Newsletter · · Ashutosh Agarwal

Fed Signals Patience After Cool PCE as 10-Year Yield Hits 5.30% - US Macro Recap - Week of October 2, 2026

US Macro Recap for the week of October 2, 2026. Podcast synthesis on a cooler-than-expected August core PCE and the methodology debate behind it, Williams and Jefferson signaling the Fed can wait as October hike odds fell to about 35%, the 10-year Treasury yield still hitting 5.30%, firm labor data and the immigration break-even ahead of payrolls, the soft landing versus reflation versus stall-speed debate, and the trades in play across bonds, the S&P, the yen and EM carry.

US Macro Recap

Week of October 2, 2026: Fed Signals Patience After Cool PCE as 10-Year Yield Hits 5.30%

US Macro Recap, Friday, October 2, 2026

On Tuesday the question was how much damage the Fed would do to bring inflation down. By Wednesday morning the Fed had two reasons to wait. New York Fed President John Williams said there was "no urgency" to hike again. Then August core PCE came in cooler than expected. Odds of an October hike fell from about 70% to roughly 35%. The odd part is the bond market. Short-term yields fell, but the 10-year still touched 5.30% on Thursday, its highest level in roughly a quarter century. Payrolls come out at 8:30 this morning, and the debate has shifted. The question is less "will the Fed hike in October" and more "why won't long-term rates come down even when the news is good?"

TL;DR

  • The inflation report was soft, but it had an asterisk. Core PCE (the Fed's preferred inflation gauge, which leaves out food and energy) rose 0.2% in August against 0.3% expected. The year-over-year rate was 3.0%, below the 3.3% forecast. Revisions knocked about 36 basis points (0.36 percentage points) off the annual rate. Part of that came from a government methodology change backdated to 2021, which Wells Fargo's Gary Schlossberg says lowered inflation "by a couple of tenths."
  • The Fed blinked, and the hike odds collapsed. Williams (Tuesday) and Vice Chair Philip Jefferson (Thursday) both signaled the Fed can wait. October hike odds fell to about 35–37%, and the current base case is a pause in October and a hike in December.
  • The long end didn't get the memo. The two-year Treasury yield fell to around 4.83%, but the 10-year hit 5.30% and the 30-year hit 5.64%. Over September the 10-year rose 55 basis points. With Q2 GDP revised up to 2.2%, claims at 197,000 and private payrolls beating forecasts, the story is now a strong economy and a big bond supply. It is no longer just about the Fed.

What's new

1. Core PCE came in cool, and skeptics say the methodology did part of the work. (Mix: a bank economist, an investment strategist, an inflation specialist, commentators.) On Bloomberg Surveillance (September 30), Bloomberg's Alexis Keenan read out the numbers. Core PCE rose "two-tenths of a percent month over month… less than the three-tenths expected," and was "up 3% versus estimates for 3.3%." Headline PCE was 3.4% against 3.7% expected. NAB's Skye Masters on NAB Morning Call (September 30) pointed to the revisions as the real story. They took "around 36 basis points off the annual rate," against the roughly 25 the market expected. The three-month annualized core rate (the last three months' pace, stretched over a year) is "now at 2%… down from 2.6%." Not everyone took that at face value. On Marketplace (October 1), Wells Fargo Investment Institute's Gary Schlossberg said the statistical change to categories like software and investment services would lower inflation "by a couple of tenths of a percent." Under the old method, headline inflation would have been "around 3.6 or 3.7 percent." Omar Sharif of Inflation Insights said "more than half the basket is rising at an inflation rate of 3 percent or higher." He thinks the Fed may "hike one more time in Q1." On Wall Street Unplugged (September 30), commentator Frank Curzio made the pundit version of the argument: the 0.3% monthly headline rise came in "exactly as expected," and "it's just the yearlies down because of the new way that they calculate this measure." Why it matters: the Fed watches the three-month trend closely, and 2% annualized is exactly its target. Even if you discount the methodology change, the monthly pace stopped getting worse.

