LISTEN TO ARTICLE
SHARE THIS ARTICLE
With the onset of an historic U.S. presidential election, a quantitative trading firm has been trying to pique hedge funds’ interest in casting big bets in political-prediction markets.
Susquehanna International Group, based in Bala Cynwyd, Pa., is a big player in options and the exchange-traded fund market, but its founders have long had an interest in the statistical side of gambling. The firm runs its own blog on gaming and poker strategy tips and has been building up a sports-betting operation in Ireland. It’s recently gauged funds’ interest in bets on election outcomes. Although few have gotten involved, Susquehanna is willing to take the other side of wagers on the presidential race for up to $100 million per bet, says a person familiar with the firm, speaking on the condition of anonymity because the plans are not public.
Funds wouldn’t be able to place bets from the U.S., where the practice is banned in most cases, and would have to consult their lawyers to ensure they’re in the clear. Susquehanna isn’t offering a betting service or product: Funds would place wagers at licensed betting platforms in the U.K., which would look for a counterparty to take them up. Susquehanna’s Dublin affiliate could then step in, essentially acting as a behind-the-scenes market maker for the bets.
Investors already express views on political outcomes in the securities they buy. Strategists at JPMorgan Chase & Co., for example, have created a“Trump basket” and a “Biden basket” of stocks expected to do well under each future administration. But traders are often wrong about what elections really mean for stocks. In 2016 futures contracts on U.S. stock indexes plummeted on election night as Donald Trump’s victory became clear. The next day the S&P 500 rallied. So if you want to make a bet, why not just wager directly on the outcome of the election? “It’s just another thing to add to the portfolio,” says Paul Krishnamurty, a professional gambler and freelance political markets analyst for Betfair, a prediction exchange with headquarters in London. “It’s just another hedge.”
A trickier question is what risk you’d actually be hedging. If it’s hard to predict how an election will affect markets, then it’s not clear how a winning political bet might offset a loss elsewhere. That’s one reason the U.S. Commodity Futures Trading Commission in 2012 stopped a company from listing derivatives contracts tied to elections. The CFTC also said such markets could damage the integrity of elections, for example by giving people a direct financial interest in voting for a candidate they wouldn’t otherwise support.
Political betting also faces more mundane objections: Even where it’s legal, fees are often high, and markets tend to be thinly traded, without much opportunity to rake in big winnings—at least in Wall Street terms. Betfair has matched about $266 million in trades on who will win the presidential race.
Markets to bet on U.S. presidential elections operated from 1868 to 1940, according to research by Paul Rhode of the University of Michigan and Koleman Strumpf at Wake Forest University. New York was a hub for it, especially the informal Curb Exchange, which later became the American Stock Exchange. Sometimes, for brief periods, betting on politics would exceed stock and bond trading. Political markets faded as scientific polling ascended and other types of gambling became more accessible.
An online market calledPredictIt has carved out a legal niche in the U.S., getting an exemption from the CFTC to run a political-betting exchange for academic research purposes. (Iowa Electronic Markets received a similar exemption.) The agreement caps the size of any individual bet on PredictIt to $850. Participants buy “shares” in possible election outcomes, so prices reflect their predictions. PredictIt has also offered markets on how many times Trump would tweet in a week, and is taking bets on when the election’s result will be called. Nov. 4—the day after Election Day—is in the lead.
Thomas Miller, faculty director of the data science program at Northwestern University, says prediction markets may pick up information polls miss. They’re faster, and people may be more honest about what they think. Miller created a prediction engine based on betting that runs a million hypotheticals on the presidential election every hour. On Oct. 28 it showed Biden with a 76% chance of winning.
Source: Read Full Article