People can foresee the future only when it coincides with their own wishes … the most grossly obvious facts can be ignored when they are unwelcome.
George Orwell
This Orwell quote about predictions is widely known. But it is the perfect foundation for why prediction markets, specifically, are so interesting. So let’s get into it.
🔮🔮🔮
In recent years Tyler Cowen has been popularizing the question, “But are you long volatility?”1 One place this line of questioning is used is when someone is very enthusiastically voicing their opinion. This often occurs when someone has a pessimistic outlook on a particular topic.
The gist of the question being, “Well, if you are so confident this is going to be a disaster, have you put your money where your mouth is?” As Cowen will tell you, if you ask people this question, a common answer is, “No, but I will be!”
People, on average, will not follow through with that promise. If they felt so confident, they would be so compelled to place that “bet” that they wouldn’t have time to voice this hollow opinion.
🔮🔮🔮
While the prediction market dates back hundreds of years, a little research will bring you to Hayek as one of the first economists to opine on the potential value of the mechanism. It is not too far a stretch to see how, if you can predict the future, you can create value.
Google was famously using prediction markets inside the company in the early 2000s. The markets were used for product launch dates, new office openings, and “many other things of strategic importance” according to the Google blog.
I find one issue with the way in which Google used prediction markets. They did not require an actual payment outlay to participate. While you can understand why it would be a nonstarter to have your employees actually outlay money to participate, in fact, I would call that a fatal flaw in the design.
A prediction, just like an opinion, is made more accurate when the predictor has skin in the game.
The outcome of a proper prediction market is a vehicle that can better allow different parties to profit from their viewpoints. In other words, one can get long that volatility aforementioned by Cowen in both more direct and more creative ways than without such a mechanism.
🔮🔮🔮
Jeff Yass, co-founder of Susquehanna, has spoken publicly about (a) his interest in prediction markets2, and (b) specifically how he would suggest the U.S. should embrace them.
In this WSJ opinion piece, Yass discloses that in the days after November 3, 2020 he wagered $500M to win just $50M3 that President Biden would indeed be the next POTUS. That wager is relatively black and white in outcome — either Biden or Trump would end up in the oval office.4 Where the gray areas overtake the black and white is where betting markets could actually be most helpful.
While political election outcomes are a simple example of a market, political policies could also be put to the market. The WSJ piece gives the glaring example of war:
On so many questions important to the public, predictions and forecasts are thrown around with little accountability. He points to the Lincoln War Department’s turning away of volunteers in 1861, saying the war would be short. Some 160 years later, Bush White House economist Lawrence Lindsey would be taken to the woodshed for predicting the Iraq War might cost $50 billion. (The actual cost turned out to be over $1 trillion.)
Asks Mr. Yass: If betting markets had anticipated even $500 billion, “would we have gone to war in Iraq?”
The theory being that politicians would have to explain why their promises stray from investor expectations.
I pressure tested this with David Epstein and Annie Duke during a recent fireside chat.
Duke did think there is a possibility prediction markets could make a difference in politics, just as Yass contemplates, and used California’s mess of high speed rail as an example:
In particular where [prediction markets] could help is politicians do have their careers at risk so one of the reasons why politicians like in the California bullet train for example … say, “I don’t want to waste taxpayer money, we have to finish this thing” is because they are afraid they are going to get voted out of office. So, if voters were to start to pay attention to prediction markets, they also might be more rationale about saying, “No, actually, thinking about that $100M you are about to spend of taxpayer money … I just think you could probably give me a lot more benefit with that money…”
🔮🔮🔮
While I can’t say I believe prediction markets will indeed become more mainstream, I can summarize why I think that would be a great outcome.
A public, liquid market could be an important tool where participants can “match wits” while contributing to public discourse on important topics such as political or economic policies.
In this vision of the world, you can profit from your highest conviction opinions rather the spray and pray approach of, almost always, pessimistic verbal diarrhea.5 After all, opinions are like assholes. Everyone has one and it’s often best they not be heard.