2. The Fed's own messaging turned, and the market moved fast. (Fed officials, reported by commentators and bank strategists.) On The Rundown (September 30), Public.com's Zaydad Mani said Williams' "relatively dovish speech" on Tuesday, saying there was "no need for urgency," pushed October hike odds from "around 70%… down to roughly 50%," and then to 35% after the PCE report. The Financial Exchange's Tucker Silva on The Financial Exchange Show (September 30) had CME futures at "a 65% probability that the Fed will not hike" in October. That is a 15-point move in one morning. Williams didn't sound finished, though. His co-host Mark Vandetti said Williams "told us all, I expect Fed funds… to be at 4," and quoted the speech: "monetary policy cannot move ships or reopen pipelines and refineries," but "it can diminish the risk that these supply shocks spill over." On Thursday, NAB's team on NAB Morning Call (October 1) said Vice Chair Jefferson, who votes on rates, said "pretty much what John Williams said… maybe we can wait," while still "hammering the point" that "inflation has been too high for too long and risks are all tilted upwards." Marketplace's reporting summed up where things landed: economists and markets "are betting the Fed will pause rates this month and hike rates in December." Why it matters: two of the Fed's most influential voices now lean toward "later, not now." That is a real shift from Tuesday's "three to four more hikes" pricing.

3. Long-term rates kept rising anyway, to a roughly 25-year high. (TV market reporters, a bank strategist, a trader.) On Schwab Network (October 1), Kevin Green said the 10-year "hit 5.3% overnight." He said those are levels "we haven't seen" in "20, 22, 24 years." NAB's Masters (NAB Morning Call) put the 30-year at 5.64% intraday and added up September: "US 10 years are up 55 basis points. And US 2s are up 54." Market pricing for the fed funds rate in the middle of next year has moved from 4.20% to about 4.80%. She also flagged the Bank of England's warning about "elevated hedge fund leverage" in government bonds: "I think that's not just an issue for gilts… it is causing increased volatility in interest rate markets." Patrick Ceresna of Big Picture Trading, on Macro Voices (October 1), said the softer inflation report "offered relief to the short end, but really didn't move the long end." His description: "when you have a 30 trillion dollar freight train moving this way it needs some form of a catalyst." AllianceBernstein's chief economist Eric Winograd on Bloomberg Surveillance offered the most useful frame. "When the Fed was cutting rates, long bond yields were going up. And now that they've raised rates, long bond yields are still going up… So that tells us that this cycle isn't really about the Fed." His list of other causes: "a 6% of GDP budget deficit in an economy that is strong," and "tariffs on, tariffs off… The Treasury Department changing its issuance calendar. We're intervening in FX markets. All of those argue for higher risk premium" (the extra yield investors demand for holding long-term bonds). Why it matters: a dovish Fed helps two-year yields, mortgages and car loans only at the margin. If long rates are driven by deficits and supply, the Fed can't fix them.

4. The job market shows few cracks ahead of this morning's payrolls. (TV market reporters, bank economists, a wealth-management host.) Kevin Green on Schwab Network gave Thursday's jobless claims: "197,000 first-time filers," with continuing claims at "1.701 million." He called them "as strong as I ever remember seeing in my career." Consensus for today's report is "something around 90,000" jobs and a "4.1% unemployment rate," after ADP's private payroll count also came in at 90,000. NAB's team (NAB Morning Call) added Challenger job cuts of 43,000, "the lowest September job cuts since 2022… down about 20% on a year ago." Masters (NAB Morning Call) noted ADP beat a 75,000 forecast. RBC senior US economist Mike Reed on Bloomberg Surveillance went looking for weakness in the less-watched numbers, like aggregate hours and part-time work for economic reasons. "It's really hard to find any weakness… at a 4-1, it's hard to argue there's weakness in the labor market right now." Tuesday's job openings report was, in Marketplace's words on Marketplace (September 29), "low hire, low fire." George Washington University economist Leah Brooks called hiring "meh." ZipRecruiter's Nicole Bachaud warned that "if you find yourself unemployed, it might take six plus months to find something." The Van Wie Financial Hour team on The Van Wie Financial Hour (September 28) set the bar for real alarm: recession signals would need claims "above 250 potentially reaching 300." That is also why the Sahm rule, a recession signal that trips when the three-month average unemployment rate rises half a point above its low of the past year, is nowhere near triggering with unemployment expected to hold at 4.1%. Why it matters: a payroll number near 90,000 sounds modest. Read alongside the immigration math below, it means the job market is not loosening.

