“You say prediction market and people are like ‘What’s that?’ – it doesn’t sound that important or lucrative or interesting.”
That’s Jeremy Gardner, best known as the director of the College Cryptocurrency Network, talking about the first reactions some may have to his latest project, Augur, the decentralized, blockchain-based prediction market founded by Jack Peterson and set to launch a token crowdsale this spring.
Common perceptions aside, Gardner and core developer Joey Krug are out to prove that a platform that allows participants to bet on the outcome of real-world events can be a vehicle for social good, rather than simply another outlet for digital gambling.
In conversation with CoinDesk, Gardner and Krug sought to portray Augur as a project that plans to use decentralized public ledgers to create a way for anyone in any field, from finance, healthcare and governance, to tap into the collective forecasting power of a global user base.
“With bitcoin, it’s a better money than what we have, but it’s a version of what we have today. We’ve never really had a wildly popular prediction market. We’ve never had these open-source tools to be able to bet on the Internet and turn that into a social good.”
Both Gardner and Krug have left college to launch the six-person project, one that boasts financial support from Ethereum creator Vitalik Buterin and the guidance of Ron Bernstein, founder of Intrade, what was until the time of its shutdown in 2013 one of the more well-known prediction markets.
Speaking to CoinDesk, Bernstein said he expressed an early interest in the project as he views it through the lens of his own struggles trying to launch a similar, though centralized, platform.
The Intrade and Tradesports CEO compared the challenges the project faces to tech giants like Uber, which have tested regulatory boundaries in order to ensure they are first to market with a potentially powerful idea.
“The potential for Augur is really about the distributed participation from the exchange operators as well as from the exchange users and the ability to distribute that in the way that they propose has never been done,” Bernstein said.
A tumultuous history
At present, there are more immediate hurdles for Augur. One of which is ensuring that the public can get behind a prediction market, especially when their rocky history has clouded public perception about what they are and what they aim to achieve.
Prediction markets allow their users to buy and sell shares in the outcome of an event. The current market price of a share is then an estimate of the probability that event will occur. Already, many academic researchers attest that such platforms, while incorporating aspects of gambling, do have practical value.
“Prediction market prices have informational value because they aggregate the beliefs of market participants and reveal what the market overall forecasts are the odds of the event at hand occurring,” a 2014 report by Mercatus Research explains.
Though it goes on to note that primary reason the market exists is speculation, the report does speak to the positive benefits such a platform can have by noting its use in US presidential elections dating back to the 1800s.
Gardner attests to being an Intrade user, placing bets during the 2008 and 2012 elections. He explained the appeal of the service, adding:
“It wasn’t gambling, there was no risk if you thought you were well-informed and that to me was a really powerful notion.”
Finding a gray area
Just as powerful a roadblock may be historical issues prediction markets have faced on a regulatory front, particularly from the Commodity Futures Trading Commission (CFTC), the body that oversees the US futures market.
In 2005, for example, the CFTC granted Intrade an exemption to operate, providing it limited participants to those with assets between $5m and $10m. By 2013, dogged by lawsuits in the US and issues abroad, Intrade shut down. Notably, its website currently promises “announcements coming soon”.
Further laws were enacted as part of the Dodd-Frank financial reform law, which sought to limit betting that was not deemed to be in the public interest. However, some prediction markets have been able to prosper, most, like the Iowa Electronic Market, under academic exemptions.
Yet these problems, Augur contends, are solved by the decentralized nature of blockchains and the overall lack of regulatory guidance.
“There’s no law that prevents us from doing what we’re doing. We’re just writing code. There’s also no law that allows us to do what we’re doing. We recognize that it’s a gray area, but it’s a gray area we’re very much comfortable with,” Gardner said, adding:
“Very clearly we write this code, we put it into the cloud and like Satoshi [Nakamoto] we could disappear.”
Gardner discussed conducting the company’s in-house legal research prior to the launch of the venture, asserting that the project should not fall under securities laws as “people aren’t partaking in a common enterprise and they’re not expecting profits from the work of others”.
