The value of the digital currency Bitcoin has soared in recent months, but can the technology also help make predictions? A research team led by Princeton faculty is designing a prediction market — in which parties predict the likelihood of future events — underpinned by Bitcoin transactions.
Prediction markets function like stock markets, with participants trading shares. These shares have payoffs that are tied to the outcome of real-world events. For instance, a share may pay out $1 if a particular candidate wins the next presidential election.
The equilibrium price of these shares indicates the probability, on average, that the market ascribes to the occurrence of the event in question; if Obama’s 2012 re-election shares were priced at $0.50, then the market theoretically believed Obama had no more than a 50 percent chance of winning re-election.
Intrade.com, one of the most popular public prediction markets, was blocked by the U.S. Commodity Futures Trading Commission as an illicit form of gambling on Nov. 12, the Wall Street Journal reported. The CFTC forbade American users from funding Intrade purchases through U.S. banks.
Computer science professor Arvind Narayanan, one of the professors leading the project, said he believes Bitcoin provides an opportunity to decentralize prediction markets from external oversight.
Unlike traditional currencies, Bitcoin is not regulated by any bank or government.
“Now that we have this beautiful decentralized system that allows two parties to transact with each other without a central authority, can we make it so that arbitration of events can be decentralized in some form?” Narayanan said.
“I think we’re converging on a successful design, but it’s too early to tell,” he continued. He declined to provide more details on their research until it is published. Mike Hearn, a Bitcoin developer who has worked with the currency since 2009, agreed that it will have the ability to decentralize trading systems.
“Making something decentralized can be a way to avoid these regulatory hassles in a way,” Hearn said. “But it’s not a silver bullet: De-centralized systems can be regulated in other ways.”
As an open-source technology, developers building instruments like prediction markets and payment systems do so for free in their spare time. Hearn cautioned against the issues inherent in developing financial software and the problem of funding development for a technology not regulated by a company.
“One of the biggest questions with a project like this is: How do you fund software development?” he said. “People usually say, ‘I’ll make a company, and then I’ll raise money,’ but at that point it’s not really decentralized anymore.”
Much of Bitcoin’s recent success hinges on its mysterious origins. The technology’s secretive founder, or founders, published work under the pseudonym Satoshi Nakamoto and ceased involvement completely in the middle of 2010.
Hearn was one of the few early adopters who communicated directly with Nakamoto during the early stages of Bitcoin development.
“Bitcoin sidestepped [the problem of centralization] by having this mysterious creator who then disappeared,” Hearn said. “He obviously did a huge amount of work in what was presumably his spare time to create this thing for free. That’s an amazing thing that doesn’t happen very often.”
Today, Bitcoin’s popularity reaches far beyond prediction markets. The exchange rate for a single Bitcoin, valued at less than a dollar two years ago, peaked at over $1,200 late last year. Experts remain divided on the currency’s long-term success.
Narayanan suggests using the current Bitcoin exchange rate as a proxy for predicting whether the currency will succeed, as it indicates investors’ belief in the currency’s probability of success.
Were Bitcoin to become a mainstream currency, its value would likely increase ten- or twenty-fold, he explained. Given that the current exchange rate hovers around $1,000 per Bitcoin, current investors cap the probability of that kind of success at no more than 1 or 2 percent. Bitcoin is now accepted by many retailers, including Virgin Galactic and OKCupid.
“Is it going to become a mainstream currency? That’s hard to answer,” Narayanan said. “We know that there is some technological soundness to the system, but it’s also unlikely that some of the more ambitious speculations will come true.”
Although Hearn said it would be difficult to “compete against PayPal and credit cards,” he said he believed the currency could gain a foothold in areas not currently served by current markets, such as prediction markets.
Narayanan said he and the rest of the team hope prediction markets can be one of the first models of that kind of success in niche markets.
Other leaders of the project include computer science professor Ed Felten, Andrew Miller of the University of Maryland and Jeremy Clark of Concordia University.