5. Immigration: 90,000 jobs a month may be above the new break-even. (An operator CIO, a bank rates team, a data journalist, a TV host.) The useful number this week is the gap between the payroll consensus and the break-even pace (the monthly job gains needed to keep unemployment steady). Conning North America CIO Cindy Boyu, on Credit Exchange with Lisa Lee (September 25), put that break-even low. "We have little to no immigration. So we are at that equilibrium where we really don't need to hire much more than 50,000 or 75,000 people each period." BMO's rates team on Macro Horizons (September 25) made the same point from the other side: "trend job growth is running comfortably above estimates of the monthly break-even pace of employment," so "the Fed will remain comfortable with raising rates further." The data journalist on Excess Returns (September 26) framed it as a supply problem: "this labor crisis is not about the unemployment rate. It's about the labor force… the supply of workers dwindling because of the backlash against immigration." On Bloomberg Surveillance, a host linked this to the inflation target itself: "We're at 3% core PCE… We're reshoring all this stuff. We're cutting immigration. So we've got some wage inflation… Maybe that's the new number." Winograd answered that the Fed will likely "be satisfied if inflation runs between 2 and 2.5% as long as inflation expectations stay contained," but needs to "reinforce their credibility" first. Why it matters: if today's number comes in near 90,000, that is above Boyu's 50,000–75,000 break-even. A low-immigration economy can post small headline numbers and still be tight, and that keeps a December hike in play.

The debate

All three camps showed up this week. The soft-landing camp gained the most ground, thanks to the PCE report and the Fed's own tone.

Camp A: Soft landing. Inflation is sticky but not speeding up, so the Fed can go slowly. (An asset-manager economist, a nowcasting firm founder, and the Fed itself.) Winograd (Bloomberg Surveillance) made the cleanest case. "When the Fed embarks on a tightening cycle, typically it's because inflation is going up. That's not really what we're seeing here. We're just seeing it not come down." The Fed is "not trying to slow the economy… They're just trying to accelerate this process of inflation convergence." Because rate hikes take "nine to 12 months" to bite, "that's a recipe for a constrained, gradual cycle rather than a rapid, aggressive one." Apurv Jain, whose firm builds real-time economic estimates from alternative data, said on Macro Hive Conversations With Bilal Hafeez (October 2) that "growth is above trend and we have been seeing that for the last few months." On inflation, "the second order effects that everyone's concerned about are mostly not that high… the corporate pricing intentions… are not so high. Like the wage pressures are not so high." NAB's three-month core PCE figure of 2% (NAB Morning Call) is this camp's best single number.

