Buterin: How to create an algorithmic stablecoin that won't turn into a Ponzi scheme or crash

Ethereum co-founder Vitalik Buterin shared two thought experiments on how to assess whether an algorithmic stablecoin is sustainable.

Buterin’s comments come amid the recent collapse of the Terra (LUNA) ecosystem and its algorithmic stablecoin, TerraUSD (UST), which caused billions of dollars in losses.

In a blog post on May 25, Buterin noted that the increased scrutiny that cryptocurrencies and DeFi have come under since Terra’s debacle is “very welcome,” but cautioned against disregarding it entirely. All algorithmic stablecoins.

“What we need is not stablecoin support or stablecoin apocalypse, but a return to principles-based thinking,” he said.

“While there are plenty of automated stablecoins that are fundamentally flawed in design and ultimately doomed to collapse, there are plenty of stablecoins that could survive in theory but with high stakes, and there are plenty of stablecoins that are very sound in theory and have worked in practice has withstood the extreme test of the crypto market environment.”

His blog post focuses on Reflexer's RAI stablecoin fully collateralized by ETH. RAI is not pegged to the value of fiat currency, but instead relies on an algorithm to automatically set interest rates, proportionally against price fluctuations, and incentivizes users to push RAI back to the target price interval.

It “represents the pure ‘ideal type’ of collateralized automated stablecoins,” Buterin said, and its structure also gives users the opportunity to withdraw liquidity in ETH at a time when confidence in the stablecoin has plummeted.

The ethereum co-founder offered two thought experiments to determine if algorithmic stablecoins are "really stable."

1: Can the stable currency be reduced to zero users in a "calm" manner?

In Buterin’s view, if a stablecoin project’s market activity “drops to near zero,” users should be able to extract the fair value of their liquidity from the asset.

Buterin emphasized that UST does not fit this parameter because in its structure, LUNA (or the volcoin as he calls it) needs to maintain its price and user demand in order to maintain its peg to the US dollar. If the opposite were to happen, it would be nearly impossible to avoid a collapse for both assets.

"First, the price of volcoin fell. Then, the stablecoins started to fluctuate. The system tried to support the demand for stablecoins by issuing more volcoins. Due to the lack of confidence in the system and fewer buyers, the price of volcoins fell rapidly. Finally, once the price of volcoins Close to zero, and stablecoins collapse.”

In contrast, since RAI is backed by ETH, Buterin believes that a drop in confidence in the stablecoin will not lead to a negative feedback loop between the two assets, leading to a broader crash that is less likely. Meanwhile, users will still be able to exchange RAI for ETH locked in vaults that support the stablecoin and its lending mechanism.

2: Negative interest rate option is required

Buterin also believes that the ability for algorithmic stablecoins to achieve negative interest rates is crucial when tracking “a basket of assets, a consumer price index, or some arbitrarily complex formula” that grows 20% per year.

"Obviously, there's no real investment that's going to get anywhere close to 20% a year, and absolutely no real investment that's going to grow 4% a year forever. But if you try, what happens?" he said.

There are only two outcomes in this case, he said, either the project “charges holders some kind of negative interest rate that basically cancels out the dollar-denominated growth rate in the index.”

Or "it became a Ponzi scheme, giving stablecoin holders amazing returns for a while, until one day it suddenly collapsed."

Buterin concluded by pointing out that just because an algorithmic stablecoin can handle the above scenarios, it doesn’t mean it’s “safe.”

“It could still be vulnerable for other reasons (such as insufficient collateral ratios), or have bugs or governance holes. But steady-state and edge-case robustness should always be one of the first things we check.”