If you want to understand why political "futures markets" are so--how shall I put this delicately?--fucking dumb, first watch this, courtesy of my man La_Rana:
The ideology that underlies the "wisdom" of markets is pretty kooky. Hell, downright nuts. The idea that agglomerations of human beings necessarily choose wisely is roundly disproven by human history. Like, all of it. Hell, Easter Island ought to be enough.
Now I am not in the least opposed to the idea of public ownership. Joint-stock ventures? Rockin. But these multiplying futures markets are just plain, old-fashioned speculations. Rather like our own perverted "markets," no? Believing themselves to be predictive, they are in fact reactive. There's nothing inherently wrong with that, but rather than reacting to the old "fundamentals"--strength and popularity of brand; profit margins; bottom lines; balanced budgets; public reputation; etc.--they react to a collection of ineffable perceptions largely held not by customers but by fellow speculators.
In the case of these political futures markets--again, just as in any other speculators' market--participants are informed by external data. Polling numbers, for instance. And so their predictions are rarely wildly deviant from other attempts at the same. In fact, they're often wrong in the same direction as conventional predictions. Witness the Democratic race in New Hampshire, where the market made the same error as the major polling outlets. The idea that speculative investment produces unique insight into future outcomes, though, is pure nonsense. Any reasonably informed individual can make intuitive predictions about political outcomes with the same or better accuracy than the "market."
One misses the days of augery.
Wednesday, January 09, 2008
Beware the Ides of March
Subscribe to:
Post Comments (Atom)
21 comments:
Dude - the idea isn't that "agglomerations of human beings necessarily choose wisely." It's that they can (not have to) predict accurately, given the right incentives.
I've got no problem saying that a prediction market has the same problems as other markets (speculators, Greater Fool theory, etc), but saying that it has the same problems as, say, voting just offends me.
(On the other hand, I'm glad to have a blog where I can use "voting" as a bad word)
"Incentives" don't confer prescience, alas. Responsibility, perhaps. The right incentives can make people better investors, but "groups can be wise" is the same as saying that individuals can be wise. Yes. And?
Video already down. Sum up ?
My take is slightly different (especially since I was poking at you with this video, which is not down). Numerous social science experiments have shown that under the right conditions, groups of people can make better decisions, on average, than individuals. The lesson, I think, is that markets only work in specific, highly scripted situations. The incentives need to be useful for the many, not the few, demand must be elastic, the ability to coerce and abuse other people must be kept to a minimum, etc. We, of the freest markets in the world, already do all of these things, but we pretend as if the child labor laws are something different than a simple, necessary, market constriction.
Nah, augery is just boring.
There are lots of problems with the idolatry of markets. One you don't mention is that investors/speculators have totally different sets of concerns from other kinds of consumers--which is kind of a problem when you're talking about, you know, voting. My favorite example of this kind of thing is an article that appeared in a history journal quite a few years ago, written by an insurance executive who had quite a few choice things to say about how silly and fatalistic people used to be and how much actuarial science has advanced since back in the dark ages. All well and good until you realize that his profit margin depends on the "silly fatalism" of his customer base.
Bravissimo, anonymous!
Surowiecki made some interesting observations about the so-called "Wisdom of Crowds," but limited them to where the conditions of diversity, independence, decentralisation and aggregation apply: each person having their own interpretation of the known facts, not determined by the opinions of those around them, specialisation and local knowledge, and a mechanism for turning private judgments into a collective decision.
The problem, I think, is in taking a limited, specific dynamic and using it as a metaphor for human nature per forza. I can haz objektivizm?
but "groups can be wise" is the same as saying that individuals can be wise.
Right. The problem is: how do you know which individuals are wise? Academic degree? New York Times column? Religious membership?
There are few reliable ways beforehand to ascertain someone's expertise. But (and I'm paraphrasing Surowiecki here), in a large enough crowd, the idiots should by-and-large cancel each other out, giving the well-informed participants a greater weight. And when those participants can weight their input even further - e.g., with dollars - the effect is pretty substantial.
Ideas markets aren't perfect, of course, but literal perfection isn't a reasonable goal. The "Clinton Democratic nominee 08" contract at Iowa's Electronic Markets, for instance, has been dropping steadily since the 1st of this month. I don't think her sudden increase in delegates yesterday invalidates that.
Any reasonably informed individual can make intuitive predictions about political outcomes with the same or better accuracy than the "market."
If this is true, then all "reasonably informed individuals" can go to InTrade using real money and cash in. Make money fa$t! In fact, you should be doing so, assuming you count yourself as reasonably informed.
Gotten rich there yet?
If we are going to get into the Wisdom of Crowds stuff, to which I alluded, the one thing no one must ever forget(as Surowiecki is wont to do), is that it only works if the participants have zero interaction with one another. The instant participants start interacting, they begin to influence one another; at 100% interaction the crowd becomes a mob. This of course means that the supposed wisdom to be gleaned from crowds will have limited applicability in the real world, i.e. Surowiecki is a moron.
