| Democracy is the worst form of government except for all the others except possibly futarchy |
[Jan. 14th, 2013|12:11 am]
|
I recently read Nate Silver's treatment of prediction markets in The Signal and the Noise. It was very good, but like most other such treatments it tended to focus on whether the best experts can do as well as prediction markets. The belief seems to be that if experts can equal - or perhaps even slightly outperform - these new market solutions, then no one can force us to switch to this complicated unorthodox system and we can safely keep relying on expert predictions. This post is about the reasons I disagree with that assessment.
A prediction market is a stock market analogue in which people buy and sell bets in order to predict the future. Someone creates a financial instrument that pays off $10 if the Democrats win the next election and $0 if they lose and people bid on the value of the instrument. If the instrument ends up priced at $6.50, that means the market thinks there's a 65% chance the Democrats will win.
These markets have some very interesting properties. The coolest is that it will always be the most consistently accurate source of information available. The proof is like so: suppose there were some source of information which was consistently better than the prediction market. In that case, whoever bet on the prediction market using that other source's predictions could consistently become rich. Many people like being rich, so someone would do this.
But this economic activity would move the prediction market's prices/predictions until they became as good as or better than the other source's predictions. Unless you faster than everyone else playing the market, this will have already happened by the time you see its predictions. Therefore, the prediction market will always be the most consistently accurate source of information available.
Another cool property of prediction markets is that they're impossible to corrupt. Suppose the Democrats wanted to make themselves look more popular in order to convince campaign donors they were a shoo-in. So they spend a million dollars bidding on "The Democrats will win the next election" and driving the price up to $9.90, or a prediction of a 99% chance that the Democrats will win. It seems that the prediction market has been corrupted.
But suppose you notice this. You know the Democrats have a less than 99% chance of winning the election; therefore you can beat the prediction market. Therefore, you can get rich. You short shares of "The Democrats will win the election" until it goes down to whatever probability you think is correct. Other people do the same. Investors start a feeding frenzy as people realize what a big opportunity such an obviously wrong prediction is, and big firms with lots of money to spend on exactly this sort of situation join in. Eventually the prediction returns back to its correct level. The Democrats' plot to corrupt the market has turned into that the Democratic Party has turned into a plot to give away a million dollars by subsidizing more rational investors. The market easily returns to the correct level.
Robin Hanson has proposed that the government should use prediction markets to inform policy decisions. For example, one of the big controversies surrounding gun control is whether it will lower the crime rate (because fewer criminals have guns) or raise the crime rate (because fewer victims have guns with which to defend themselves). In a futarchy, we would resolve this question by setting up a prediction market in which people predicted the future crime rate conditional upon gun control passing or failing. Since this would be the most accurate possible assessment of the evidence around guns and crime, we could use it to inform what legislation we wanted to pass.
So according to the conventional wisdom, this is a mildly interesting idea, but it depends a lot upon whether prediction markets can do better than the best of the experts who are already informing the debate on this subject. Nate Silver himself is a good example; he was, by many measures, more accurate than InTrade this election.
I am a big prediction market groupie, and I don't care whether top experts are a little better than prediction markets or vice versa. If you told me that Nate Silver can beat even a highly liquid prediction market by 5%, I would gain a little respect for Nate Silver but continue to push futarchy (government via prediction markets) over argentocracy (government by Nate Silver).
The reason is similar to the reason I (unlike a growing number of rationalists) continue to think democracy is a better system than monarchy, and it was most coherently explained by sci-fi writer/occasional antipope Charlie Stross.
The Mandate of Earth
We tend to think governments in general, and democracy in particular, should be optimized for good decision-making. To argue for democracy along these lines, one might suggest that democracy takes advantage of the wisdom of crowds, or that the population as a whole knows what it wants better than out of touch elites, or that monarchs would make bad decisions because they're corrupt and interested only in their own power.
To argue against democracy along these lines, we might point out that elites are better-educated than the common people and less prone to populist arguments like "let's take resources from small unpopular groups and redistribute it to the majority". One might also just look around at how democratic judgments actually work. As @aristosophy puts it, "it was the 236th year of the reign of what would later be known as King The-American-People the Terrible."
Stross' argument for democracy says it was never intended as a means to optimize policy. It's got a few more modest goals, at which it succeeds admirably.
First, it's supposed to place an upper bound on how terrible a leader can be. America can do some stupid things sometimes, but we would never elect a Stalin, a Pol Pot, or a Kim Jong-Il - whereas military governments and monarchies often do end up with those kinds of people. One might counter-argue that a democracy elected Hitler, but this seems sufficiently explained by the majority of Hitler's badness focusing on minority groups without much voting power - a democracy would have trouble electing someone who was Hitler-level bad towards the average voter.
A democracy never has to worry about the crown prince being a psychotic bastard. It never has to worry about being forced to accept the last leader's feeble-minded son as the successor. I mean, we did it anyway, back in 2000. But we weren't forced to.
But second, and more important, a democracy provides a Schelling point. A Schelling point, recall, is an option which might or might not be the best, but which is not too bad and which everyone agrees on in order to stop fighting. The President might not be the best leader. But he is very clearly the leader.
The importance of this cannot be overstated. The history of the world before democracy was a history of legitimacy squabbles. Some were succession squabbles - the king's psychotic younger son wants to seize the throne from the king's feeble-minded older son, or the Grand Vizier wants to murder the Sultan and start his own dynasty. Others were peasant revolts, where everyone just decides at the same time that they hate the king and decide to have a bloody civil war to overthrow him. Democracies get to avoid that.
In the six hundred fifty years between the Norman Conquest and the neutering of the English monarchy, Wikipedia lists about twenty revolts and civil wars, all the way from the Barons' Wars to the War of the Roses to the English Civil War. In the three hundred years since the neutering of the English monarchy and the switch to a more Parliamentary system, there have been exactly zero.
