Stablecoins have another weakness

Stablecoin growth may have stalled, but the lull might be short-lived if expert predictions are to be believed.

As my colleague Tim Craig noted last week, US Treasury Secretary Scott Bessent recently forecast US dollar stablecoins would swell to $3 trillion by 2030. Citibank predicted $4 trillion. S&P Global gave euro stablecoins a range of $250 billion to $1.1 trillion.

Put another way: it’s critical that investors and policymakers know whether their stablecoins are, well, stable. And new research from MIT suggests there are risks lurking in plain sight, and that last year’s Genius Act did nothing to address these risks.

Even when stablecoins are fully-backed by highly liquid assets, they can break their peg in times of market stress as underlying infrastructure buckles, the researchers found.

“A stablecoin’s ability to trade at par under stress depends not only on its asset quality, but on the functioning of redemption mechanisms, markets, and operational infrastructure,” they write. “The Genius framework leaves these stress-contingent dynamics largely unspecified.”

Stablecoins need to keep their peg. And virtually every conversation about stablecoin resilience centers on the issuer’s reserve assets.

“It implicitly treats stablecoin stability as a balance-sheet problem, resolvable through conservative asset holdings and supervision,” the researchers note.

It’s why Tether got a rather poor grade from S&P last year. S&P cited Tether’s reliance on supposedly high-risk reserves, such as gold and Bitcoin, when it said the issuer was vulnerable to a market downturn.

Even if Tether were invested entirely in short-duration US Treasuries, as money market funds are, a “run on the bank” could challenge its ability to quickly sell those treasuries.

Broker-dealers stand between buyers and sellers of treasuries. And they can be a bottleneck that limits stablecoin issuers’ ability to honour mass redemption requests, according to the researchers.

For example, when the coronavirus hammered markets in March 2020, bid-ask spreads on treasuries widened. If that were to happen again, a run on a stablecoin could knock it off its peg, setting off a doom loop.

There is an easy fix: let stablecoin issuers borrow directly from the Federal Reserve, and address the bottleneck issue. But then they would be subject to stringent bank-like regulations, which could make them unprofitable in a low-interest rate environment.

Traditional markets aren’t the only threat to stablecoin stability.

Issues with blockchains and smart contracts could also knock a stablecoin off its peg, according to the researchers.

On a matrix that ranks such risks by severity and likelihood, they identify two that are most worrisome: Smart contract logic flaws and bridge failures.

Then there are more quotidian problems. Last year, Paxos accidentally minted $300 trillion PYUSD. Though it corrected the problem minutes later, Aave had to pause PYUSD activity and the stablecoin briefly traded below par.

In the US, banks are attempting to re-litigate the Genius Act, annoyed that crypto companies seem to have found a loophole allowing them to pay customers yield on their holdings. The debate has held up landmark legislation that would bring the rest of the crypto industry out of its legal grey area.

If there are additional risks the law never considered, this might not be Congress’ last stablecoin battle.

Top DeFi stories of the week

This week in DeFi governance

VOTE: GMX DAO votes to fund GMX Labs for two years

PROPOSAL: Fluid DAO considers creation on nonprofit foundation

VOTE: Stablecoin issuer Angle Protocol votes on incentive program

Post of the week

Crypto sleuth ZachXBT made waves when he recently said he would soon provide evidence that employees at one of crypto’s most profitable companies were engaging in insider trading. That has led to speculation as to the company’s identity, a prediction market, and many, many jokes.

Aleks Gilbert is DL News’ New York-based DeFi correspondent. You can contact him at aleks@dlnews.com.