PredictIt: Close the prediction market

ive here. Rajiv Sethi focuses on the possibility that PredictIt will have to terminate operations, and the chaos and gameplay opportunities that will result from having to terminate contracts that expire after PredictIt terminates operations. He only mentions in passing that this event will also shut down the last important online election forecasting market. There is as much enthusiasm for forecasting markets as the priests, but it may be because of the disastrous state of opinion polls as the supposed virtues of prediction markets (the pollsters have great difficulty getting anyone to answer the questions, and their contact lists skewed toward the landlines, as in old farts).

Prediction markets work best when they represent voters well. The reason the Brexit betting markets were so wrong (expect the remaining six-point win the night before the vote) is that in gambling, “votes” are weighted in dollars, or here in pounds sterling, with the total bet per side. But the actual number of bettors favoring Brexit was much greater than the stay-on bettors!

As Barry Ritholtz wrote in 2008:

Over the years, I have been a critic of the prediction and futures markets. In particular, the specific ways in which some parties abuse it (ie the policy).

However, I am a firm believer that markets can generate valuable economic and investment data that can be very useful when handled appropriately…

In each of these cases, we rely on deep and diversified markets and trade in dollar volumes measured in the trillions. This is part of my criticism of the futures markets: they are weak, volumes are weak, and the dollar amounts at risk are very small. Thus, these markets are prone to failure at times.

In fact, the excuse-making for the futures market failure began almost as soon as the New Hampshire primaries were over (see Why online prediction markets are blowing up New Hampshire, forecast markets with comfort tools that feed on advanced indicators, etc.)

None of these really looked at what makes the futures markets work, how they might fail, and what their strengths and weaknesses are. This morning, I want to delve into some of the problems these markets can face, why they fail, and what value they provide.

First, let’s start by reviewing some of the more exciting market failures to forecast:

• Iowa Elementary 2004 (Howard Dean vs. Everyone Else)

• The Republican Party retained the Senate in 2006

• New Hampshire Primary 2008 (Obama vs. Hillary)

On politics, I agree with Dan Gross, who wrote “These are less future markets than those of the recent past.” In other words, bettors are essentially compiling polling data that already exists – rather than predicting the unknown future.

When it comes to politics, it’s not as if the Iowa online marketplace or the Intrade gamblers have any special insight or inside knowledge. They don’t know anything—individually or collectively—that the rest of the public (insiders included) don’t already know. They all read the same newspapers, blogs, polls, etc., and respond to any broad narrative that happens on that day or week. They respond just like any other focus group, from the same information that voters have.

I sometimes think of the political futures markets as a focus group in their own right. Here’s where things get really interesting: when the group is underrepresentative of their target market, they misrepresent it alarmingly. In fact, the closer the merchants as a group are to the target decision makers/voters, the better their track record.

Written by Rajiv Sethi, Professor of Economics, Barnard College, Columbia University. Posted via From “Incomplete Information”

Market Predictions has been operating under a no-action letter from the CFTC since 2014. This permission has now been withdrawn, and the exchange has been ordered to liquidate all open positions by February 15 of next year:

The Market Oversight Division (DMO) of the Commodity Futures Trading Commission (DMO) announced today that it will withdraw CFTC Letter 14-130 effective immediately. When the DMO issued the letter on October 29, 2014, it took the position of taking no action in connection with the operation of a non-profit marketplace for certain event contracts and the offer of such contracts to American persons by the Victoria University of Wellington, New Zealand… The DMO has determined that the University of Victoria has not managed its market in accordance with the terms of the letter and as a result has withdrawn it… all relevant and outstanding listed contracts and positions that include all open interest associated with this market must be closed and/or liquidated no later than 11:59 PM (EST) on February 15, 2023.

Permission to act was based on meeting nine conditions, one or more of which the CFTC claims has been violated. It was not clear exactly what the nature of the violation was, but the conditions for the continuation of the operation were as follows:

The DMO granted relief based on the university’s assurances that the proposed event holding market would:

  1. be small in size and not-for-profit;
  2. It is operated for academic and research purposes only;
  3. To be supervised by university faculty, without separate compensation;
  4. Offering event contracts consisting of two sub-markets for binary options contracts related to political election results and economic indicators;
  5. Limited to 5000 traders per contract, with an investment limit of $850 per participant in any contract;
  6. does not provide brokerage services or charge commissions to participants;
  7. engage a third-party service provider to perform know-your-customer due diligence (“know-your-customer”) on its participants;
  8. Charge only fees necessary to cover KYC implementation, regulatory compliance, and basic expenses for operating the proposed event contract marketplace; And the
  9. Restrict advertising to media outlets where there is a high potential to reach those interested in the subject matter of the event contracts, provided that such advertisement clearly discloses that the platform is not regulated, experimental and is operated for academic purposes.

