Back in January I wrote a one-page playbook for my stablecoin book. The consensus I signed onto was a cut story: bill yields would fade, tokenized Treasuries would stop leading on-chain, and I should let bills mature and rotate into lending and synthetic-dollar carry before policy rates fell. I dated the note, filed it, and spent nine months watching it be wrong.
On September 16, 2026, the FOMC hiked by 25 basis points to a 3.75–4.00% target range — interest on reserve balances at 3.90% and the ON RRP rate at 3.75%, effective the next day. As of October 1–2, the December futures strip priced a possible further move, and 12 of 18 dot-plot officials saw at least one more hike this year. The next meetings are October 27–28 and December 8–9. This is a hiking regime, so I stopped forecasting and replaced the playbook with a measured Friday rule. Here are the numbers I logged, the triggers, and worked math on a $50,000 sleeve — educational, not financial advice.
TL;DR. My January "cut rotation" was wrong: the FOMC hiked to 3.75–4.00% on Sep 16 (IORB 3.90%, ON RRP 3.75%). Logged Oct 1–2: sUSDS 3.60%, Aave V3 Ethereum USDC 3.71% supply APR (7d 3.80, 30d 3.62, utilization 93.9%, $2.31B TVL), Morpho Prime 4.40% Base / 5.27–5.28% Ethereum, OUSG 3.43% 30-day, sUSDe 3.71% — day two after ENA incentives ended forever Sep 30. I now run five Friday triggers instead of a rate view; 50bp on $50k is only $250/year, so caps, cooldowns and spread all count.
The plan versus what actually happened
The January plan had three legs: OUSG would roll into lower bill yields, so I'd stop topping it up; the Sky Savings Rate would be cut by governance, so I'd trim sUSDS ahead of the vote; and I'd pre-position into pooled lending and sUSDe, where carry was supposed to stay firm as rates rolled over — the Ethena vs Sky convergence I expected across two wrapped-dollar yields.
Instead the September 16 statement moved the range up. By my October 1–2 review the dot plot still showed 12 of 18 officials seeing at least one more hike this year, the December futures strip priced a possible further move, and front-end bills were already above 4%. Every leg of the rotation was mistimed. The lesson: I have no edge on the policy path; what I control is measurement, thresholds and what I pay to be wrong. The forecast became a rule.
The venue scorecard, logged October 1–2
Every Friday I log the same fields from two sources per number. This week's table:
| Venue | Rate logged Oct 1–2 | What sets it | Exit / access |
|---|---|---|---|
| sUSDS (Sky) | 3.60% | Sky Savings Rate, governance-set around IORB | Instant via 1:1 USDC/USDS PSM; blocked for US persons |
| Aave V3 Ethereum USDC | 3.71% supply APR (7d 3.80, 30d 3.62) | Pool utilization, logged at 93.9% | $2.31B TVL; withdrawal latency risk at high utilization |
| Morpho Prime | 4.40% Base / 5.27–5.28% Ethereum | Curated vaults, utilization and reward layers | Top Ethereum vaults ~$99–102M; curator risk |
| OUSG (Ondo) | 3.43% 30-day (Aug 3 review) | T-bills rolling daily, net of fees | ~$378M AUM; KYC; $100M/$50M global daily caps |
| BUIDL (BlackRock) | No official yield; proxies 3.47–3.78% Sep 30–Oct 1 | T-bills, repo and cash equivalents | QP-only, $5M subscription minimum |
| sUSDe (Ethena) | 3.71% trusted tracker (30d 4.15, 90d 3.76); 5.28% on another | Perp funding + stETH; ENA incentives ended Sep 30 | 7-day unstaking cooldown |
Behind the BUIDL row: BlackRock FedFund's 7-day SEC yield was 3.78% on September 30, and informal BUIDL proxies printed 3.47–3.78% across September 30–October 1 — neither an official BUIDL number, since the fund publishes no daily yield the way OUSG publishes 30-day figures. Access differences are in the BUIDL vs OUSG comparison and the Ondo vs BUIDL table.
Why a hike moves each engine differently
sUSDS follows IORB with a governance lag. The Sky Savings Rate is voted, not automated: after administered rates stepped up September 17, IORB stood at 3.90% while sUSDS still logged 3.60% on October 2; in a cut cycle the lag runs in reverse — sUSDS falls only after a vote. Spread and calendar are the whole subject of the Sky protocol review.
Bill tokens roll daily. OUSG's 3.43% 30-day figure reflects bills maturing into new bills, net of fees, so it adjusts gradually rather than on FOMC day. The raw bills already priced the hiking regime on September 24:
| Treasury bill tenor | Discount-equivalent yield, Sep 24 | What it means for a tokenized sleeve |
|---|---|---|
| 4-week | 3.86% | Reprices fastest, close to administered rates |
| 3-month | 4.08% | Core of OUSG-style portfolios |
| 6-month | 4.22% | My lending-spread benchmark |
| 1-year | 4.27% | Embeds regime expectations |
On $50,000, each 25bp of rolling bill yield is about $125 a year — my unit for comparing bills to lending.
