NextFin News - The Federal Reserve is holding interest rates at 3.50% to 3.75%, the bond market is repricing from cuts to hikes, and the 10-year Treasury yield is trading near its highest level in nearly three years. Through all of it, one variable is doing most of the work. Ketaki Sharma, founder and CEO of Algorithm Research, said in a television interview on September 4 that inflation is the main factor to watch in assessing the bond market and the central bank's next move. It is a deceptively simple framing, and it is exactly the point: after a year of shifting narratives, the entire rates complex has narrowed to a single question — whether inflation data forces the Fed's hand before growth forces it to blink.
The Setup: A Hold That Leans Hawkish
The policy backdrop explains why Sharma's read carries weight. At its July 29 meeting, the Federal Open Market Committee kept the federal funds target range unchanged at 3-1/2 to 3-3/4 percent, but the decision was far from unanimous. Nine members supported holding; three preferred an immediate 25-basis-point increase. The Board of Governors set the interest rate on reserve balances at 3.65% and the primary credit rate at 3.75%, effective July 30. The statement framed the pause as a balance between a labor market near maximum employment and core inflation still above the 2% target, with energy prices adding fresh pressure.
The committee's own projections had already shifted before the July meeting. In June, officials raised their 2026 inflation forecast to 3.6% on the headline measure and 3.3% for core, which excludes food and energy — up from 2.7% for both in the March update. The median estimate for the fed funds rate at the end of 2026 moved to 3.8%, up from 3.4%, signaling that at least one rate increase is likely before year-end. Of the 18 officials who responded, nine projected at least one hike, eight saw rates unchanged, and one still anticipated a cut. Growth was penciled in at 2.2% and unemployment at 4.3%.
Markets have moved in the same direction. Earlier in 2026, investors assigned roughly a 35% probability to a rate increase and were broadly positioned for cuts. After the July meeting, that positioning flipped toward one to two hikes by the end of the year. The 10-year Treasury note yield reached 4.78% on September 4, up 0.17 percentage points over the prior month and 0.71 percentage points from a year earlier. On September 2 the 10-year touched 4.81%, its highest level since late 2023, while the 30-year bond yield climbed to 5.286% and the 2-year note sat near 4.39%. The front end of the curve is anchored by the Fed's 3.50% to 3.75% range; the long end is being priced by something else entirely.
That gap is where Sharma's comment lands. The Fed controls the policy rate. The bond market controls the term premium. When the two diverge, the bond market is usually telling you what it thinks the Fed will eventually have to do.
Why Inflation Is the Only Variable That Matters
Inflation has returned to the center of the policy debate for a concrete reason: core inflation is sticky at 3.3% year over year, well above the Fed's 2% target, and the most recent monthly reading came in at 0.2% — in line with forecasts but not falling fast enough to clear the target on any comfortable timeline. Core inflation excludes food and energy by design, but energy prices feed through transport, production costs, and inflation expectations, so a sustained energy shock can lift headline inflation even when core is merely stuck rather than accelerating.
The transmission runs through two channels, and both are active. The first is the expectations channel: when households and businesses expect higher prices, they act in ways that produce them — workers demand larger wage increases, managers build inventory ahead of anticipated cost rises, and buyers pull purchases forward. The second is the term-premium channel: investors demand more compensation for holding long-dated debt when they are uncertain whether the Fed will stay ahead of inflation. That is why the 10-year yield can climb even while the policy rate sits on hold. The Fed can hold the front end steady and still watch financial conditions tighten through the long end.
Institutional investors are reading the same signal. Mike Mitchell, head of US Treasury and inflation trading at Goldman Sachs, noted that with a month to go until the Fed's next rate decision, inflation is increasingly in focus for the US bond market. He pointed to real yields in the 10-year sector running near 2.5% and closer to 3% in the 20-year sector — historically elevated levels that indicate the market is pricing a persistent inflation premium, not a transitory one. After the July inflation print, he observed that roughly 9 basis points of movement were priced into the September meeting, a sign that policymakers are not being forced into an immediate decision but are also not being handed room to cut.
"With a month to go until the Federal Reserve's next rate decision, inflation is increasingly in focus for the US bond market."
Sharma's framing cuts through the noise by refusing to treat every data point as a separate story. Jobs, spending, production, and confidence all matter, but they matter to the Fed through the inflation function. If they push prices up, they become relevant. If they do not, they are background. That is a harsh filter, but it is the filter a committee with a 2% inflation mandate and a dot plot pointing to higher rates is forced to use.
Cyclical Shock or Structural Shift: The Call That Decides the Trade
This is the question that determines whether the bond rally investors hoped for earlier in the year simply arrives late, or does not arrive at all. If 2026's inflation pressure is mostly cyclical — an energy spike, a supply-chain kink, a temporary labor squeeze — then it will mean-revert. The Fed will not need to hike much, real yields will fall back, and the 10-year yield will drift toward the 3.75% level that some year-end forecasts still target.
The evidence for the cyclical read is genuine. Energy shocks are episodic by nature; the risk premium embedded in oil prices can unwind as quickly as it appeared once geopolitical tensions ease. Core services inflation has historically cooled as labor markets soften, and unemployment at a projected 4.3% leaves room for that adjustment without a sharp rise in joblessness. The Fed's own near-term pricing — roughly 9 basis points implied for September — says policymakers do not feel cornered into an immediate hike. A committee under genuine pressure does not price in 9 basis points.
