Stablecoin Weekly: Structural Risk Shows Up

This week made the case for the base layer. A cross-chain bridge exploit routed about $200M of bad debt into Aave, roughly $13B of DeFi TVL rotated out in forty-eight hours, and the parts of the stack built for institutional legibility (tokenized treasuries, regulated-settlement rails, fiat-backed issuer growth) kept maturing through the same week. The commercial resolution of the Drift incident from two weeks ago also landed this week, in a way that reshuffled competitive positioning between the two largest issuers. For allocators and treasury teams, this is the kind of week where the difference between shared-risk and isolated-risk design, and between base-layer and strategy-layer yield, stopped being a theoretical distinction.

• On April 18, a misconfigured LayerZero DVN setup let an attacker mint roughly $292M of rsETH across chains, deposit it as Aave V3 and V4 collateral, and borrow real assets against it. Aave was left with roughly $196M to $230M of bad debt, about $6.6B of TVL exited in twenty-four hours, and the broader DeFi TVL dropped roughly $13B over forty-eight hours.
• The Drift incident from two weeks ago resolved commercially: Tether led a $127.5M rescue package with $147.5M total from partners, and Drift's relaunch plan explicitly expands USDT usage as a settlement asset on Solana. Circle shipped new cross-chain USDC infrastructure routed through CCTP burn-and-mint across more than seventeen chains, with managed gas handling and upfront fee disclosure.
• The institutional rails kept opening. Superstate announced that Invesco Advisers will become the investment manager of USTB in Q2. Ondo, Clearstream, and 360X went live with ten tokenized U.S. equities and ETFs on a Deutsche Börse-backed venue. Visa took an anchor validator slot on Stripe and Paradigm's Tempo chain.

The aggregate supply picture stayed quiet in totals but meaningfully bifurcated underneath. Centralized fiat-backed stables grew, yield-bearing synthetic stables shed supply, and tokenized treasuries continued their steady upward drift. The deltas are small in absolute terms and directional in signal.

• Total stablecoin market cap is roughly $320.6B, up about $2.9B (+0.9%) from last week's baseline.
• USDT sits at about $187.2B, up roughly $3.0B (+1.6%).
• USDC is effectively flat at about $78.3B.
• USDS is down about 9.2% week-over-week to $7.98B, and USDe is down about 5.1% to $5.53B.
• RWA.xyz shows tokenized U.S. treasuries at about $13.80B across 76 live assets, up from $13.53B across 74 assets. The category 7D APY sits at 3.39%, a small tick up from 3.34%.

• USYC at about $2.90B with 3.18%, up about 8.8% week-over-week, the largest gainer in the category
• BUIDL at about $2.47B with 3.47%, up about 2.1%
• USDY at about $1.87B with 3.55%, roughly flat
• JTRSY at about $1.50B with 3.14%, a new entrant into the top five on Centrifuge's platform
• BENJI at about $0.97B with 3.54%, down about 5.1% and pushed a rank lower
• OUSG exited the top five at about $620.5M, down about 10.7% week-over-week

The treasury mix is rotating. Circle's USYC is taking share at the top, Centrifuge's JTRSY has climbed into the top five for the first time, and Franklin's BENJI and Ondo's OUSG both contracted. This is less a story about yield (the products print inside a narrow mid-3% band) and more a story about distribution, share-class flexibility, and which tokenized money-market wrapper institutional allocators are actually picking up.

The tokenized-treasury base layer tightened marginally, printing around the high-3% area this week, with the category 7D APY at 3.39%. That movement is trivial on its own. What matters is the context: in a week where DeFi lending saw real loss events and double-digit TVL migration, the base layer absorbed capital rather than losing it. For institutional mandates that treat the tokenized Treasury floor as the baseline every higher-yield product has to beat, the week was another data point that legibility pays for itself.

The picture here splits into a distressed part and a structural part.