China is justly hailed as doing much better than the West with this because of their idea of the Mandate of Heaven, but even they collapsed into multiple feuding states around a dozen times in their history, for a total death toll in the tens of millions. Romance of the Three Kingdoms, which I reviewed recently, famously begins: "The empire, divided, seeks to unite; united, seeks to divide". For the vast majority of human history, there was this fatalism that there was going to be a civil war that destroyed your state, it was just a question of whether it happened tomorrow or next century.
In a non-democratic form of government, you're always going to have someone thinking they have more of a right to be in charge than the guy who's there now. In a democracy, the criterion for legitimacy is an objective and easily verifiable one - they got the most votes in an election. If there's any dispute, you can just hold another election. As a Schelling point, it's hard to beat.
Yes, reactionaries, I totally just went there. I just said democracy was better than the Mandate of Heaven, because it promotes stability.
Prediction Markets Are Unimpeachable Experts
Democracy doesn't always perform optimally, but it always performs fairly. There are some biases in particular democracies, like the way the US primaries work, but the general concept of democracy is scrupulously fair, and that is enough to prevent people from starting civil wars.
Academia is different. Its state resembles that of pre-democratic governments, when anyone could choose a side, claim it was legitimate, and then get into endless protracted fights with the partisans of other sides. If you believe ObamaCare will destroy the economy, you will have no trouble finding a prestigious academic who agrees with you. Then all you need to do is accuse the other academics of bias, or cherry-picking, or using the wrong statistical test, or any of the other ways to discredit scientists you don't like (which are, to be fair, quite often true).
A democratic vote among the scientific establishment is insufficient to settle these topics. The most important problem is that it gives massive power to the people who determine who gets to be part of "the scientific establishment". A poll of theologians would establish that God exists; a poll of African Studies professors would establish that affirmative action is effective and morally obligatory; a poll of Sociology professors would establish that capitalism is destroying the country and should be dismantled. Further, exactly which fields are biased in this way is itself a politically charged question: climate change deniers would argue that polling climatologists on global warming is exactly as messed up as polling theologians on God's existence.
This also creates overwhelming pressure for the government or special interest groups to take over scientific establishments. If we consider the intelligence community an "academic establishment" this is what happened during the Iraq War; the petrochemical industry is doing its best to subvert climatology and the pharmaceutical industry has quite a bit of power over medicine. If whether or not a drug worked was decided by a straight vote of all doctors, I bet the pharmaceutical companies would work a lot harder at gaining influence.
So not having any Schelling point - being hopelessly confused about the legitimacy of academic ideas - sucks. But a straight democratic vote of academics would also suck and be potentially unfair.
Prediction markets avoid these problems. There is no question of who the experts are: anyone can invest in a prediction market. There's no question of special interests taking it over; this just distributes free money to more honest investors.
Not only do they escape real bias, but more importantly they escape perceived bias. It is breathtakingly beautiful how impossible it is to rail that a prediction market is the tool of the liberal media or whatever. You just tell Limbaugh: "Wait, you think the prediction market has a consistently liberal bias? Then invest on the conservative sides of issues and you get free money for having discovered this startling economic fact!" If Limbaugh invests his fortune and turns out to be right, he's laughing all the way to the bank and improving the system. If Limbaugh claims the market is biased but refuses to invest it in, everyone knows he's just spouting hot air.
Nate Silver might do better than a prediction market, I don't know. But Nate Silver is not a Schelling point. Nobody chose him as Official Statistics Guy via a fair process. And if someone objected to his beliefs, they could accuse him of bias and he would have no recourse until it was too late.
If a prediction market is almost as good as Nate, and it is also unbiased and impossible to accuse of bias, we have our Schelling point. Barack Obama can say something like "Obamacare won't be unaffordable, in fact it will cut the size of the budget deficit!" And if Rush Limbaugh says "You're lying, or relying on data collected by liberal hacks", Obama can just retort "No, seriously, the prediction market says there's an 80% chance I'm right", and Limbaugh will just have to admit he's right and slink away.
Just as democracy made it harder to fight over leadership, prediction markets make it harder to fight over beliefs. We can still fight over values, of course - if you hate teenagers having sex, and I don't care about it, we can debate that all day long. But if we want to know whether a certain law will raise the pregnancy rate, there will be only one correct answer, and it will only be a mouse-click away.
I think this would have more positive effects than anyone anticipates. If people took it seriously, not only would the gun control debate be over in an hour, but it would end on the objectively right side, whichever side that was. If single-payer would be better than Obamacare, we could implement single-payer and anyone who tried to make up horror stories about how it would destroy health care would be laughed out of the room. And once these issues have gone away, maybe we can reach the point where half the country stops hating the other half because of disagreements which are largely over factual issues.
Right now we're going backward from this future. A prediction market has to be very liquid (ie have many users spending lots of time on it) before it becomes any good at predicting things. But the US government just cracked down on the largest prediction market, InTrade, because they classify it as "online gambling". This has much reduced its liquidity and set the entire field back by years. IARPA, a government intelligence agency thing, has a toy prediction market going, but it's much more limited without real money.
I hope that someone soon starts a bitcoin prediction market outside the US government's reach. It might fail - prediction market users and bitcoin users are both small minorities, and the disjunction of two small minorities might be too small to provide the necessary liquidity - but maybe later when dollar-bitcoin convertibility becomes more fluid, its time will come. This is the sort of idea I would totally pursue myself if I had money, time, technical knowledge, business acumen, entrepreneurial spirit, legal advice, and about twenty other abstract and concrete resources I do not possess. As it is I sort of fantasize about making enough money in medicine to fund someone who has the other nineteen. |
|
|
squid314
