Live Market Rates

The Fed's rate-cutting cycle did more than pause this week: the market began pricing the chance it reverses. Treasury yields, the interest the government pays to borrow, climbed to their highest of 2026, led by the 2-year note (a read on what traders expect the Fed to do over the next year or two), which rose 18 basis points (hundredths of a percent, so 0.18 points) to 4.31%. Its longer cousins followed, the 10-year settling at 4.67% and the 30-year at 5.15%. The trigger was not domestic data but a spike in oil prices tied to Middle East escalation, which markets read as fresh fuel for inflation and therefore reason for the Fed to hold or move higher.

Here is the tell. The move was concentrated at the front end (short-maturity notes most sensitive to Fed policy) rather than the long end (10- and 30-year bonds, which track long-run inflation and growth). Front-end yields rise when traders push out or cancel expected cuts, and that is what happened: CME FedWatch, which infers Fed odds from futures prices, now shows roughly a one-in-three chance of a hike at next week's meeting and essentially no chance of a cut. A month ago the debate was when the next cut would land. The question the week raised is whether 'higher for longer' has quietly become 'higher, full stop.'

Fed officials handed the market the script. Governor Christopher Waller argued that if inflation ran hot, the committee would need to consider tightening 'in the near term,' a firm line with the July 28-29 meeting days away. It lands on real numbers: CPI still sits near 3.7% year over year and core PCE, the Fed's preferred inflation gauge, near 3.4%, both comfortably above the 2% target. For yields to reverse, the oil move would need to fade or the labor market to crack. June payrolls at plus 57,000 were already soft, but one weak print has not outweighed the inflation picture.

Two things stayed calm amid the drama. The 10s-2s spread (the gap between the 10-year and 2-year yields, a classic recession gauge that turns negative, or 'inverted,' when short rates top long ones) flattened to 36 basis points but held firmly positive, so the bond market is signaling caution, not contraction. And the overnight funding market barely flinched: SOFR (the Secured Overnight Financing Rate, the benchmark for what it costs to borrow cash against Treasuries overnight) sat at 3.64%, and Prime held at 6.75%, because both only move when the Fed actually moves its target rate. The action was all in expectations, not in the cash rate the Fed controls today.

The yield curve steepened this week, and the Fed had little to do with it. Front-end rates, the short maturities most sensitive to Fed policy, sat still: SOFR, the Secured Overnight Financing Rate that measures the cost of borrowing cash overnight against Treasuries, held at 3.63%, and Prime stayed at 6.75%. All the action was at the long end. The 30-year Treasury yield rose 12 basis points (hundredths of a percent, so 0.12 points) to 4.98%, and the 10-year added 9 to 4.49%. When long rates climb while short rates hold, the curve steepens, and that shift usually says more about supply and inflation than about the next Fed meeting.

Start with the front end, which barely moved. The 2-year Treasury yield (a read on where markets think the Fed is headed over the next year or so) rose just 5 bps to 4.14%, even after June payrolls landed at a soft 57,000 against roughly 115,000 expected. A cold jobs report would normally drag short yields lower. This one did not, because the Fed is not cutting: with CPI at 4.27% and core PCE at 3.41%, both well above the 2% goal, June meeting minutes showed nine of eighteen officials open to a hike this year. New Chair Kevin Warsh, asked whether he would revisit the 2% target, said 'that is the Federal Reserve's long-held objective.' Cuts are not in the conversation; the fight is hold versus hike.

So the long end had to find its own reason to move, and it did. Long yields reflect expectations for growth and inflation over decades plus a term premium, the extra yield investors demand to lock up money for thirty years rather than roll short-term paper. That premium is climbing. A July 9 auction of 30-year bonds cleared at 5.058%, the highest since 2007. When an auction goes weak, meaning investors demand a higher yield to absorb the debt on offer, it signals the market is straining to digest supply, and between federal deficits and a wave of corporate borrowing for AI buildout, there is plenty of it. That, not the Fed, is what steepened the curve.