Leonard - I think that's a confused metaphor. The behavior of political futures markets don't actually affect--except perhaps in some tenuous, tangential way--the outcomes of political contests. The claim, instead, is that such markets can predict the future disposition of something external to the market itself--in this case, who will win elections.
To follow your comparison, I'd have to be claiming that a person could reasonably predict the day-to-day fluctuations of opinion within that futures market itself. That wasn't my claim at all.
Remember what these phony market enthusiasts are claiming: that their enterprise of collective wagering produces accurate predictions about the outcomes of events that are not determined by the markets.
Speaking of the lack of wisdom of the Easter Island crowd, fairly convincing fairly recent evidence suggests that Jared Diamond's take on the story isn't really what happened. Brief summary - the trees disappeared not because the people foolishly chopped them all down, but because the rats (who arrived at the same time as the people) ate the seeds.
See http://tinyurl.com/ldwbm
IOZ (may I call you "your fuzziness"?), I don't understand your rebuttal. What is being bet on are actual, future, real-world outcomes. We are agreeing on that. And we also agree that the bets in the market have essentially no effect on those real-world outcomes. From this you somehow then leap to the conclusion that you have to "predict the day-to-day fluctuations of opinion within that futures market itself" to make money. No, you don't. You only have to be able to pick winners with better accuracy than the "market" is picking them.
You seem to have some model where the prediction markets never judge outcomes, but that's not how they work.
Put another way: what I was saying was not a metaphor. If you can predict politics better than those markets, you can make money in them. Not metaphorically, literally. Here's how: determine, using your reasonably informed intuition, who will win one of the contests bettable on at InTrade. Let's take, as an example, the South Carolina primary that will happen in a week. You can buy contracts, right now, as follows:
obama: 77.9
clinton: 29.0
edwards: 3.0
Say you think Obama will win, just using your intuition. Now you take $7790 of your hard-earned cash (I know you're loaded), and buy DEM.SC.OBAMA. That comes to 100 shares. Now you wait until Obama wins, and boom - each share pays off $100. You get $10000: a profit of $2210 in under a week! Zoiks!
Of course, if your intuition is not quite up to snuff, you lose your whole investment. So, this is not without risk. To really make money via superior information in a market, you need an understanding of risk and how to parcel out your total stake into a portfolio of investments to minimize your risk of bankrupcy. But that's rather a detail.
The main booster of prediction markets seems to be Robin Hanson. He responds to recent pooh-poohing here. Unsurprisingly, Bryan Caplan has his back.
Leonard,
I think what IOZ is saying as that these "markets" do not have actual predictive power that is superior to a "reasonably well informed" individual.
It's not that you can't make money by being right. It's that the market as a whole doesn't have a better chance of being right than you do, which is what the linked video appears to claim.
ultima - I hope you realize the foolishness of your statements above. The "smart" people will rise above the rabble in a crowd. In what world?!?
You have heard of mob mentality, the Milgram experiments, Britney Spears.
The group does not raise up to the intellect of its smartest member(s). I wish it did, but experience tells me otherwise.
I make it a point to discount heavily any common wisdom before taking any advice from any group think "findings."
There is speculation about what element is missing from political prediction markets at the Organizations and Markets blog here.
Political markets would likely be a lot more accurate if there was significantly more liquidity.
If I research a major NYSE stock to find evidence of a 5% pending price change, I can buy millions of dollars of stock or options to try to earn back on my research.
These political markets don't allow it. Hell, a thousandaire is a high-roller in most of them. As such, there's no reason for anybody to put significant effort into beating the official predictions.
Agreed with anonymous above.
Also, remember that these political markets, like sports betting markets, aren't really prediction markets... they're probability markets.
I do agree that any market is going to be affected by emotion and wrong information and therefore not accurate.
I've rarely seen such limbo dancing in an attempt to justify "the market."
The idea that the uninformed simply "cancel each other out" is ludicrous, it is a forecast of fantastic imagination.
The idea "if informed, groups vote better than individuals" is irrelevant. Totally irrelevant.
Individual decisions on spending money are NOT scripted, they are NOT done as informed decisions, and they are NOT likely to trigger mutual cancellation of "bad decisions."
Slave away at the excuse-making. It's nothing more than denial of reality. There is no such thing as "the market" and therefore the notion of a "free market" is equally specious.
Love you libertarian nincompoops, you're mighty entertaining. Gimme a shout when you return to the world of the living human being, will ya? I"ll be interested in how your lovely theories have squared with reality. I'll bring you a box of Kleenex so that you may mop up the sad sorry tears you cry at the revelation of your long-term delusional state.
Hi, I do believe this is an excellent web site.
I stumbledupon it ;) I will revisit yet again since I bookmarked it.
Money and freedom is the greatest way to change, may you be rich
and continue to help other people.
Here is my website - Mulberry Outlet UK
Post a Comment