Still, he admits the CFTC is “a huge concern”, though one that will be alleviated by the fact that no one involved with the project will operate a prediction market.
“It’s about narrowing down who we might piss off and reducing the likelihood of that happening,” Gardner said.
Harnessing blockchain tech
Questions of market viability aside, Augur is already at work on its technology, though even its construction remains in flux. Augur intends to use two types of tokens to facilitate its market.
First, it will plan to use the as-yet-unlaunched sidechains technology to transfer bitcoin to addresses that can be held while a user is actively holding shares as part of a wager. Second will be the token of focus during its upcoming crowdsale, Reputation, which is set to be sold over 45 days in May.
Reputation tokens will be sold during the crowdsale in part to raise money for the completion and continued development of the platform, but be used by Augur in practice to incentivize users to reliably report on the outcome of events, say that Hillary Clinton is elected president in 2016.
“After the election happens, because there’s not a centralized source that confirms that it’s happened, there has to be a decentralized reporting system. That’s where reputation comes in. Reputation holders are asked to report on the outcome of events and that ensures the integrity of the system,” Gardner said.
Those who are dishonest, in turn, have a percentage of their tokens redistributed among the trustworthy users in the network. How much will be distributed is determined by an equation, and will vary depending on factors such as how many people report on the event and how accurately.
Lastly, there will be a ‘seigniorage modeled coin’ that will help ensure that long-term bets can be made without being subject to the volatility of bitcoin. For example, if a user wanted to bet on the 2020 election, he or she would need to know the funds would retain value at this time.
“What we’ll probably do,” Krug continued, “is there is a thing called the subcurrency API on Ethereum, and we’ll probably support the best seigniorage out there, so we’ll probably use the best of them.”
Already in place, according to Krug, is the equation that would allow for the exchange of shares in event outcomes, and the consensus mechanisms that determine binary outcomes.
“You can even have up to 75% liars and get the correct outcome reporting,” Krug said, though he noted the project is still working on how to deal with multidimensional outcomes.
Connecting with potential users will be another challenge, as evidenced by the somewhat lackluster reception of the group at The North American Bitcoin Conference’s Startup Stage.
Recalling the event, Gardner remarked: “I think there is a huge gap that keeps people from understanding the project.”
Focusing on this question will be the job of director of marketing Tony Sakich, a former marketing manager at payment processor BitPay.
Describing the project’s members as a “dream team”, Sakich indicated that he intends to focus his efforts on keeping Augur’s ad campaigns “old-school”, while appealing to a broad base beyond those in the bitcoin community.
“We’re hoping to go to a lot of a schools, academia has wanted something like this for a while. I think the people who are into prediction markets are not the crypto types, and we’re trying to make that crossover,” Sakich said.
Calling the project a “real challenge” when compared to BitPay, Sakich said he’s in the midst of lining up conferences and events that will allow Augur to reach audiences outside of the bitcoin community.
Although he said he won’t be using the betting platform, he added he expects Augur to “become one of the definitive prediction markets”, provided it can be maintained by its decentralized community.
Despite these challenges, the Augur team is confident their product can solve real problems.
For example, Krug mentions that Augur could be used by farmers in Argentina to hedge against weather cycles or by Chinese traders who are unable to access the US stock market.
“Take the problem of asking your own employees for information,” Krug continued. “Say you’re a fashion company and you want to see what will be the best-selling product of that year. You can ask your employees. What you could do is ask the general public and get a much more accurate result.”
On an even more abstract level, Krug went so far as to suggest that prediction markets could be used by doctors to more accurately diagnose patients.
Perhaps most notably, Krug and Gardner aren’t interested in turning Augur into a long-term endeavor or even a business, or at the very least, that matter is still up for debate.
“A lot of team members are pretty young, this is really an intellectual experiment. Let’s say that Augur is successful, there’s all sorts of consumer applications, there will be consulting opportunities,” Gardner said, comparing the career path to those blazed by bitcoin developers like Peter Todd.
“If Augur is successful, there’s a huge opportunity to make money, but we believe that if this succeeds, we’ll have created this world-changing forecasting tool.”
The full white paper for the project can be found here.