Lending follows utilization, not the Fed. Aave's 3.71% live USDC supply APR is set by borrow demand — utilization was 93.9%, with a 3.80% seven-day and 3.62% thirty-day average. Coin Metrics data put Morpho's median supply APR +65bp versus bills since January 2026 but with 3.3× Aave's rate volatility; Aave ran 31bp under the 1-year T-bill and underperformed it on 78% of days. The premium is real, and so is its variance — the core Aave vs Morpho tradeoff.
sUSDe can disconnect from the Fed entirely. Its carry is short-ETH perp funding plus stETH (roughly 11%+ RWA in the backing mix), so it pays in any rate regime — or inverts. On September 30, day two of my window, ENA incentives ended permanently after more than $750 million distributed since 2024, per Ethena Foundation governance; every sUSDe quote from here is organic carry, the distinction in real yield vs emissions.
Why trackers showed three different numbers
The most dangerous thing on a rotation day is a stale quote wearing a live timestamp. This week gave three clean examples:
| Quote | Number A | Number B | Why they differ |
|---|---|---|---|
| Aave V3 Ethereum USDC | 3.71% live on Aavescan | 12.67% in a DefiRate permanent Sep 30 snapshot | Lag and a utilization spike frozen into an annualized point-in-time print |
| sUSDe | 3.71% trusted tracker Oct 2 (30d 4.15, 90d 3.76) | 5.28% on a second tracker | Different windows, annualization and treatment of the final ENA days |
| BUIDL | Proxies 3.47–3.78% Sep 30–Oct 1 | No official published daily yield | Proxies track nearby funds and repo rates, not the fund's own NAV series |
My rule: log source, timestamp and window with every number; trust live pool contracts over permanent snapshots at 94% utilization; never rotate on a single feed. I cross-check in the stablecoin APY tracker before touching anything.
The Friday rotation rule: five triggers
Every Friday I read five triggers. A rotation needs one that has actually fired — conviction is not a trigger — and the table is the whole rulebook:
| Trigger | Threshold | Action |
|---|---|---|
| 1. Spread | Morpho Prime median supply APR − 6-month T-bill ≥ +75bp for two consecutive Fridays | Move 10–20% of the bill sleeve; 75bp pays for Morpho's 3.3× volatility vs Aave |
| 2. Regime | IORB direction confirmed across two FOMC meetings plus a dot-plot majority | Hike regime: keep or expand bill tokens and sUSDS core. Cut regime (first cut plus dots): let bills mature into USDC and raise lending/sUSDe before SSR cuts land |
| 3. Structural | Emissions are >30% of stated APY, or an incentive ends inside 90 days | Value the vault at its base rate only |
| 4. Circuit breaker | USDe/sUSDe off-peg >1% on a major venue, or funding–staking spread negative for 5 days | Halt new sUSDe; honor the 7-day cooldown |
| 5. Friction | Round-trip gas plus spread >5% of the expected 90-day edge | Don't rotate; on a $50k sleeve the 90-day dollar edge must clear roughly $60 |
Trigger 3 already earns its keep: the Morpho Steakhouse High Yield USDC vault on Base showed 5.6% all-in, but 2.94% was base and 2.77pp Merkl incentives on roughly $449M. Post-ENA I write the base rate into the ledger and treat emissions as a maybe, per the curator checks in Morpho vault due diligence.
$50,000 worked example, dated this week
I ran a hypothetical $50,000 sleeve at each logged rate for a year, with real frictions instead of free entry and exit:
| Sleeve | Rate | Year on $50k | Friction and fine print |
|---|---|---|---|
| sUSDS | 3.60% | $1,800 | L1 round trip <$0.60 (3 tx ~$0.19 each; 0.3–2 gwei); SSR lag |
| Aave Ethereum USDC | 3.71% | $1,855 | Pool-driven; 93.9% utilization can slow withdrawals |
| Morpho Base Prime | 4.40% | $2,200 | Round-trip gas under $0.01; curator risk |
| Morpho Ethereum Prime | 5.28% | $2,640 | Vaults sized $99–102M; rate moves with utilization |
| OUSG | 3.43% | $1,715 | Bills roll ~±$125/25bp; caps $100M/$50M global, $25M per address |
| sUSDe | 3.71% | $1,855 | 7-day unstaking cooldown ≈ $36/week foregone at this rate |
Two points fall out. A 50bp edge on $50,000 is $250 a year — real, but one forced bridge, thin-pool swap or week in a cooldown can give it back. And with L1 gas under $0.20 per transaction, gas is no longer an excuse to skip a rotation that clears trigger 5; the binding costs are spread, caps, queue time and governance lag. I net them with the gas fees vs yield math in the yield calculator.