The distressed part is Aave. USDC and USDT supply APYs on Ethereum spiked into the mid-teens, with figures near 13%, as utilization ran near 100% during mass withdrawals. That is not yield. It is a liquidity squeeze on a lender now carrying unresolved bad debt. Real-money allocators should read those headline numbers as a stress reading, not a deposit opportunity, and wait for resolution of the rsETH-related remediation before treating Aave stablecoin deposits as producing a stable rate again.

The structural part is quieter but real. Spark's new Savings Vaults V2 are printing around 4.25% on spUSDC, up from roughly 3.75% last week, and sUSDS sits flat at 3.75%. Sky-governance-set rates kept their usual week-over-week quietness, while the Spark IRM spell from April 9 modestly reshaped equilibrium utilization on ETH and USDT markets. Maple's syrupUSDC and syrupUSDT are quoted in the 6% to 8% band, modestly firmer than the 4.18% and 3.63% cited at last issue, with AUM continuing to expand. Morpho's broader curated USDC vault range sits roughly in the 3% to 8% range depending on curator and market configuration, broadly in line with last week.

Net read for allocators: outside the Aave stress window, the genuine spread above the tokenized-treasury base layer widened only modestly this week. The more durable shift is structural. Deposits that had been defaulting into the largest pooled lender are a candidate to rotate toward isolated-market venues like Morpho and governance-set-rate products like sUSDS and Spark Savings V2, because those designs did not propagate the Kelp incident. Morpho's team publicly noted isolated-market exposure to the Kelp-linked collateral in the low single-digit millions, against Aave's nine-figure bad-debt residual. That comparison is likely to shape curation mandates and counterparty selection for weeks.

Active credit held up quietly through the week. Maple continues to screen in the 6% to 8% range on its USDC and USDT products, and curated Morpho vaults continue to operate more as a menu than a benchmark, with quoted APYs ranging from low single digits to the high single digits depending on strategy and risk.

On the basis side, sUSDe screens around 3.72%, roughly in line with or modestly above last week's 3.58%. USDe supply contracted by about 5.1% on the week, which is consistent with reduced demand for synthetic-dollar carry rather than any mechanical product change. As with prior weeks, the point is not the single-period print but the regime: basis carry is currently paying only modestly over the tokenized-treasury floor, so the marginal allocator has to explicitly want to take funding-rate and collateral-structure exposure rather than assuming a clean yield premium.

Two large issuer moves were tightly coupled to the Drift aftermath. Circle shipped a new cross-chain USDC transfer product routed through CCTP burn-and-mint across more than seventeen chains, with managed gas, upfront fees, and near-real-time tracking. Tether launched tether.wallet, a self-custodial wallet for USDT, USAT, XAUT, and BTC with human-readable addresses and native gas-in-sent-asset handling. Separately, Tether led the $127.5M Drift rescue alongside partners and Drift committed to expanding USDT usage as a settlement asset on Solana as part of its relaunch. Tether Investments also participated in a $134M raise for SDEV, a NYSE-American-listed vehicle anchored in a roughly nine-percent position in SKY tokens. Two different playbooks: Circle leaning into native-rails infrastructure, Tether leaning into direct capital deployment and distribution wins.

The institutional rails kept opening alongside those issuer moves. Superstate announced the extension of its partnership with Invesco, with Invesco Advisers scheduled to become the investment manager of the roughly $1B USTB tokenized Treasury fund in Q2, which will be renamed but retain its ticker and contract identity. Ondo, Clearstream, and 360X went live on April 15 with ten tokenized U.S. equities and ETFs (including SPY, QQQ, AAPL, MSFT, NVDA, and Circle's own CRCL) on a Deutsche Börse-backed regulated venue accessible to EU broker-dealers, with Clearstream custody integration queued as the next step. Ripple and Kyobo Life Insurance, a roughly $92B AUM Korean insurer, announced a pilot for Korea's first tokenized government bond settlement over XRPL. Visa joined Stripe and Paradigm's Tempo chain as an anchor validator alongside Zodia Custody. Each of these is a distribution move rather than a marketing move, and none of them needed a yield narrative to make sense.

