Rethinking Stablecoin Redemption
Challenging the assumption of mass issuer redemptions as a Business-as-Usual expectation
This article questions the prevailing assumption of central banks and regulators, that mass redemption at par, on demand and within a T+1 timeframe, should be a business-as-usual (BAU) expectation for stablecoin issuers. This expectation diverges from international approaches (e.g. MAS, ESMA, GENIUS proposals). It also fails to distinguish between two fundamentally different concepts - redemption and conversion - thereby misdiagnosing the market risk and proposing an unnecessarily rigid solution that could undermine both innovation and stablecoin stability.
Redemption is typically the process by which a stablecoin is returned to the issuer to be burned, removing it from circulation, and accompanied by liquidation of the corresponding backing assets held in the issuer’s treasury, with the subsequent proceeds returned to the requesting entity. Redemption therefore has economic impact and operational cost to the issuer. Issuers - and distributors or other intermediaries - therefore typically seek to minimise redemptions. Conversion, in contrast, refers to the exchange of a stablecoin into another form of money (e.g. bank deposit, other stablecoin, or, in future, a retail CBDC) via intermediaries or exchanges. Conversion has no economic impact to the issuer, but forms a core part of the intermediary’s business model, which is typically predicated on market making and liquidity provision.
In practice, conversion - not redemption - is the dominant mechanism by which end users enter and exit stablecoins, and is the outcome that should be sought in future interoperable markets that support multiple forms of digital currencies. Redemption serves two distinct but complementary purposes for a stablecoin issuer. First, it underpins conversion: by guaranteeing secondary market participants the ability to redeem or issue at par, the issuer enables market makers and conversion providers to confidently offer 1:1 exchange in the secondary market, either as a flat service with a fee or through arbitrage activity on exchanges. This access to direct redemption ensures stable secondary market pricing and liquidity for end users. Second, redemption provides a guarantee of last resort, a backstop mechanism designed to preserve trust in the stablecoin even under stress. It is important to note that this is not intended as a continuous, daily operational flow for all holders. Just as retail bank customers do not withdraw all their funds in cash at once, but instead rely on seamless access to spend and transfer their money, stablecoin users will typically rely on conversion and everyday transaction use, with redemption remaining a safeguard rather than a constant operational service.Forcing redemption to act as a route for all users to exit a stablecoin risks over-engineering the regime and distorting the role of the issuer. Furthermore, it risks imposing a blanket regulation based on a high-risk scenario rather than the true realities of the risks posed on a BAU basis.
Expanding upon this point, the real risk to address in relation to redemption is not issuer liquidity under BAU conditions, but rather the ability to make coinholders whole in failure. If a stablecoin issuer holds 100% high-quality liquid assets (HQLA), for example in the form of short-term UK government debt, demand deposits, qualifying money market funds and, in future, central bank reserves, then all coinholders can be made whole in full over time, even if mass redemptions cannot be met instantly. For example, in the event of an issuer’s failure, it is important to observe that provided all backing assets are, to give an extremely conservative example, a mixture of (remunerated) central bank reserves and HQLA in the form of UK gilts having a maturity of 6 months or less, then the longest a coinholder will have to wait is 6 months until all the backing gilts have matured, at which point they can be made whole again. Understandably, there may be concerns around in-flight payments. Given that stablecoin payments are near-instantaneous and do not require lengthy clearing or settlement periods via an intermediary, it is arguable that the same risks do not exist with respect to in-flight payments as they do with a traditional payments system. Indeed, in a market stress scenario, the notion that all HQLA can be liquidated immediately at no loss (rather than through orderly wind-down in which assets are allowed to mature) is inconsistent with both reality and with Basel III’s own caveats on systemic fire sales. The quality of backing, not the speed of liquidation, should be the primary focus of the regulatory regime.
Furthermore, mandating universal T+1 redemption concentrates liquidity risk entirely on the issuer, rather than allowing it to be dispersed across the market via conversion providers and arbitrage participants. This creates a self-fulfilling liquidity crunch: under stress, coinholders rush to redeem en masse, draining issuer resources and potentially triggering failure. In contrast, a market-based model, in which liquidity is provided by intermediaries, with redemption acting as a backstop, naturally spreads risk across intermediaries, exchanges, and arbitrageurs, as is already the case in secondary markets.