Access and liquidity are part of the yield
Headline rates ignore who is allowed through the door. sUSDS is blocked for US persons but instant for eligible users through the 1:1 USDC/USDS PSM. OUSG requires accredited-investor plus qualified-purchaser status via Ondo I LP (Rule 506(c)/3(c)(7)), a $5,000 minimum, a KYC'd wallet whitelist, and atomic USDC mint/redeem through the Instant Manager within daily caps. BUIDL is QP-only: a $5 million minimum subscription, 250k USDC minimum redemption, USD T+0/T+1 via Securitize, a 24/7 instant-USDC-at-$1 Circle facility, and monthly token dividends. Wrappers add their own restrictions — see how to buy tokenized Treasuries.
Regulatory backdrop: under GENIUS Act §4(c), payment stablecoins generally can't pay interest, so on-chain dollar yield lives in a wrapper — sUSDS, sUSDe, a lending receipt or a fund token — and that wrapper is exactly where the gates above sit. The structural case is in tokenized Treasuries vs DeFi lending.
What breaks the rule
The case study I reread most is October 10–11, 2025: USDe wicked to $0.65 on Binance and spent 23 minutes below $0.90, traced to a single-source oracle issue. Roughly $1.9B — about 13% of supply — redeemed in 24h, the Reserve Fund went unused, and Aave liquidations were only ~$47k. Protocol redemption held; reacting on Binance's print did not, and the New York Fed's Liberty Street Economics analyzed the episode on June 23, 2026. My response is pre-written: off-peg beyond 1% on a major venue trips the breaker, and I wait out the 7-day cooldown rather than sell into it.
Other breakers: a funding–staking spread negative for days removes sUSDe's reason to exist; an Aave utilization spike can turn instant redemption into queueing (withdrawal queues and redemption delays); and ENA expiry is now a known structural break, so post-ENA numbers get re-underwritten from the base rate. My peg playbook is stablecoin depeg risk.
Into the October and December meetings
Into October 27–28 I watch four things: administered rates versus the 3.90%/3.75% marks, Sky's governance calendar for SSR lag, how fast 4.08% three-month bills roll inside OUSG, and whether the Morpho-vs-bill spread holds +75bp for a second Friday. For December 8–9, the dot majority and futures strip decide whether trigger 2 stays in hike regime or flips. I pair this with my monthly portfolio review cadence, the yield allocation framework for sizing, and the best stablecoin yield 2026 league table.
If you run your own book, answer this before the next meeting: which single trigger would make you move capital this month, and is its threshold written down — or would you decide live? Mine fits on an index card; that's the whole point.
Sources and further reading
- Federal Reserve — FOMC statement, September 16 2026 — the hike to 3.75–4.00%.
- Federal Reserve — H.15 Selected Interest Rates — administered rates and the bill curve.
- Federal Reserve — FOMC calendar — the Oct 27–28 and Dec 8–9 meetings.
- Sky Money — live sUSDS Sky Savings Rate.
- Aavescan — Aave V3 Ethereum USDC — live supply APR, utilization and TVL.
- Morpho — Prime vault rates and base-versus-incentive splits.
Frequently asked questions
What happens to sUSDS and tokenized T-bill yields if the Fed cuts — or hikes?
SSR governance-set tracks IORB with a lag: after Sep 16 2026 hike (IORB 3.90%) sUSDS sat at 3.60% Oct 2; in a cut cycle it falls after a vote, not automatically. T-bill tokens reflect bills as they roll (3M 4.08% Sep 24), adjusting only as holdings mature.
Is sUSDe safe when perpetual funding goes negative?
Not "safe" like T-bill tokens; carry can shrink or reverse. Oct 10–11 2025 USDe wicked $0.65 on Binance for 23 minutes (a single-source oracle issue); ~13% (~$1.9B) redeemed in 24h with no Reserve Fund use and ~$47k Aave liquidations; redemption held, but the 7-day cooldown blocks instant reaction.
Can US residents buy OUSG or BUIDL?
OUSG requires accredited investor plus qualified purchaser status, a $5,000 minimum, and a KYC wallet whitelist through Ondo I LP; BUIDL is qualified-purchaser-only with a $5 million minimum subscription and a 250k USDC minimum redemption processed via Securitize; permissionless wrappers around either product carry their own restrictions.
sUSDS vs sUSDe — where does each yield come from?
sUSDS distributes the Sky Savings Rate from short-duration RWA/T-bill exposure (rate-policy income); sUSDe stakes USDe backed by short-ETH perp hedges + stETH and ~11%+ RWA — funding + staking income with no mechanical Fed link, organic-only after the Sep 30 2026 ENA incentives ended forever.
How fast can I redeem OUSG, BUIDL and sUSDe?
OUSG supports atomic USDC mint and redeem through its Instant Manager within daily caps ($100M global subscribe, $50M redeem); BUIDL settles USD T+0 or T+1 via Securitize plus 24/7 instant USDC at $1 through the Circle facility; sUSDe enforces a 7-day unstaking cooldown, the slowest exit of the group.
Run Friday's numbers in one place
Log sUSDS, Aave USDC, Morpho, OUSG and sUSDe side by side — live rates, windows and sources — before your next rotation.
Open the Stablecoin APY TrackerRelated reading: best stablecoin yields in 2026, Ethena vs Sky, Aave vs Morpho, real yield vs emissions, and the DeFi yield calculator.