The cleanest read is the 10s-2s spread, the 10-year yield minus the 2-year, which widened 4 bps to +0.35%. When it is negative the curve is inverted, short rates sitting above long ones, a classic recession warning; it turned positive earlier this year and keeps climbing. The steepening here is not the friendly kind that arrives when the Fed cuts into a slowdown. It is the term-premium kind, driven by heavy supply and sticky inflation. For it to reverse, the long end needs softer inflation data or lighter issuance. Neither turned up this week.

The strange part of the week: a jobs report weak enough to end the debate about a rate hike, and yet Treasury yields rose anyway, with the long end leading. That is a bear steepener (long-term yields rising faster than short-term ones, so the gap between them widens), an odd response to soft data. Payrolls grew just 57,000 in June against a 115,000 forecast, and May was revised down to 129,000. On most weeks that print pulls yields lower. Instead the 30-year yield rose 12 basis points (hundredths of a percent, so 0.12 points) to 4.98%, and the 10-year rose 9 to 4.49%. Something other than growth was setting the price of long money.

Start with the front end, the short-maturity yields most sensitive to the Fed. The 2-year Treasury (a read on what markets expect the Fed to do over the next year or two) rose 5 basis points on the week to 4.14%, even after dipping to 4.13% on the jobs report itself. Two forces pulled against each other. Inflation is still hot: headline CPI ran at 4.27% year over year and core PCE, the Fed's preferred gauge, at 3.41%, both well above the 2% target. That kept alive a question that would have sounded strange a year ago, namely whether the Fed's next move is a hike rather than a cut. The weak labor data pushed back, and the net was a small rise.

The long end saw the real move, and its reasons sit further out in time. Thirty-year yields track long-run inflation and growth rather than the next Fed meeting. June meeting minutes, released this week, showed officials divided, with many warning that strong demand for AI infrastructure could sustain upward pressure on prices for technology and electricity. New York Fed President John Williams called AI-driven demand his main inflation concern: 'If this creates a sustained impulse to demand relative to supply in inflation, I do think that's the kind of situation where you don't look through this.' Supply did not help. A soft 30-year auction (when the Treasury sells bonds and buyers demand higher yields than expected, a sign of thin appetite) capped a run of weak long-bond sales.

The result is a curve telling two stories at once. The 10s-2s spread, the gap between the 10-year and 2-year yields, widened 4 basis points to 0.35%. When that gap is negative the curve is inverted, a classic recession warning that short-term rates sit above long-term ones. It is positive now and steepening, which reads as: the Fed stays put in the near term, while the inflation problem is a longer-run one the long end has to price. For that to change, the market would likely need a cool CPI print to take the inflation scare down a notch, or a labor market that cracks hard enough to swing the Fed toward cuts rather than hikes.

The week's real question is whether a suddenly soft labor market hands the Fed a reason to stop fretting about a rate hike. June payrolls came in at 57,000, roughly half the 115,000 economists expected, and revisions erased another 74,000 from April and May. The 2-year Treasury yield (a read on where markets think the Fed is headed over the next year or two) fell 12 basis points (hundredths of a percentage point, so 0.12 points) to 4.07%. That is the market saying a July hike, which had been a live worry, now looks off the table.

Here is the awkward part. Inflation is not cooperating. CPI is running at 4.27% year over year, a three-year high, lifted in part by energy costs tied to the conflict in the Middle East. The Fed held its target at 3.50% to 3.75% on June 17 in a unanimous vote, and new Chair Kevin Warsh made the priority plain: "The commitment to deliver is strong, unanimous, and unambiguous, and that's I think an important message we've missed for five years." Some officials had been eyeing a hike, not a cut. A weak jobs print does not erase hot inflation; it just makes tightening into softening demand a harder sell.