It is also important to recognise that stablecoin redemptions, once a coin achieves scale and confidence, will likely become even more infrequent, as are total withdrawals of one’s bank deposits from a retail bank. As with other forms of regulated money (e.g., bank deposits or e-money), widespread utility and trust reduce the demand for cashing out. If redemption requests exceed, for example, 40% of outstanding supply, it is no longer BAU; it signals a stress event and should trigger provisions for the issuer to redemption timeframes to allow for orderly liquidation of assets, or even trigger recovery or resolution frameworks. It should not force the issuer to conduct a firesale in order to meet a regulatory requirement for 100% liquidity in 24 hours.
In our view, a universal T+1 redemption model increases risk. Mass redemption requests could cause overwhelming operational challenges and a firesale of backing assets. This in turn exacerbates market concern around the issuer’s ability to meet the T+1 timeframe, leading to a self-fulfilling run of the stablecoin. Holders would find themselves unable to redeem or sell at par and the stablecoin would cease to exist. By contrast, in a model in which redemption is reserved for a small number of customers and is not T+1, the same mass redemption event would not put the same pressure on the issuer. The issuer’s terms of redemption with the minority of holders who are customers would provide for a longer redemption timeframe permitting orderly liquidation of the backing assets. Holders who are not customers would not seek to enforce redemption against the issuer at all. Instead, the result would be downward price pressure on secondary market exchanges, as holders accept prices below 1:1 in order to convert into bank money. However, the arbitrage mechanism would to some extent counteract this pressure. In this scenario, losses would be limited and dispersed across holders and the secondary market rather than being concentrated on the issuer and forcing its failure. Losses would be suffered by secondary market participants as either delays in conversion compared to the T+1 ideal or as the acceptance of a price below 1:1. In a rare stress scenario, this is a safer outcome than the failure of the issuer, in which holders would suffer greater losses.
The right for customers to redeem at par is not in question. This is a key part of the product and necessary to enable a liquid secondary market. But insisting on immediate execution of that right in all cases for all holders imposes liquidity constraints that do not even exist for banks at present, without providing access to bank-like infrastructure, such as central bank liquidity facilities. It is also inconsistent with many forms of bank accounts (e.g. restricted access savings accounts offered by commercial banks, usually in return for (in theory at least) a higher fixed interest rate, or “'fixed term” products which can be multi-year with no access) where there are time constraints on withdrawals. Regimes proposing T+1 redemption standard, such as the UK’s FCA (at the time of writing), in seeking to operationalise universal, near-instant liquidity, risk solving for the wrong problem.
If the regulatory objective is to ensure that holders can ultimately be made whole, the focus should be on quality of backing assets, not liquidity at all times. A backing asset portfolio composed of HQLA will allow holders to be repaid in full, even in failure, without creating a cliff-edge redemption dynamic that can trigger issuer insolvency. Unlike stablecoin issuers, no banking or financial institution is able to make 100% of its liabilities available to all customers simultaneously, not in BAU, and not even in failure. The risk of financial loss arising for holders of regulated stablecoins even in a failure scenario is therefore vanishingly small compared to the risk presented by commercial bank deposits, yet the current stablecoin proposals risk escalating that risk by creating self-fulfilling liquidity spirals and undermining the long-term viability of UK-based stablecoin issuance.
Regulators should therefore be urged to reframe the role of redemption within the stablecoin regime. It is not, and should not be, the primary mechanism by which users exit a stablecoin. That role is fulfilled by the market, via intermediated conversion. Redemption should be preserved as a guaranteed right of last resort (consistent with holding backing assets on trust for holders), not a BAU obligation, and thresholds should be established beyond which redemption is considered a stress or recovery scenario, with appropriate supervisory treatment. This would align the UK with international regulatory best practice, support issuer resilience, and ensure that the regime reflects how stablecoins function as part of a broader financial ecosystem.
A more balanced approach would:
Uphold the right to redeem at par only for direct customers or in the event of the failure of the issuer, and would allow redemption timeframes to vary under BAU vs stress scenarios.
Recognise conversion as the primary exit mechanism for coinholders, with redemption supporting liquidity, arbitrage and long-term trust.
Ensure that redemption rights and backing assets together provide a credible, enforceable means of maintaining stability of a stablecoin - and preserving the singleness of money - without requiring immediate liquidity across the entire backing asset pool.
If the regulatory goal is to ensure that stablecoins remain reliably interchangeable with other forms of money at par, then the focus should be on ensuring credible backing, transparent governance, and a supportive market structure. Stablecoin stability does not require instant redemption for all users at all times. It requires a regime that enables trust, supports arbitrage, and provides for orderly redemption under stress. In this respect, current regulatory proposals in jurisdictions such as the UK risk conflating liquidity and solvency, and in doing so, may inadvertently threaten both.