Stable LP vs Stable Lending — Where to Park USDC in 2026
If you have stablecoins on-chain, you have two real choices for yield: lend them to a money market (Aave, Fluid, Kamino) or provide liquidity to a stable pair (Curve, Aerodrome, Orca).
Most newsletters tell you the LP option earns more. They're right on the headline rate. They're usually wrong on the after-risk return. This article is the decision tree the PassiveBlocks bot uses to pick between the two — with the numbers that matter.
The honest comparison
Same amount of capital ($1,000 USDC), same time horizon (12 months), as at 2026-05-19:
| Option | Headline APY | After realistic friction | Worst-month risk |
|---|---|---|---|
| Fluid USDC lending (Arbitrum) | 4.63% | ~4.55% | Protocol exploit (low) |
| Aave USDC lending (Ethereum) | 3.80% | ~3.75% | Protocol exploit (low) |
| Curve USDC/USDT/DAI (stable LP) | 5.20% | ~4.80% | One stable depegs 1% = -0.5% NAV |
| Aerodrome USDC/USDT (stable LP) | 6.50% | ~5.40% | Emissions decay + depeg risk |
| Orca USDC/USDT whirlpool | 7.10% | ~5.20% | Concentrated range = active management |
The LP rows look higher. They are higher — until something moves.
What you're actually being paid for
Lending pays you for taking protocol risk and rate risk. The borrower pays interest, the protocol takes a cut, you get the rest. Your USDC is always worth $1. You can withdraw whenever (subject to utilisation).
Stable LP pays you for taking peg risk and impermanent loss risk. You're providing liquidity for two assets that should both be worth $1. As long as they trade at parity, you earn swap fees plus (often) protocol emissions. The moment one of the two stables drifts — even temporarily — the pool rebalances against you, and you've now bought the cheaper stable at the higher price.
Same dollar of yield. Different reasons it shows up. Different reasons it disappears.
The 2023 USDC depeg, costed
In March 2023, USDC briefly traded at $0.88 after Silicon Valley Bank exposure was revealed. It re-pegged within 72 hours.
If you were lending USDC during that window: - You held a deposit token that was always denominated 1:1 in USDC. - You felt nothing in your position value. USDC re-pegged, your balance stayed identical.
If you were LP'ing USDC/USDT during that window: - The pool rebalanced as arbitrageurs swapped out cheap USDT for cheap USDC. - You exited the event holding more USDC and less USDT — but your dollar value was down ~4–6%, and you'd locked in the depeg loss. - A full year of swap fees (5–6% APY) would have been wiped out by the single weekend.
This is not a hypothetical. It happened. It will happen again to a different stablecoin. The 12-month yield gap between lending and stable LP did not survive one bad weekend.
When stable LP actually beats lending
LP is the right call in two specific situations:
1. Battle-tested pairs with peg-defended stables. USDC/USDT on Curve or Aerodrome, both stables backed by named issuers (Circle, Tether), with multi-year depeg-recovery history. The LP earns 1–2 percentage points more than lending in normal markets. The depeg risk is real but bounded.
2. Active concentrated liquidity, large capital. Orca and Aerodrome Slipstream let you concentrate liquidity inside a narrow band ($0.998–$1.002). On $50K+, the swap fees compound to genuine yield. On $1K, you're paying gas to rebalance the range more often than the fees pay.
For everyone else — sub-$50K, no time to manage ranges, no appetite for re-pricing a stable mid-week — lending wins on a risk-adjusted basis.
The PassiveBlocks bot's actual allocation
For full transparency, here is how the bot is currently positioned (2026-05-19):
- Stablecoin lending (Fluid, Arbitrum + Base): ~99% of stablecoin capital.
- Stable LPs: 0%.
- Volatile LPs (WBTC/WETH range): small allocation, separate strategy slice.
The bot does not run stable LPs right now. Not because they can't earn — they can — but because at the current capital size the 0.5–1.5pp uplift over Fluid USDC doesn't clear the depeg-risk premium the bot prices in. At $50K+, that math changes. At $1K, it doesn't.
That's the rule + receipt: the rule is "LP only when the after-risk return clears lending by 2pp+." The receipt is the allocation above.
The two-question decision tree
Before you put USDC into anything, answer two questions:
1. Is the headline APY mostly fees or mostly emissions? - Mostly fees → it's earnable yield, sustainable. - Mostly emissions → it's a marketing budget. Discount it 40–60%.
DeFiLlama shows the split for most pools. If you can't find it, assume emissions until proven otherwise.
2. Does the position survive a 1% depeg of either stable for 48 hours? - Lending → yes, your balance is unchanged. - Stable LP → no, you've eaten a 0.5–1% NAV hit on that side of the pair.
If you can't accept a stable-depeg drawdown, the only correct answer is lending.
A note on hardware
If your stablecoin balance is over $5,000 — whether in lending or LP — the gating risk is no longer protocol yield. It's wallet hygiene. A single phishing signature wipes a year of careful yield work.
A hardware wallet (Ledger or Trezor) costs ~$120–$190 once and removes that entire category of loss.
Get a Ledger Nano X → (affiliate — we earn a commission at no cost to you)
The takeaway
Lending earns slightly less and bores you to sleep. Stable LP earns slightly more and asks you to defend two pegs at once. On portfolios under $50K, the bored option wins on after-risk return almost every year.
The PassiveBlocks bot picks the boring option every cycle. That is the strategy, not a missing feature.
This is educational content, not financial advice.
More in Yield Strategies
Get analysis like this every Monday — free.
Subscribe to PassiveBlocks →