Predict asserts that it has not violated any of these Terms. In an email to traders, the exchange argues that it “disagrees with the Commission’s decision and affirms that all active markets are not only within the terms of the no-action letter but also comply with the commission interpretations that have been communicated to us over the past eight years.”

If the stock exchange is really going to close in a few months, the important unresolved question is about resolving the markets that point to the post-February 2023 events. And there are many of those, with very active trading in the two major party presidential nomination markets, and the results of the 2024 presidential election. .

For example, consider the Democratic nomination market, where Gavin Newsom is (for some reason) trading at 20 cents to the dollar, behind Joe Biden but ahead of Kamala Harris. The market allocates the probability that one in five will be the candidate. Here is the price history in this decade for the past 90 days:

There was a huge spike in trading on June 14, with around 40,000 contracts trading at 5 cents a piece. Someone who bought 10,000 contracts on that date paid $500, and those contracts are now worth $2,000. But since the event will not be resolved by February 2023, what will happen to this situation?

The exchange will have to figure out some way to liquidate these contracts before the listed termination date. One possibility that can be immediately ruled out is the reversal of all transactions in such markets. This reversal will essentially cancel all trades, so the (hypothetical) trader in question will get a refund of $500 and the contracts will disappear from his account. This is not possible – for example, this trader could sell all contracts at different prices in the meantime, withdraw all the funds, and close his account.

So what can the stock exchange do? It can stop trading at some point (with or without notice) and liquidate all contracts at the price of the last trade. But there are two problems with this. It can lead to people trying to manipulate the price of the last transaction in ways that will lead to a solution more favorable to them. This can result in some very strange prices – for example, the sum of all contracts in the filter markets can exceed 100 cents to the dollar. This happens anyway from time to time, due to the fee structure of the exchange, but the effect will be amplified exponentially as people try to manipulate the system to their advantage. The usual arbitrage-based mechanisms that keep prices roughly consistent with probability will be weakened since there is no guarantee that the final decision will meet the laws of probability.

So there’s a real problem here, and it’s not clear how the exchange will deal with it. Given the uncertainty, I would expect there will be a spike in withdrawals, and people with virtual profits will now be liquidated and cash out. But others may see opportunity in the midst of a frantic and chaotic trade. When Intrade abruptly shut down nine years ago, accounts were frozen and withdrawals halted. I expect the shutdown this time to be more orderly.

But here’s the thing – with PredictIt gone there will be no comprehensive market-based election forecasting system in the US. Kalshi is CFTC approved and regulated but has not yet entered the field. Iowa’s leading electronic marketplaces are still operating but with very limited size and coverage. Model-based forecasters like FiveThirtyEight exist, of course, but it would be a little difficult to assess their performance relative to the markets. So, if there is a rationale for closing this exchange, I hope the CFTC will clarify it at some point.

In a previous post, I explained that prediction markets are unusual among online platforms in facilitating a diversity of perspectives, by creating strong incentives for people who disagree with one another to interact:

People actively seek information that largely confirms their current beliefs, and social media platforms are accommodating and increasing this demand. Predictive markets play an interesting and usual role in this environment. They encourage people with opposing viewpoints to interact with each other in anonymous, credible, nonviolent ways. Meaning it is the opposite of echo chambers. A market with homogeneous beliefs will not have a trading volume, or it may attract those with different opinions who are attracted to what they see as wrongly priced contracts.

While most online platforms facilitate and deepen ideological separation, prediction markets do just the opposite. It provides financial reinforcements to those who get it right, and forces others to question their own assumptions and inclinations. Although they are known as mechanisms for generating expectations through the wisdom of the masses, they also bring opposing views of the world into direct and subordinate contact with each other. This is a useful function in an increasingly separate digital ecosystem.

In fact, there is a real sense that Predict gave us an advance warning about January 6. Roughly a month into the 2020 election, with a decisive number of states ratifying their vote totals, the exchange was still giving the incumbent one vote. Eight chance to win re-election. The common explanation at the time was that the markets had deteriorated. The fact is that there was information, most of us had not seen, that there was a big surprise in the store.

Easy print, PDF and email

Related Posts

Leave a Reply

Your email address will not be published.