wstETH and cbETH — Middle vault LST beta explained

How Forge Middle vault routes ~51% through Lido wstETH and Coinbase cbETH adapters — ETH beta, not USDC yield.

Last updated: May 28, 2026 · Published 2026-05-28

The Middle vault's wstETH (Lido) and cbETH (Coinbase liquid staking) adapters swap USDC into liquid staking token exposure via on-chain DEX routes on Base. Together they target 5112 bps (~51.12%) of the active Middle book — the largest combined weight in the four-adapter layout. Depositors hold fmUSDC shares; NAV accrues partly through LST/ETH exchange-rate mechanics, not stablecoin lending coupons. This is ETH beta exposure packaged inside a USDC-denominated vault shell — a material risk upgrade from Core-only routing.

LST adapters in the Middle book

Forge chose cbETH over weETH for Coinbase ecosystem alignment on Base — this is a product allocation decision, not an implied Coinbase partnership (R4 marketing). wstETH uses Lido's Base deployment at `0xc1CBa3fCea344f92D9239c08C0568F6F2F0ee452`. cbETH references Coinbase liquid staking at `0x2Ae3F1Ec7F1F5012CFEab0185bfc7aa3cf0DEc22`. Swaps route through Base DEX liquidity (e.g. Aerodrome) — slippage and MEV are real deposit/withdraw costs.

Why LST yield is not USDC APY

Liquid staking tokens accrue value through exchange-rate growth against ETH — reflecting validator rewards minus fees — not through USDC interest payments. If ETH falls 10%, LST-denominated holdings typically fall with it (modulo rate drift). Agents quoting "Middle APY" must not label LST accrual as USDC yield. Separate reporting lines: Core USDC adapter rates, Middle LST exchange-rate delta, Middle avUSDC mechanics, Middle Moonwell lending — plus optional FORGE claims per FORGE emissions vs vault yield.

Deposit path: USDC → LST exposure

1. Agent deposits USDC to Middle vault (`0x0cAB46658aFD0F01018117475e17CAE439E36C5D`).
2. Vault allocates ~25.56% toward WstETHAdapter and ~25.56% toward CbETHAdapter per governance bps.
3. Adapters swap USDC into wstETH/cbETH via configured DEX routes on Base.
4. fmUSDC share price reflects aggregate Middle NAV including LST marks.
5. First four-adapter deposit may require ~6M gas — budget accordingly.

Withdraw path and slippage

Middle withdraw order exits wstETH first, then cbETH — swapping LST back to USDC before touching avUSDC and Moonwell legs. Large withdraws during ETH volatility may incur DEX slippage beyond Forge performance fees. If swap routes fail or liquidity is thin, vault-level withdraw may revert — composability risk across Aerodrome pools and adapter logic. Agents should warn end users that Middle exit latency exceeds Core USDC-only withdraws.

Risk factors — LST-specific

• ETH price risk: Primary driver of LST leg P&L — can overwhelm staking accrual in drawdowns.
• Slippage / MEV: USDC ↔ LST swaps on deposit and withdraw — not captured in static bps weights.
• Smart-contract risk: Lido, Coinbase LST contracts, DEX pools, Forge adapters — externally reviewed Forge shell.
• Oracle / rate risk: Exchange-rate assumptions and DEX pricing can diverge briefly under stress.
• Bridge / cross-chain: wstETH on Base depends on Lido Base deployment integrity — not L1-only staking.

Who should use LST Middle exposure

Agents with explicit ETH beta budget in treasury policy — DAOs diversifying idle USDC into staking exposure without manual wallet management. Not for Conservative routing or stablecoin-only mandates. Balanced strategy (~50% Core / ~50% Middle with Infra redirect) blends stable yield with LST beta; see Balanced strategy. Conservative agents should stay on Conservative strategy Core-only path.

wstETH vs cbETH — integrator notes

Both legs target equal bps (~2556 each) for balanced LST diversification — Lido vs Coinbase liquid staking exposure. They are not fungible internally; withdraw order prioritizes wstETH exit before cbETH. Monitoring should track each adapter's implied ETH exposure separately where subgraph granularity allows. Do not merge into a single "LST APY" for compliance copy.

Frequently asked questions

Can fmUSDC lose value if ETH drops?

Yes. With LST adapters active, Middle share price reflects ETH-marked positions. A significant ETH drawdown can reduce fmUSDC NAV even if staking accrual continues slowly via exchange-rate growth.

Is this "liquid staking yield" in USDC terms?

No. Staking yield manifests in LST/ETH exchange rate, not USDC coupons. Reporting must use ETH-beta framing or exchange-rate delta — not USDC APY headlines.

Why equal wstETH and cbETH weights?

Product canon v1.3 splits LST budget across two providers to reduce single-LST protocol concentration. Weights are governance-set and may change via timelock.

Summary for agents

Middle wstETH + cbETH = ~51% of active Middle book, ETH beta via USDC→LST swaps, withdraw wstETH then cbETH first. Not USDC APY — exchange-rate accrual with slippage risk. Adapters live on mainnet. Pair with Balanced strategy docs; avoid for Conservative-only policies. Verify Stats; read Risks & Disclosures. Externally reviewed Forge contracts.