But there is a structural layer underneath the cyclical noise, and it is the more important one. Real yields near 2.5% on the 10-year are not just an inflation story; they are a supply and term-premium story. Large fiscal deficits require more Treasury issuance, and more issuance means investors need a higher yield to absorb it. That dynamic is not fixed by a single soft inflation print. It is a regime change in how the bond market prices government debt — the end of the ultra-low real-yield era that defined the 2010s and early 2020s.
The honest call is that both forces are at work, and conflating them produces the wrong conclusion. The cyclical leg — energy prices and near-term data — will decide whether the Fed hikes in September or waits until later in the year. The structural leg — deficits, issuance, and term premium — will decide where the 10-year yield settles over the next several years. An investor who bets on mean reversion because energy is cyclical may be right about the next three months and wrong about the next three years.
History offers a warning. In the 1970s, the Fed repeatedly treated inflation spikes as cyclical and temporary, pausing after each round of tightening only to find inflation re-accelerating. The pattern broke only when policy accepted that the problem was structural and held rates high for long enough to reset expectations. The 2026 setup is not a repeat of the 1970s — labor markets are tighter, expectations are better anchored, and the committee is more transparent. But the lesson holds: the cost of calling a structural shift cyclical is paid in credibility, and credibility is the one asset a central bank cannot reprint.
The Second-Order Effect: When the Bond Market Tightens Policy for the Fed
The first-order read is straightforward: higher inflation means higher rates, which hurts bonds. The second-order question — the one most market commentary skips — is what higher long-term yields do to the rest of the financial system, and how that loops back to the Fed.
The Fed sets the policy rate, but it does not set mortgage rates, corporate bond spreads, or equity discount rates. Those are set by the long end of the curve. When the 10-year yield rises on term premium rather than on growth expectations, financial conditions tighten without the Fed taking any action at all. Mortgages reprice higher, cooling housing activity. Corporate borrowing costs rise, squeezing capital expenditure and refinancing plans. Equity valuations, especially for long-duration growth stocks, compress because cash flows far in the future are discounted at a higher rate.
This creates an asymmetry the Fed has to navigate. If it hikes into a bond-market-driven tightening, it risks overdoing it — tightening policy twice over, once through its own rate and once through the curve. That is how well-intentioned tightening turns into an unnecessary recession. But if it holds steady while the long end sells off and inflation expectations drift higher, the whole exercise loses credibility and the market will demand an even larger move later. The committee is walking a line where both mistakes are available.
That is why Sharma's framing matters beyond the next rate decision. Watching inflation is not just about guessing the next 25-basis-point move. It is about reading whether the bond market believes the Fed will win. A 10-year yield at 4.78% with a 2.5% real component is the market saying it is not yet convinced. Until that changes, the path of least resistance for long-term yields is sideways-to-higher, regardless of what the Fed does at the front end.
The Counter-Thesis: Growth, Not Inflation, Is the Real Constraint
The strongest case against the inflation-first view is that the Fed's binding constraint is not prices but growth. Unemployment is projected at 4.3% and GDP growth at 2.2% for 2026 — a soft-landing setup, not an overheating one. If the labor market cracks, inflation becomes a secondary concern very quickly, and the market will pivot back to pricing cuts regardless of what the latest inflation print says. Rate cuts priced for 2027 in economist surveys reflect exactly this view: a Fed that holds now to avoid reigniting prices, then eases once the lagged effects of restrictive policy show up in hiring.
There is also the argument that the Fed has already done the heavy lifting. With the policy rate at 3.50% to 3.75% and real rates positive, policy is genuinely restrictive. Another hike may do more damage to employment than to inflation, which is precisely why three officials dissented in favor of raising rates immediately rather than waiting — they judged that the cost of moving late exceeds the cost of moving early. The opposite view, that patience is the safer path, rests on the belief that growth will slow enough to do the disinflationary work without additional tightening.
Both views are internally consistent, which is why the falsifying signal matters more than the opinion. The threshold is specific: if core PCE prints at or above 0.3% month over month for two consecutive months, the cyclical-cooling thesis is wrong and the structural-inflation read takes over. At that point, the bond market's current positioning — and the Fed's patience — would both be under pressure. The July reading was 0.2%, in line with forecasts, so the signal has not fired. But it is the number to watch, because it is the number that would prove the softer view wrong.
What Comes Next
The base case is that inflation stays sticky but does not accelerate, the Fed holds through the September meeting, and the 10-year yield trades in a wide range around current levels — elevated enough to keep pressure on rate-sensitive sectors, but not high enough to break something. The upside case for bonds is a clean string of soft inflation prints that lets the market reprice back toward cuts and pulls the 10-year yield back toward the low 4% range. The downside case is a second energy shock that pushes headline inflation higher and forces the Fed's hand, sending the 10-year toward 5% and the 30-year above 5.5%.
Short term, the next two inflation prints and the September FOMC meeting are the catalysts. Medium term, Treasury supply and the term premium are where the structural story lives — watch issuance plans and auction demand. Long term, the question is whether the era of ultra-low real yields is over, and the 2.5% real yield on the 10-year says the market is already voting yes.
Inflation is the variable everyone is watching. The price of that inflation — the term premium the bond market demands for believing the Fed — is the story everyone should be watching.
Explore more exclusive insights at nextfin.ai.