The move was concentrated at the front end, the short-maturity yields most sensitive to Fed policy. Within that group the 2-year fell 12 bps, the 10-year eased only 8 to 4.38%, and the 30-year barely budged, down 3 to 4.87%. Long yields answer to long-run inflation and growth more than to the next meeting, and on that score little changed this week. The 10s-2s spread (the gap between the 10-year and 2-year yields, and a classic recession signal when it turns negative, or "inverted") sits at a positive 0.31%, up 4 bps. That widening comes from the front end dropping faster than the long end, the opposite of the inversion that flashed warnings in prior cycles.

What would flip the story is the inflation data, not the jobs data. The unemployment rate actually fell to 4.2% in June, but for an unflattering reason: the labor force participation rate dropped to 61.5%, its lowest since early 2021, so the rate improved because people stopped looking for work, not because they found it. Markets read the whole report as mildly disinflationary and trimmed hike odds accordingly. Should June CPI, due July 14, run hot again, the front end could hand back this week's rally in a hurry. For now floating rates sat still: SOFR (the Secured Overnight Financing Rate, the benchmark that prices much floating-rate debt) held at 3.62% and Prime stayed at 6.75%, both anchored to a Fed that is not moving.

The week's defining move was a flatter yield curve, meaning the gap between short and long term Treasury yields shrank, and it tells a coherent story about a Fed that just changed its tune. On June 17 the Federal Open Market Committee held its target rate at 3.50% to 3.75%, the fourth straight hold, but the accompanying dot plot (each official's anonymous projection for where rates belong) flipped hawkish. The median 2026 estimate jumped to 3.8% from 3.4% in March, with nine of nineteen participants now favoring at least one hike this year and six of those wanting two. Markets spent months betting on cuts. That wager is now off the table.

Short-dated yields took the hint. The 2-year Treasury, which mostly reflects what investors expect the Fed to do over the next year or two, rose 6 basis points on the week to 4.11%. That is the front end of the curve, the maturities most sensitive to policy. It moved up because the probability of a near-term cut, the thing that would pull short yields down, has largely evaporated. The May reading on the personal consumption expenditures price index did nothing to argue otherwise: headline PCE ran 4.1% over the year, the hottest since 2023, with the core measure at 3.4%.

The long end did the opposite, and that is the genuinely interesting part. Yields on the 30-year fell 7 basis points to 4.86% and the 10-year slipped 2 to 4.41%, even as data confirmed inflation above 4%. Long maturities care less about next quarter and more about where inflation and growth settle over a decade. A central bank willing to hold rates high, or raise them, to break inflation is one whose long-run price stability looks more believable, which pulls those distant yields down. Warsh leaned into exactly that. Asked about the 2% goal, he said the two belongs to the left of the decimal point, with zero to the right, a tidy way of insisting inflation should start with a 2, not a 4. The bond market took him at his word.

Put the two ends together and you get a flatter curve. The gap between the 10-year and 2-year yields, known as the 10s-2s spread and watched closely because a negative reading (an inverted curve, where short rates exceed long ones) has preceded past recessions, narrowed 8 basis points to 0.30 points. It is still positive, so no recession flag here, just a market pricing a Fed that means business on the front end and is trusted on the back end. What would change the story is simple to name: a cool inflation print, or any hint from Warsh that the hike talk was a bluff, would let cut bets and the front end fall again. Neither arrived this week.

The week's real news was not that the Fed held, but that it changed its mind about which direction it might move next. Policymakers kept the target range at 3.50% to 3.75% for a fourth straight meeting, a unanimous 12-0 vote and Kevin Warsh's first as chair. The surprise sat in the projections, the dot plot (each official's anonymous forecast for where rates go): nine of eighteen now pencil in at least one 2026 hike, and the median end-year rate moved up to roughly 3.8% from 3.4% in March. Warsh framed the 2% goal bluntly, calling the Fed's commitment "strong, unanimous, and unambiguous" and saying it has "missed for five years, and we're going to fix that."

