The stablecoin risks institutions actually underwrite

Institutions that settle in stablecoins underwrite five risks: depeg, issuer and reserve quality, issuer concentration, operational failure, and regulatory divergence. Every one of them has a documented precedent, and every one of them has an established mitigation. The institutions that use stablecoin rails well are the ones that can name each risk precisely, price it, and put a control against it, which is what this page does.

The honest starting point is that the risks are not hypothetical. The two most instructive episodes in the market’s history are a regulated, reserve-backed coin losing its peg for a weekend, and an unbacked one losing everything.

Depeg risk: the weekend USDC traded at 86 cents

In March 2023, Circle disclosed that $3.3 billion of USDC’s cash reserves, about 8% of the total, were held at Silicon Valley Bank when regulators closed it. USDC broke its dollar peg almost immediately and traded as low as 86 cents at its trough. When US authorities announced that SVB depositors would be made whole, the price recovered, and once Circle resumed redemptions on the Monday the peg was fully restored.

The episode set the template for how institutions should think about depeg risk in reserve-backed coins. The token traded below par exactly as long as there was doubt about the reserves behind it, and not a day longer. Depeg risk in a regulated coin is a window of uncertainty, not a permanent loss, but an institution holding large balances through that window is carrying real mark-to-market exposure, and one that needed to redeem during it would have taken a haircut.

TerraUSD was a different kind of failure. UST was algorithmic: backed by a mechanism rather than by reserves, it relied on an exchange link to a floating sister token to hold its dollar value. In May 2022 the mechanism entered a spiral and roughly $45 billion of market value across the two tokens evaporated within a week. There was no window to wait out. The lesson institutions drew is categorical: a coin without off-chain reserves is a structurally different instrument, whatever the ticker suggests, and post-crisis regulation in both the US and EU effectively excludes that design from institutional payment use.

Issuer and reserve risk: is the backing what the issuer says it is

Depeg risk asks what the market believes; reserve risk asks what is actually there. The formative case is Tether’s 2021 settlement with the CFTC, which found that between 2016 and 2018 the issuer held sufficient fiat reserves to back its tokens on only 27.6% of the days sampled, and fined it $41 million for misrepresenting its backing. Tether’s disclosures and reserve composition have changed substantially since, but the case established the principle: an issuer’s claims about its reserves are an underwriting question, and the answer has at times been no.

This is the risk regulation has moved hardest against. The GENIUS Act requires US payment stablecoin issuers to hold one-to-one reserves in high-quality liquid assets under federal or state supervision. MiCA requires e-money token issuers in the EU to be regulated institutions that redeem at par on demand. For coins inside these regimes, reserve risk now has a supervisory floor. For coins outside them, it remains what it always was: reliance on the issuer’s own attestations, which institutions should read the way a credit officer reads any unaudited balance sheet.

Concentration risk: two issuers, most of the market

The stablecoin market is large and lopsided. Of a roughly $300 billion market, the two largest issuers account for approximately 85% between them as of July 2026. An institution that builds stablecoin settlement without deliberate diversification is, by default, building a concentrated exposure to one or two private issuers, on top of whatever chain and custodian that flow settles through.

Concentration also has a corridor dimension. The coin with the deepest liquidity in emerging-market corridors and the coin with the cleanest regulatory standing in the EU are different coins. An institution locked to a single issuer either forgoes corridors or carries compliance risk in some of them. The mitigation is structural rather than clever: cap per-issuer exposure, maintain redemption arrangements with more than one issuer, and treat coin selection as a per-transaction routing decision rather than a platform commitment.

Operational risk: networks, keys, and custody

Stablecoin settlement inherits the operational risk profile of the ledgers it runs on. Networks have had outages; transactions are irreversible once final, which turns operational errors that fiat rails can recall into losses; and the security of balances reduces to the security of keys. Institutions answer this the way they answer custody risk everywhere else: qualified custodians, segregated accounts, dual controls on movement, and allow-listed counterparty addresses, so that a payment can only go where policy says it can go.

The subtler operational exposure is fragmentation. The same coin exists on many networks, and moving value between them introduces bridging and liquidity steps that are themselves operational surfaces. Institutions that standardize on a small set of networks per corridor, with fallbacks defined in advance, remove most of this surface before it becomes an incident.

Regulatory divergence: compliant where, exactly

A stablecoin’s regulatory status is jurisdictional, and the divergence is now concrete rather than theoretical. USDT, the largest coin by supply, did not seek MiCA authorization, and MiCA-licensed exchanges removed it for EEA customers through 2024 and 2025. USDC and EURC are authorized e-money tokens in the EU. The same instrument can be the market’s deepest pool of liquidity in one region and unavailable through regulated venues in another.

For an institution, this means the compliance question is per-corridor, per-coin, and it changes. A payment policy written around one coin’s status in one jurisdiction quietly becomes non-compliant when a regime tightens or a coin’s authorization changes. The durable answer is to encode the rule rather than the coin: EU corridors route over MiCA-authorized tokens, US corridors over GENIUS-regulated ones, and the routing updates when the regulatory facts do.

What the mitigations have in common

Walk back through the five risks and the mitigations converge on one design principle: no single point of dependence. Not one issuer, not one coin, not one network, not one jurisdiction’s rulebook. Institutions that hold that line convert stablecoin risk from an existential question into a portfolio of managed exposures, which is the only form in which a risk committee should accept it.

Where Frame fits

Frame is the settlement layer for global finance: one integration to orchestrate payments at scale across fiat rails, stablecoins, and tokenized deposits, with compliance enforced on every transaction and settlement in seconds instead of days.

Frame’s design assumes everything above. Because Frame is rail-neutral, stablecoins are one rail among several, and a payment that should not touch a stablecoin at all can settle over fiat rails or tokenized deposits through the same integration. Within the stablecoin rail Frame is coin-neutral, so concentration limits and jurisdictional rules are expressed as routing policy rather than re-integration work. Frame’s Rules Engine evaluates every transaction against its governing policies before it settles: a transfer that would breach an issuer cap, use an unauthorized coin in a regulated corridor, or move outside approved counterparties does not settle. The risks stay real. The controls stop being manual.