The Fed Held. The Curve Did Not. What the Bond Market Is Pricing Now
An 8-4 Fed left rates at 3.50-3.75%. The 10-year jumped 6 bp anyway. The curve, not the statement, is the news after the April 29 meeting.
The Fed held. The curve did not. That gap is the news after the April 29 FOMC meeting, and in our view it tells you more about the next two meetings than any sentence in the statement does.
The committee left the funds rate at 3.50 to 3.75 percent for a third straight meeting, with a 8-4 vote that produced the most dissents since 1992. The headline writers are focused on the split. The bond market wrote a different story. The 2-year closed near 3.90 percent, up about 5 basis points. The 10-year rose roughly 6 bp to 4.42 percent. The 30-year ticked up to 4.97 percent. Equities barely moved on the day, with the S&P 500 closing at 7,135.95, down 0.04 percent.
The shape of that curve move is the signal. It is not a textbook bear-steepener and it is not a flight to the front end. It is the belly leading. We believe that distinction is what separates this print from a routine no-change meeting.
Why the belly led
A normal “Fed-on-hold-with-hot-inflation” tape would push the 2-year hardest, since the funds rate is the discount factor that anchors the front end. That is not what happened. The 10-year moved more than either the 2-year or the 30-year, leaving 2s10s slightly steeper while 10s30s flattened on the day. We covered the structural mechanics of this dynamic in our piece on the Treasury term premium regime, and the April 29 print is a near-clean example.
When the belly leads a sell-off without the front end matching it, the cleanest interpretation is that the move is not about expected policy rates over the next year. It is about the compensation investors require to hold long-dated paper across a wider distribution of outcomes. That compensation has a name. It is term premium, and it has been quietly re-emerging since the Hormuz news cycle began. For readers who want the underlying primer, our yield curve explainer walks through how each part of the curve responds to different shocks.
Term premium is doing the Fed’s tightening
The mechanic worth keeping in front of you is that real financial conditions tighten through long-dated yields, not the funds rate. Mortgage rates, corporate IG spreads, EM dollar funding, and capex hurdle rates all reference the long end. When the 10-year moves up 6 bp on the day the committee declines to act, the curve is doing the tightening the Fed chose not to do.
| Channel | Anchored to | What 4/29 did |
|---|---|---|
| Mortgage rates | 10-year Treasury | Tightened (10Y +6 bp) |
| Corporate IG spreads | Long-end real + credit | Tightened (real yields up) |
| EM dollar funding | Long-end + DXY | Tightened (long end +) |
| Capex hurdle rate | 10Y nominal + risk premium | Tightened (term premium re-priced) |
| Funds rate channel | Policy rate | Unchanged |
That table is the answer to why one dovish dissent could co-exist with three hawkish ones without the committee tearing itself apart. Governor Stephen Miran, who preferred a 25 bp cut, can argue plausibly that long-end yields are doing the Fed’s job. The hawks can argue plausibly that the long end is repricing for the wrong reason. Both can be correct at the same time, which is the real reason the vote split four ways.
Energy passthrough is where the long end gets the message
Brent rose roughly 6 percent to close near $118, with an intraday print near $120 on the back of the U.S.-Iran standoff. WTI cleared $106. We mapped the supply-side scenarios in detail in our Hormuz blockade portfolio impact piece, and the curve action on April 29 is the cross-asset transmission of that same shock.
History suggests the chain runs in one direction. Energy passthrough lifts headline CPI, headline lifts long-dated breakevens, breakevens lift nominal long yields, and the long end tightens conditions on housing and capex. The March CPI release from BLS showed headline at 3.3 percent year over year, driven heavily by gasoline and energy, while core CPI ran at 2.6 percent and core PCE printed near 3.0 percent. That headline-versus-core wedge is the entire story. It is also the reason the front end barely budged while the long end repriced.
We believe the EIA’s framing on inflation-adjusted oil levels matters here. The current Brent print is the highest nominal level since 2022, but in real terms it sits below the 2022 episode. The Fed’s own statement language calls inflation “elevated, in part reflecting the recent increase in global energy prices.” That is calibrated language, not panicked language, and it reads as a committee that learned a pattern from the prior cycle.
What the bond market is actually pricing
Stack the curve move against the March 2026 SEP, which still showed one cut in 2026 with core PCE projected at 2.7 percent by year-end. The April 29 curve does not contradict that median dot. It widens the distribution around it. Front-end forwards are roughly where they were two weeks ago. Long-end real yields are not.
| Tenor | Pre-meeting (4/28) | Post-meeting (4/29) | Change |
|---|---|---|---|
| 2-year | ~3.85% | ~3.90% | +5 bp |
| 10-year | ~4.36% | ~4.42% | +6 bp |
| 30-year | ~4.94% | ~4.97% | +3 bp |
| 2s10s spread | ~51 bp | ~52 bp | +1 bp |
| 10s30s spread | ~58 bp | ~55 bp | -3 bp |
That configuration is not what a market repricing imminent cuts looks like. It is also not what a market repricing imminent hikes looks like. It is what a market repricing wider tail outcomes looks like, which is exactly what term premium is supposed to capture.
Our meeting preview flagged that this would be a meeting where the press conference moved less than the curve. That call held up.
Allocator implications
For an RIA audience the framing is cleaner than the headlines. The bond market is repricing for a wider range of outcomes, not a specific one. Long-end duration is where mark-to-market pain shows up first when term premium re-emerges, and history suggests that channel runs faster than the policy-rate channel. The front-end view (anchored to the funds rate) and the long-end view (anchored to term premium) are now telling different stories.
We believe two plausible Fed paths look real from here. The first is a Fed that holds through the summer and lets the long end do the conditioning work. The second is a Fed that has to acknowledge breakeven dis-anchoring in the next statement and remove the easing bias the three hawks already voted against keeping. Neither path is a cut. Neither path is a hike. The risk lives entirely in the term premium and breakeven channel.
Skanda Amarnath of Employ America called the facts “decisively hawkish.” We would qualify that on core. The facts have moved hawkish on headline. They have not on core. That wedge is the wedge between the front end and the long end on April 29.
The next core PCE print and the May FOMC statement are the swing factors. Until then, in our view the curve is the cleanest signal in the room. It is telling investors that the Fed is boxed in on the long end while it sits still on the short end. The 8-4 vote is a footnote to that move. The same pattern, visible in the global rates complex underlying the original Bloomberg Markets Wrap, suggests the curve is the story.
Forward-looking statements reflect Ferrante Capital’s current analysis and are subject to change without notice. Actual results may differ materially from any forward-looking views expressed. Ferrante Capital LLC is a Registered Investment Adviser. This article is for informational and educational purposes only and does not constitute investment advice, tax advice, legal advice, or a recommendation to buy or sell any security.
Conflict disclosure: Ferrante Capital and its associated persons, as well as clients of the firm, may hold positions in U.S. Treasury securities, broad equity index funds, and energy-related securities referenced in this article. No specific issuer recommendation is intended. Holdings change without notice and this is not personalized investment advice.
Please consult a qualified financial professional before making investment decisions.
