On Tuesday night, the UMA Project community approved contracts that allowed creating its first token: ETHBTC. This is a synthetic token whose value tracks the relative value of ETH to BTC, so if ETH is worth $200 and BTC is worth $10,000, an ETHBTC should be worth $0.02.
The intriguing thing about ETHBTC, though? No ETH or BTC is needed to make it.
This will be the first deployment of what UMA, a decentralized finance (DeFi) project, calls the priceless token model, one built from the start to minimize the need for oracles.
“ETHBTC was selected as the first test for UMA’s priceless synthetic design because it’s DeFi-centric but not too serious,” Hart Lambur, UMA’s co-founder, told CoinDesk in an email. “This first token is still experimental, so it felt wise to choose a product that appeals to hardcore DeFi natives – the type of people that might want to bet on this rate, and who best understand the risks of ‘new’ things.”
ETHBTC is available now on the new Uniswap, though the team is warning interested buyers: “The mechanisms behind this design have not been proven in the wild. Users should proceed with extreme caution.”
Representing BTC on Ethereum to prove its value as collateral has become a prominent theme for DeFi in 2020.
Below is an explanation of how ETHBTC is created, and it should basically describe how any other synthetic tokens might be created on UMA, though some of the variables will likely change.
To generate ETHBTC, a user posts dai as collateral to the smart contract. Based on the collateralization rate of 120%, the contract will allow the user to generate a specific amount of ETHBTC. They can then sell the new ETHBTC on the open market or they can use it to add liquidity to the ETHBTC pool that will be created on Uniswap (most interested buyers will probably choose to just acquire it directly on Uniswap).
UMA’s synthetic tokens trade like any Ethereum-based token until their contract comes to an end. At that moment, the staked DAI will be split between token holders and stakers. If the value of ETH vs. BTC has gone up at the close of the contract, the token holder will get a profit on what they paid for it. If it hasn’t, the staker will earn a profit on that original sale as the contract releases more dai back to them.
So token holders are long and stakers are short ETHBTC. Lambur described it as “a sort of meta-bet on DeFi as a whole,” because the most likely explanation for growth in ETH uncorrelated to BTC would be more people using DeFi products.
This is all quite new. The UMA team noted in their message that this is very much an alpha test in the real world. While it has been audited by OpenZeppelin, users should be very cautious about the amount of risk they take on.
“We strongly encourage interested users to do their own research and proceed with caution in this experiment,” Lambur wrote.
Under the priceless token model, UMA does not need an oracle to function on a day to day basis.
“What we’re saying is: Let’s not do any on-chain price ever,” Lambur told CoinDesk in an interview. “This is how you’re going to have to scale DeFi,” he added later.
The idea here is that everyone knows that the contract is going to have this defining moment when it comes to the end and the stakes get split up between stakers and coin holders. If the definition of the price is clear and transparent to everyone, the truth of the world should not be confusing at that moment. If so, then an oracle will never be necessary. People will just see what the truth was and accept that outcome.
“Minimizing the dependency you have on your oracles is just good system design,” Nik Kunkel, on the oracles team at MakerDAO, told CoinDesk. “This type of oracle’less design is very unique to their system and the characteristics of the UMA system. It can’t really be applied anywhere else.”
This was a point that Sergey Nazarov, creator of Chainlink, a network of oracles, also emphasized.
“If you say, ‘I’m not going to build data feeds to build financial products,’ the number of financial products you can build is very small,” Nazarov said. “I think what they are doing is essentially an interesting experiment.”
That said, Lambur compared UMA’s approach to paper contracts in the real world. Traditional contracts don’t have to be publicly posted to function and most of the time no one but the parties ever see them because every one honors their side of the deal.
“We are really trying to frame the oracle itself as being like taking someone to court,” Lambur said.
In order to backstop price throughout the life of the contract, UMA also has a liquidation model. If anyone spots an undercollateralized position, they can trigger a liquidation event. Again, if they initiated it accurately under the conditions, ostensibly there will be no need to turn to oracles.
If there’s a dispute, UMA token holders will settle it by coming together to vote. The holders who voted on the winning side will be rewarded in new token emissions. The economic model of UMA is designed so that it will always be unprofitable to buy up UMA tokens in order to vote through a false choice.
“The tokens themselves, the token holders, form this court system which ultimately is the security of the whole platform,” Lambur said. “The whole overall premise for our token economics is we need the cost of bribing the system to be greater than the value of the system.”