The backdrop explains the tone. May CPI ran at 4.27% year over year, a three-year high, with energy prices up 23.5% over twelve months as the conflict with Iran pushed oil higher. That is the kind of figure a new chair cannot ignore, even if much of it is a supply shock the Fed cannot drill or refine away. So the hawkish dots double as a credibility signal: a central bank that has overshot its target for half a decade telling markets it will not let an oil spike serve as the excuse.

Here is the tension. The Fed all but threatened a hike, and the bond market faded it. On the announcement the 2-year Treasury yield (a read on where markets expect Fed policy over the next couple of years) jumped about 11 basis points (hundredths of a percent, so 0.11 points) to 4.15%, then handed the entire move back to close at 4.05%, down 8 on the week. CME FedWatch, which infers Fed odds from futures prices, still puts roughly 89% on a hold in July. The market heard the hawkish dots and judged them posture, not a plan.

The disconnect comes down to growth. Q1 GDP grew just 1.6% annualized, payrolls added 172,000 in May, and unemployment sits at 4.30%. A market braced for a hike would drive the 2-year well above the 3.63% funds rate; instead it sits about 40 basis points above, the shape of a long hold rather than tightening. Longer maturities say the same: the 10-year fell 10 basis points to 4.43% and the 30-year dropped 19 over the month to 4.93%, the move you get when traders bet that tighter policy plus an energy tax on consumers cools the economy. Even SOFR (the Secured Overnight Financing Rate, the benchmark under most floating-rate commercial loans) only firmed 4 basis points to 3.63%, going nowhere fast.

What would turn the dots into action is core inflation, which strips out food and energy. Core PCE, the Fed's preferred gauge, sat at 3.29% in April, and a renewed climb there would mean the energy shock is leaking into the rest of the economy, the outcome Warsh said he wants to prevent. Absent that, the market is betting the Fed talks hawkish and sits still. A dot plot is a forecast, not a promise, and this week the two halves of that sentence pulled in opposite directions.

The hottest inflation print in three years landed this week and barely dented the Fed's path. Headline CPI rose to 4.27% year over year in May, the highest since 2023, yet almost the entire jump came from energy: gasoline climbed about 7% on the month as the Iran war disrupted Middle East oil supply. Strip energy out and core inflation held near 2.9%, calmer and broadly in line with forecasts. That split, a scary headline over a steady core, is why a 4-handle CPI number did not blow up the rate outlook the way it might have a year ago.

The clearest evidence sits in the shape of the Treasury curve. Short maturities (the front end, most sensitive to what the Fed does next) sold off hard: the 2-year yield rose 18 basis points (hundredths of a percentage point, so 0.18) over the month to 4.13% as traders erased their remaining bets on 2026 rate cuts. Long maturities (the long end, which tracks expectations for inflation and growth over decades) hardly budged, with the 30-year at 5.03%, up just 5 basis points on the month. Read together, the bond market is saying the Fed will stay put longer, not that inflation has permanently reset higher.

Overnight funding stayed anchored, as expected. SOFR (the Secured Overnight Financing Rate, the benchmark cost of borrowing cash overnight against Treasury collateral) held at 3.60%, and Prime, the bank reference rate behind many business and consumer loans, sat unchanged at 6.75%. Both move only when the Fed moves the policy rate, and with a hold near-certain this month, neither had a reason to budge. The 30-day average of SOFR even ticked down 5 basis points, a reminder that the trailing averages lag the overnight rate rather than predict it.

What would actually change the story is breadth. An energy shock is a supply problem, and rate hikes do little to bring barrels back online, so the Fed can look through it as long as it stays contained. If core services or rents started reaccelerating, or pricier fuel bled into everything from airfares to groceries, the long end would move and the last cut bets would vanish. For now the yield curve is positive: the 10s-2s spread, the gap between the 10-year and 2-year yields that turns negative (inverted) before most recessions, sits at 0.42%. It flattened a touch as the front end climbed, but it is nowhere near flashing red.