and even Romney himself ignored the prediction markets because, well, they just don’t care about markets.  Markets work for them politically and psychologically, but that’s as far as they go — they don’t love or understand markets enough to believe what the markets have say about the odds of Romney’s victory, the prospects of inflation, or the government’s credit-worthiness).  They are conservatives, first and foremost.

Most commentators have offered a different explanation, which is that conservatives chose to believe the facts they wanted to believe were true, not the facts that actually were true.  Hayek, incidentally, offered up the exact same explanation in 1960 for why, among other reasons, he was not a conservative:

Personally, I find that the most objectionable feature of the conservative attitude is its propensity to reject well-substantiated new knowledge because it dislikes some of the consequences which seem to follow from it – or, to put it bluntly, its obscurantism.

That’s the explanation most commentators have offered for the stunningly bad election predictions made on the right, and I don’t see any reason to disagree, but I do find the following, from the same essay by Hayek, just as plausible, and in at least as much need of attention:

It may well be asked whether the name really matters so much.  In a country like the United States, which on the whole has free institutions and where, therefore, [conservatism] is often a defense of freedom, it might not make so much difference if the defenders of freedom call themselves conservatives…  [But] even when men approve of the same arrangements, it must be asked whether they approve of them because they exist or because they are desirable in themselves…

Belief in integral freedom is based on an essentially forward-looking attitude [i.e. liberalism] and not on any nostalgic longing for the past or a romantic admiration for what has been [i.e. conservativism].  The need for a clear distinction is absolutely imperative… the believer in freedom cannot but conflict with the conservative and take an essentially radical position, directed against popular prejudices, entrenched positions, and firmly established privileges.  Follies and abuses are no better for having long been established principles of folly.

It’s not so much a love of freedom that distinguishes the Republican party, but an uneasy alliance between conservatives, who say they love freedom but really love tradition, and libertarians, who genuinely do love freedom and believe in the free market’s wisdom.

The deeper irony is that in some sense Romney supporters were right all along: the market worked very well!  It predicted the election better than they did.  It did something none of them could have: turn near-infinite bits of decentralized information into an accurate signal about the probability of an Obama victory, via a profit motive.

Maybe a libertarian should have won…

As for libertarians who called the election wrong too (I’m looking at you, Russ Roberts, though you were at least humble about your prediction), I assume they were fooled by their conservative fellow-travelers.

 

Update (11/30/2012):  It seems there was one Romney adviser who was not surprised by the election’s outcome.





  • http://aviewfromthecave.com Tom Murphy

    Nice analysis, Jacob. I would include the necessary level of boosterism that is required by each party. To me, what needs to be picked apart are the predictions that were wrong on a massive scale.

    All data pointed towards Obama winning, but it being relatively close. Given the need to pander to the party and maintain positions on conservative news sources, I could understand a prediction that called for Romney to win in a close election.

    The blow out predictions by Morris and others are simply ludicrous. That goes exactly to your point that politics and party devotion can trump political philosophy. All sides are great at muddying political philosophy for the sake of politics, but I am always struck by participating in a conversation with a person who can sweepingly declare a love for freedom while advocating for policies that restrict freedoms.

    Of course, you know that the answer is quite simple. Conservatives doubted the prediction markets because they did not provide the answer that they liked. That moment of cognitive dissonance forced them to grasp to the notion of winning which meant they had to slam the markets. The irony is quite fantastic.

    • http://jacobageller.com/ Jacob A. Geller

      Thanks Tom.

      Speaking of boosterism and bad predictions, here is Dick Morris on why he predicted a landslide for Romney:  

      “I think that there was a period of time when the Romney campaign was falling apart, people were not optimistic, nobody thought there was a chance of victory and I felt that it was my duty at that point to go out and say what I said.”

      Reminds of something Richard Dawkins has said: “American political opportunities are heavily loaded against those who are simultaneously intelligent and honest.”  (Source: http://www.youtube.com/watch?v=VxGMqKCcN6A#t=17m15s)

      Yes, I buy the cognitive dissonance argument (like Hayek), but I find it inadequate.  The mental gymnastics would have to be radically, stunningly, almost implausibly skillful to explain an inconsistency like this.

      I augment that argument with another one (also borrowed from Hayek), that despite what they say, conservatives have a much different, less freedom-friendly political philosophy from the one they claim to have.  They found it easy to disregard the prediction markets, because they don’t much care about markets in the first place.  They care more about preserving tradition than about the efficiency of markets and the importance of liberty and limited government.

