Confidence lost: Understanding the risks of stablecoins. Part 1: Custodial stablecoins.
In this article, we’re decomposing the fundamental risks of stablecoins. Part 1 focuses on custodial stablecoin and Part 2 on non-custodial stablecoins.
Nearly everyone who uses cryptocurrency owns stablecoins. However, few understand the risks of various stablecoin designs. Let’s dive in.
Introduction
A stablecoin constitutes a protocol-based, chain-native token which has a stable price relative to an arbitrary value (i.e., USD $1) and aims to track the return of the sovereign currency on a public blockchain. In the cryptocurrency realm, peg refers to a given price the token aims to stay at. While there is no generally accepted taxonomy for stablecoins, we prefer to use a risk-based approach and distinguish between two basic categories: stablecoins which require trust in a third party (custodial stablecoins) and stablecoins which replace the trust required with economic mechanisms implemented through smart contracts (non-custodial stablecoins).[1]
Note: Aren’t a centralized and a collateralized stablecoin (or a decentralized and a non-custodial stablecoin, respectively) the same thing? Short answer: No. While these terms often are used interchangeably, they actually characterize two different properties of the stablecoin: Centralization describes the governance layer of a stablecoin, while (non-)custody is an asset layer property. There can be centralized non-custodial stablecoins designs and vice versa.
Note 2: Isn’t off-chain and custodial collateral in a stablecoin design the same thing? Short answer: No. While these terms are often conflated, off-chain and on-chain simply describe the technological mechanics of the collateral. On-chain collateral can be custodial or non-custodial, but not vice versa. Custodial stablecoins are, inter alia, subject to counterparty and censorship risks. “Off-chaining” exacerbates these risks. This is important to understand, as decentralized non-custodial (and therefore strictly on-chain) stablecoins are one of the best and arguably the trillion dollar use case for crypto.
Custodial Stablecoins
1. Risks
Custodial stablecoins are backed (collateralized) by traditional assets off-chain and require a custodian for safekeeping these assets. The issuer (in liaison with a custodian) then offers digital representations (tokens) on-chain of the safekept assets and users get a peg maintaining token representing a claim against the assets held by the custodian.
The drawbacks of custodial stablecoins are manifold. First, trust in centralized asset custodians is required, which is deeply antithetical to the ethos of cryptocurrencies. Trust in custodians not only entails the trust that the custodian will live up to his promise if called to do so, but also that the collateral truly exists in the quantum or value promised.
Because these stablecoins are centralized, they can be easily censored and wallets have been repeatedly blacklisted in the past.[2] Further, custodial stablecoins are subject to heightened (disclosed, undisclosed or unknown[3]) custody[4]/counterparty risks.[5] If a custodian is unable to fulfil its obligation[6] to return the safekept[7] assets[8] due to mismanagement[9], value deterioration[10] (economic risks of capital assets)[11], fraud, theft, freezing orders by authorities[12] etc. the value of the token loses its peg and can go down to zero. In other words, custodial stablecoins suffer from downward price instability due to moral hazard and/or from financial risk.[13]
2. Reserves
Custodial stablecoins can be divided into full reserve stablecoins (each stablecoin is backed 1:1 by a unit of the reserve asset, i.e. USD $1) or fractional reserve stablecoins (a mix of reserve assets and capital assets). Intuitively, full reserve stablecoins appear to be the apex custodial stablecoin design, as reserve assets always fully account for the entire stablecoin supply value. However, at this point there are no relevant full reserve stablecoins available.[14]
The most popular stablecoins such as Tether and USD Coin are fractional reserve stablecoins. The issuer holds reserve assets to facilitate redemptions, usually in the form of bank deposits.[15] The remaining assets constitute capital assets and account for the remaining stablecoin supply value. The issuer tries to earn a higher interest rate and sell capital assets to address redemptions if required. However, this design is always subject to price and default risk.[16]
3. Misaligned Incentives
The design suffers from misaligned incentives: The issuer gets to profit from investment gains without taking on the economic risks.[17] Subsequently, these risks are transferred to the token holder, who may not be able to redeem his token at par value and suffer substantial losses because of these misaligned incentives.
