What Are Prediction Markets?
Prediction markets leverage the power of the crowd to predict the outcome of future events. Presidential elections or primaries winners, the likelihood of natural disasters or terrorist attacks, the future price of bitcoin or even the outcome of a sporting event, are set up as contracts which expire at either zero or usually $1.00. The price of each contract depends on the probability that such an outcome will occur based on crowd betting behaviour, with likelier events carrying a price closer to $1.00. For instance, leading prediction market PredictIt has Trump as the favorite to win the New Hampshire primary, what a share price currently at $0.40. If Trump does indeed go on to win the New Hampshire primary, those who hold a Trump contract will earn $1.00. If he loses, the Trump contract will be worthless.
What Makes Prediction Markets Valuable?
Prediction markets have at times demonstrated greater accuracy than traditional polling in predicting future events. For instance, the University of Iowa’s not-for-profit Iowa Electronic Markets (IEM) has a lower average margin for error on presidential elections since 1988 than the final polls, by approximately 25%.
While presidential elections have been the most widely used predictive market, there are applications in business as well. From a macro perspective, predictive markets can give insight as to future trends, which could impact company strategy. More specifically, companies have used predictive markets to gain greater insight into how future products might perform, and empirically predictive markets have demonstrated greater “accuracy” than standard opinion polls or surveys. There are a number of possible reasons for this, but most importantly in our opinion is that the win/loss dynamic creates a self-selection process by which only people with real insight into a particular issue will participate. And not only do they seem to be more accurate, predictive markets are cheaper and easier to set up than standard surveys.
Most interesting, at least from our perspective, is the disruptive potential predictive markets could have in the future in the insurance markets. Imagine, for instance, the ability to buy contracts predicting natural disasters. Rather than purchase insurance from brokers, home or business owners could leverage the predictive markets, predicting yes, there will be an earthquake within the next year that will cause substantial damage. Theoretically, the lack of overhead, combined with the predictive power of the masses, would lead to lower and more accurate pricing.
Why Predictive Markets are a Good Fit for Cryptocurrencies
Regrettably, predictive markets have been termed gambling in the United States, resulting in most of the exchanges exiting the market. Notable exceptions are PredictIt and IEM, both of whom have received no-action letters from the SEC, which basically state that no enforcement action will be taken. While these two markets have achieved excellent liquidity levels, they have limited their topics for predictions to political topics and a smattering of financial events. Intrade, on the other hand, left the US in 2012 due to the sudden refusal of banks and credit cards to process transactions, as well as fear on the part of executives as to legal action against them. Shortly thereafter they shuttered their operations entirely – a stunningly fast reversal from market leader to out of business.
Bitcoin offers a solution to the former issue – lack of willing payment processors – and there are a number of different centralized products attempting to reintroduce prediction markets using bitcoin instead of credit cards or bank transfers. However, many within the cryptocurrency arena see potential in a decentralized prediction market platform to circumvent government censorship. Without an address to which the government may apply pressure, the prediction markets are immune to legal pressures, similar to bitcoin itself. But how could such a market operate autonomously? Probabilities, of course, are easily adjusted automatically based on the volume of users predicting one outcome or another. However, a decentralized prediction market gives rise to a pretty tricky question: who can be trusted to accurately report the result of any given prediction contract? Certainly not the users themselves!
Essentially, in order to facilitate accurate reporting of events the decentralized market must ensure greater incentive in telling the truth. We will delve into specific platforms elsewhere, but basically the method used, with slight alteration, is to “equitize” reputation, which is given and taken away based on an individual’s history of honesty. Reputation is used to report on the outcome of events, with truthful reporting leading to greater amounts of this equitized reputation, and lies resulting in reputation being lost. Reporting is done anonymously, meaning that nobody is able to see how others are reporting, which works to incentivize truth – if everyone else is telling the truth and you are lying, you will simply lose reputation and the truth will out in any event!