The substantive policy action this week was publication rather than rule-making. Treasury, FinCEN, and OFAC's April 10 NPRM on AML, CFT, and sanctions obligations for Permitted Payment Stablecoin Issuers was published in the Federal Register on April 10 with a comment deadline of June 9. The FDIC's NPRM implementing GENIUS Act prudential standards (reserves, redemption mechanics, capital, a twelve-month liquidity buffer, and a 10%-of-supply-in-twenty-four-hours redemption circuit breaker for bank-affiliated PPSIs) was published the same day with the same June 9 deadline. The NCUA's earlier PPSI-for-credit-unions comment period closed on April 13, the first GENIUS Act comment window to close formally. None of this introduces new rule text, but it moves the rulemakings from announced to operationally open for public comment and turns the comment cycle into a defined calendar.

The more politically live item is the CLARITY Act yield fight. The Senate returned from recess on April 13 with CLARITY back on the docket. White House crypto adviser Patrick Witt signaled that non-yield obstacles have mostly cleared. A White House CEA report from April 8 argued that a full ban on stablecoin-level yield would lift U.S. bank lending by only about $2.1B, or 0.02% of the current lending base. The American Bankers Association publicly pushed back on that estimate on April 13. The fight over whether issuers, exchanges, or third parties can pay any yield on permitted stablecoin balances remains the single most unresolved question shaping the economics of USDC, USDT, RLUSD, and successor products for institutions. Reporting later in the week suggested the bill may slip into May; none of that is confirmed by a Senate Banking Committee release.

Outside the United States, no material primary actions landed in-window from the SEC, CFTC, NYDFS, ESMA, FCA, MAS, or HKMA. The HKMA's April 10 issuance of Hong Kong's first stablecoin issuer licenses to HSBC and Anchorpoint sits just outside the coverage window but is worth flagging as context for the Asia distribution map.

The shape of this week is cleaner than the noise around it suggested.

A large pooled lender took a structural hit from a cross-chain bridge misconfiguration and is now carrying real bad debt, while an isolated-market lender that priced the same kind of collateral in a narrower context took low single-digit millions of exposure. That is not a moral point about protocols. It is a design point about risk segmentation that maps cleanly onto how institutional allocators already think about counterparty, concentration, and collateral identity. ArkenYield underwrites exposure by collateral identity and venue design first, yield second. That ordering is why a week like this lands as analysis, not remediation.

At the same time, the Drift aftermath delivered a concrete lesson in distribution: capital and settlement position are both contestable when a venue is forced to recapitalize. Tether put the check down and won the seat. Circle's response was to harden its own rails. Both responses are rational. Both are also signals about what institutional stablecoin competition looks like in a year when compliance and reserve discipline are becoming part of the product definition rather than the marketing wrapper.

Through all of that, tokenized treasuries expanded. New institutional rails opened. The base layer did what a base layer is supposed to do, which is stay quiet during the weeks when everything else is not. That is why stablecoin allocation is increasingly a portfolio-construction and operating-infrastructure problem, not a product-selection problem. The question is not whether these developments matter. It is which layers of the stack absorb this kind of week and which layers transmit it.

• Aave's rsETH incident remediation: how the roughly $200M bad debt residual is absorbed, whether safety module or treasury resources are used, and whether governance tightens caps, oracles, or LRT-identity rules as a result
• Whether the TVL and deposit migration out of pooled DeFi lending into isolated-market venues and governance-set-rate products proves structural or snaps back once Aave's situation stabilizes
• The CLARITY Act yield negotiation through late April and into May, including any Senate Banking Committee markup schedule
• Whether USDS and USDe supply continues to contract or stabilizes, and how yield-bearing synthetic stable supply behaves as lending stress settles
• First wave of public comment letters on the Treasury / FinCEN / OFAC and FDIC NPRMs before the June 9 deadline
• Whether Tether's distribution win on Drift begins showing up in other Solana-native venues, and whether Circle's new cross-chain USDC product drives meaningful flow through CCTP rather than competing bridges