Put differently, for both full and fractional reserve stablecoins a run similar to a run on the bank or money market funds can happen at any time. If the issuer fails to provide timely and accurate information about the reserves, users question the quality of the reserve, lose confidence, and are strongly incentivized to convert the stablecoin to cash with immediate effect, triggering a run.
4. Fragility
It is important to understand that stablecoin runs don’t occur because the issuer is insolvent. Rather, a stablecoin run can cause a healthy issuer to become insolvent. If users panic and rush to convert their stablecoins to cash, it will bring the issuer down even if the fears were initially unfounded. This is the reason all issuers holding reserves with price risk have a potentially fatal fragility.
5. Fractional reserve stablecoins?
Like banks hundreds of years ago,[18] custodial stablecoin issuers are likely to embrace regulation and governmental oversight in the foreseeable future to be able to drop the less lucrative narrow banking model in exchange for a fractional reserve model similar to fractional reserve banking, all while claiming 1:1 exchangeability and adapting the modus operandi of the legacy banking system. Arguably, the result is a quasi CDBC.
6. Stablecoins as reserves
Some variants of custodial stablecoins use other custodial stablecoins as a reserve asset to avoid the price risk of capital assets or fractional reserves. In turn, these designs not only add another layer of protocol risk and complexity, but more importantly these designs inherit all the risks from the underlying custodial stablecoin, i.e. risks they neither own, nor control, nor possess the ability to fully assess or understand or influence or mitigate in any relevant way. In other words, these designs adopt a black box of operational risks from an unrelated underlying custodial stablecoin issuer and extend them[19] on to the token holders.[20]
7. Interim conclusion
In summary, risks of custodial stablecoins designs are generally well understood, but largely ignored by users and market participants.
This article continues in Part 2: Non-custodial stablecoins.
[1] Compare to Klages-Mundt, A., Harz, D., Gudgeon, L., Liu, J. Y., & Minca, A. (2020, October). Stablecoins 2.0: Economic foundations and risk-based models. In Proceedings of the 2nd ACM Conference on Advances in Financial Technologies (pp. 59–79).
[2] Tether blacklist on Ethereum https://dune.com/phabc/usdt---banned-addresses
[3] From launch in October 2018 — April 2021 the monthly independent reports by Grant Thornton LLP stated that “As of the Report Date and Time, the issued and outstanding USDC tokens do not exceed the balance of the US Dollars held in custody accounts”. As of May 2021, this confirmation stopped appearing in the monthly reports. https://www.centre.io/hubfs/pdfs/attestation/Grant-Thorton_circle_usdc_reserves_06092021.pdf
[4] Circle’s Investor Presentation names no less than 79 risks, including penalties to be paid to the SEC of over USD $10m for unauthorized trading of securities. https://www.circle.com/hubfs/investors/Circle-Investor-Presentation-July2021.pdf
[5] After accumulating a deficit of $378.4m in 2020/21 Circle required, inter alia, a loan from the US government (“PPP Loan”) and $441m in additional funding from multiple lenders to be able to keep operating for another minimum of 12 months. https://www.sec.gov/Archives/edgar/data/0001824301/000121390021036070/ea143875ex99-6_concordacq.htm:
“The Company has historically experienced unprofitable financial results due to the Company’s reliance on trading revenue (…) Last year there was substantial doubt of the Company’s ability to continue as a going concern.”