Camp B: Reflation. The economy is running too hot, and bonds aren't done. (A bank economist, a macro research firm, a bank rates team.) RBC's Mike Reed (Bloomberg Surveillance) looked past the soft headline: "we got a weaker number than expected, but I think that's due to methodology." He sorts inflation into three buckets. Housing disinflation is "largely done for this year." Core services excluding housing sees "really little disinflation" with the labor market this strong. And core goods face freight costs that look like 2020 "in terms of a year-over-year change." His conclusion: "inflation is moving in the wrong direction." One Bloomberg host did the math aloud: a "6.1% GDP price index" plus 2.2% real growth is "8% simplistic nominal GDP. That's a banana republic." Darius Dale of 42 Macro, on Bitcoin Magazine Podcast (October 1), said his models put fair value for long Treasury yields at "somewhere around six and 6.15%. So we got another like 80, 90 basis points to go." He credits "a massive, massive amount of stimulus coming from both the AI CapEx bubble" and last year's tax law, which is pushing businesses "to allocate the marginal dollar… from OPEX to CapEx." BMO's Macro Horizons team said they would be "looking for opportunities to fade a dovish overreaction in the front end of the yield curve" (that is, bet against short-term yields falling too far).

Camp C: Stall speed. The consumer is already cracking, and more hikes risk a recession. (A bank economist, a NASDAQ economist, a consumer-lending CEO, commentators.) Reed is in this camp too, which shows how close together the camps are. His warning sign is non-mortgage interest payments (credit cards, auto, student and personal loans) as a share of disposable income, now "about 2.5%." "The number that concerns me is 2.8%. If you take out the COVID recession, the past three recessions prior to that, when you've hit 2.8%… we've gone into recession." On spending running ahead of income (+0.9% against +0.2% in August): "That's not sustainable." NASDAQ economist Michael Normile on IBKR Podcasts (September 29) argued that rates barely touch today's inflation drivers. "AI capex in general, it's not rate sensitive. And businesses that are rate sensitive, they're already showing that rates are restrictive." So "a more significant rate hike cycle would be riskier." The hosts of The Morning Market Briefing (September 30) gave the commentator version: "the idea that rising oil… causes all things to go up double digits is just not true because what happens is people… stop buying stuff. I mean, mortgage applications were down 6% week over week."

The honest read: Camp A won this week's data. Camp B owns the bond market. A 5.30% 10-year after a soft inflation print is the strongest evidence anyone has that something besides the Fed is driving rates. Camp C has the best early-warning indicators (Reed's 2.8% line, the consumer confidence numbers below), but none of them has tripped yet. Today's payroll report decides which camp gets December.

The trades in play

Operators putting money to work:

  • Short stocks, maximum long bonds, a little long yen (Andy Constan). On The Julia La Roche Show (September 29), Constan said he is at "75% of my max risk target… short equities, long, long-term bonds each," with "a slight long in the Japanese yen and a slight long, very slight long in short-term interest rates. I'm long a SOFR contract" (a futures bet that short-term rates fall). "I got to max long bonds on Friday into that sell-off. And of course, it's going in my face this morning. But you can't pick the bottom." His case for bonds: real yields (yields after expected inflation) are now "two and three quarters to three and a quarter on the long end," unlike 2020, when bonds "had no upside." His case against stocks is his "not enough pie" argument. AI companies' implied earnings add up to more than GDP can support: "these companies are claiming more than 100 percent of the pie." Financing the build-out is "a bunch of people saying, I will pay you Tuesday for a hamburger today."
  • A protective S&P put through year-end (Patrick Ceresna, Big Picture Trading). On Macro Voices, Ceresna laid out buying "the January 14, 2027 SPX put option at the 7,430 strike," about 3% below the market, "quoted at 136 index points." That costs "just under 2% of the index notional," with a break-even near 7,300. His reason is weak market breadth: "the median stock in the S&P 500 is 16% off of its 52-week highs," and only 20% of stocks are above their 50-day average. A drop of "even 150 S&P points" could set off systematic selling worth "tens of billions of dollars." On currencies he sees the euro, having broken "below the 113 handle," heading "down to 110 or even 108." On gold, it "will be very likely that we can even retest the 4,000 level," and it needs to hold "north of 4,500" to turn bullish again.