      There is a lot of evidence for this in the real world.  I listed some bullet points above, but you could add to that the sweeping GOP profligacy during the Bush years (and I’m referring not just to the Bush administration but to GOP Members of Congress):  http://baselinescenario.com/2011/01/17/deficit-hawkoprite-watch/

      • http://aviewfromthecave.com Tom Murphy

        I 100% agree on this point: “I augment that argument with another one (also borrowed from Hayek), that despite what they say, conservatives have a much different, less freedom-friendly political philosophy from the one they claim to have. ” That to me is where the dissonance occurs. There is a desire and claim to be one thing, but when something personally challenges it there are a lot of back-flips necessary to reach that point.Someone argued it well in the context of the Republican rape gaffes. They said that the problem begins with an unflinching opposition to abortion. When exceptions are presented (which, in a separate conversation, are silly themselves), the opponent is confronted with the example of rape he (in these cases) must try to rationalize a way to make it work.

  • Mark

    It is also worth noting the large discrepancy between prediction markets. There was a very large arbitrage between Betfair and Intrade, even net of any fees, commissions, currency exchanges etc. I know because I took advantage of it. I also know of people who put up 6 figure sums to attempt to profit from this arbitrage but were unable to perform their efficient market duties to close the gap because the weight of money on Romney on Intrade was too big.

    One could hypothesize many reasons why this might occur, but only two I can think of make sense. One, that Intrade is the only American-facing market out of the two and hence given the roughly even support of the two parties the weight of money would naturally be more Romney biased than Betfair, which does not take American accounts. This weight of money was significant enough that not even a significant number of professional arbitrageurs could close the gap.

    However, I prefer option Two. Intrade is the only one of the two that was ever quoted by the American media and hence it would be beneficial for the Republican party (or related supporters) to attempt to move the price on Intrade so as for the race to appear far more even between the two parties and not a lost cause for the Republicans. Compared to the cost of say, an attack ad, it’d probably be quite cheap even fighting against the arbitrageurs that would inevitably show up.

    • Jacob AG

      See Robin Hanson on this subject:  http://www.overcomingbias.com/2012/11/was-intrade-being-manipulated-over-the-last-month.html

      Note that despite the segmented markets and the high transaction costs and the discrepancies and the lack of liquidity and the inefficiency etc etc, the odds were consistently in Obama’s favor, all across the board.  The markets, imperfect as they might be, were very much in agreement, consistently, that Obama’s odds were better than Romney’s.

      Now a clever conservative could try to make the case that the prediction markets were inefficient all across the board, which would solve the conflict between their simultaneous faith in and disregard for markets, but I have not seen one single conservative make that case.  Romney didn’t; his spokespeople didn’t; and the pundits didn’t.

      Cognitive dissonance theory and the conservative-libertarian divide still seem most plausible to me.

  • http://twitter.com/OmnivoreBlog The Omnivore

    The reason the Republicans (as a party) discounted the betting markets is because they believe, not irrationally, that betting is driven in large part by the polling analysis (i.e. people doing the betting are reading 538). Why they didn’t bet *contrary* to that probably has to do with (a) the general difficulty of setting up an account (I understand the barrier to entry is not super high–but somewhat high–especially if I want to use non-US markets) and (b) the the ObamaCare decision where a lot of bettors got burned–in their opinion totally irrationally (and, to be fair, who saw THAT coming in THAT way).

    I also suspect that with such contrary indicators (the GOP pundits vs. the polling) even the most hard core believers were a bit nervous. I saw lots of people predicting a big Romney win–but few willing to place money on it: Not knowing what’s going on between state and national polls can, even if you hold to the state polling, produce uncertainty.

    • http://jacobageller.com/ Jacob A. Geller

      Hi Omnivore,

      Thanks for commenting.  To your points:

      “The reason the Republicans (as a party) discounted the betting markets is because they believe, not irrationally, that betting is driven in large part by the polling analysis (i.e. people doing the betting are reading 538)”

      If this is true, then conservatives implicitly believed the markets were behaving irrationally and giving bad price signals which, aside from being contrary to their stated ideology, should have presented them with great betting opportunities.
      And to your second point:

      “Why they didn’t bet *contrary* to that probably has to do with …”

      a) the barriers are real but not overly high — and not any higher for Romney-bears than for Obama-bears… and

      b) the Obamacare decision surprised everybody, not just conservatives, in which case conservatives should have either assumed the price signals were still accurate OR that the pundits’ predictive powers were nonexistent.  Yet they did neither!

      There is still a mystery to ponder…

      And finally:

      “…even the most hard core believers were a bit nervous.”

      Maybe so, but few of the elites seem to have let it show.  Some pundits hedged their predictions with admissions that predicting is hard, uncertain work, but many seemed quite confident.  Wasn’t Mitt “shellshocked”?  Didn’t he only write one speech?  Where was this nervousness?  Not at the top of the conservative elite… A Romney adviser told CBS ”I don’t think there was one person [on Romney's staff] who saw this coming.”  Also witness the shock and confusion on Fox News on election night… Karl Rove didn’t seem nervous about a Romney loss, he seemed confident, followed by in-denial.

Read previous post:
Assorted Links

1.  Urine-powered generator (but does it work?).  HT Chris Blattman. 2.  Nominal pice rigidity. 3.  Chrome plug-in that automatically scans email...

Close