[6] For example, Tether’s Terms of Service fully exclude the right of Tether Token holders to redeem the Tether Tokens for cash, defying the one single purpose of a custodial stablecoin. Further, the Terms reserve the right to postpone redemptions in theory indefinitely, stating “Tether reserves the right to delay the redemption or withdrawal of Tether Tokens if such delay is necessitated by the illiquidity or unavailability or loss of any Reserves held by Tether to back the Tether Tokens, and Tether reserves the right to redeem Tether Tokens by in-kind redemptions of securities and other assets held in the Reserves”. https://tether.to/en/legal. A similar wording can be found on Circle’s website: “We reserve the right to (i) change, suspend, or discontinue any aspect of the USDC Services at any time, including hours of operation or availability of any feature, without notice and without liability and (ii) decline to process any issuance or redemption without prior notice and may limit or suspend your use of one or more USDC Services at any time, in our sole discretion.” https://www.circle.com/en/legal/usdc-terms
Essentially, Tether or USDC token holders are unsecured creditors. Even if they weren’t, chances for legal recourse (considered the capital and time required) against offshore entities like Tether’s are de facto close to zero.
[7] It’s unclear if Tether adheres to basic governance principles such as asset segregation. Their Terms of Service state: “The composition of the Reserves used to back Tether Tokens is within the sole control and at the sole and absolute discretion of Tether.”
[8] Tether does not publish reserve attestations by a Big 5 accounting firm. Their attestations come from a recently hired Cayman based accounting. https://www.coindesk.com/markets/2022/01/26/tethers-new-accounting-firm-is-the-old-one-with-baggage
[9] Tether makes substantial legal efforts in public courts to keep the reserves backing their stablecoin USDT a secret https://coingeek.com/tether-desperately-trying-to-stop-nyag-probe-results-from-going-public
[10] Example of more than half of Circle’s reserves not being in invested Cash & Cash Equivalents. USDSC Reserve Breakdown August 2021: Cash & Cash Equivalents: 46%. Corporate Bonds: 16%. Yankee CDs: 15%. US Treasuries: 13%. Plus, $100m are held in Municipal & US Agency Bonds. As per November 2021, Circle stopped reporting reserve details. https://www.centre.io/hubfs/pdfs/attestation/Grant-Thorton_circle_usdc_reserves_08242021.pdf. Notably, Circle’s reserves are capital assets and subject to price risk, however no investment risk related disclosure can be found in their public list of “USDC Risk Factors” here: https://www.circle.com/en/legal/usdc-risk-factors
[11] Circle’s attestations are no longer “correctly” stated, but “fairly” stated. https://www.coindesk.com/layer2/2022/03/24/correctly-was-wrong-circles-accountant-tweaks-fine-print-of-usdc-attestation
[12] Paragraph 6 of Tether’s Risk Disclosure states that: “(…)Tether relies on financial institutions and counterparties to hold funds (…) Reserves held at or through financial institutions or intermediaries may be subject to the risk of loss, theft, insolvency, and governmental and regulatory freezes and seizures.” https://tether.to/en/legal/#risk-disclosure-statement
[13] Kwon, Y., Kim, J., Kim, Y., & Song, D. (2021). The Trilemma of Stablecoin. Available at SSRN 3917430. 17.
[14] Because sitting on idle cash is not a business model.
[15] Here, a token holder additionally takes on bank run risk.
[16] Makarov, I., & Schoar, A. (2022). Cryptocurrencies and Decentralized Finance (DeFi) (No. w30006). National Bureau of Economic Research. 37.
[17] It’s worth noting that the business model of a stablecoin issuer like Circle and the failed CeFi platform Celsius are fundamentally the same: Acquire custody of a user’s financial assets and invest them with a) the minimum transparency necessary and b) the maximum returns possible (for the custodian, not for the user); while legally limiting liability to the largest extent possible. The difference is Circle gets to keep all the returns while Celsius gets to keep the excess return over a promised return.
[18] Fractional reserve banking dates to the 17th century, popularized by the Bank of Amsterdam.
[19] For details revisit previous custodial stablecoins chapter.
[20] Essentially just a wrapped version of the underlying token.