Strategist views on instruments:

  • Dollar-yen has peaked, and EM carry trades are unwinding (State Street FX strategist). On Street Signals (October 1), the guest strategist said "165 would be top of the range" for dollar-yen, with a possible "move… below 150 even by the end of this year, and then naturally… back down towards 140." The reason: US officials have been unusually "outspoken… about wanting to drive dollar-yen lower" while the Bank of Japan keeps raising rates. Carry trades (borrowing in low-rate currencies to buy high-yielding ones) are unwinding. Mexico's rate advantage over the US is "only about 250 basis points now," the peso's year-to-date gains have been "completely unwound," and "Dollar Colombia is almost 10% off the lows."
  • Stay away from bonds until 6% (Darius Dale, 42 Macro). On Bitcoin Magazine Podcast, Dale said "we wouldn't touch bonds with a 10 foot pole until we at least get back to fair value," which his model puts around 6–6.15%. That is the opposite of Constan's trade.
  • Fade the front-end rally (BMO). As noted in the debate, the Macro Horizons team sees "a durable bull steepening trend" (short-term yields falling faster than long-term ones) as "premature."
  • Hyperscaler credit is holding up (Odd Lots). On Odd Lots (October 1), the market-strategist guest noted that even as yields exploded, "hyperscaler credit spreads, none of them are at their wides of the year at the long end." He said hyperscaler investment-grade bond issuance in September was "a zero… they got their business done ahead of time."

Read-throughs

  • Corporates: AI earnings keep the index up while consumer names fall. On Odd Lots, the guest cited Goldman Sachs: "about half of earnings growth this year is expected to be just driven by hyperscaler CapEx." Market breadth is so narrow that the last comparable reading "was literally the day after the dot-com bubble peak," though he noted breadth "was also that bad in 1998" and the market rose for two more years. Consumer stocks tell the other side. Nike "was a $180 stock in 2021. It's 35 now," and McDonald's "peaked on March 3rd," around when the Iran war began. Micron shows the AI side. Per Schwab Network, it reported revenue of "$54.23 billion up… 379% year over year" with "gross margins, 87%." Curzio (Wall Street Unplugged) flagged the earnings-season risk: Dick's Sporting Goods lost "30% of its value in a day" after a guidance warning, and McDonald's CEO said before a shareholder meeting that "things aren't that good."
  • GDP: about 5% growth, mostly from business investment. Q2 GDP was revised to 2.2% from 1.5% (NAB Morning Call). For Q3, a host on Bitcoin Magazine Podcast noted the Atlanta Fed's live estimate has consumer spending "flat relatively to Q2" but business investment at "11.6 annualized… up four points from Q2 and three times that of consumer spending. So the growth story right now… is really more a story of AI investment." Normile (IBKR Podcasts) said real business investment in AI categories is "growing at a 20% year-over-year rate" for three quarters, while every other category "has been negative for seven straight quarters." There is a caveat on how much of this shows up in US GDP. Oxford Economics' Bernard Yaros said on Marketplace that "a lot of the AI spending by businesses is imported from abroad," which boosts Taiwan and Korea more. The Van Wie team (The Van Wie Financial Hour) noted the S&P Global flash manufacturing PMI (a business survey) "jumped to 57… the highest level since 2022."
  • The K-shaped consumer: the wealthy are starting to wobble too. The Conference Board's consumer confidence index fell to "the lowest level in more than 12 years" (Marketplace). The new detail: households earning "$125,000 to $150,000… saw the sharpest decline." Bates College economist Paul Shea warned that "if this is a sign that the higher-income households, who account for more of the consumption… are starting to lose a little bit of faith, that could be a real red flag." Case Western economist Jonathan Ernest said people are "treating this like the rainy day," with savings rates falling. Normile (IBKR Podcasts) put numbers on it: tax refunds were "about 15% bigger this year," but Oxford Economics finds the extra gasoline cost "is about to surpass" that boost, and "the savings rate [is] down to 3%." Jain (Macro Hive) sees the same split in real-time data: "aggregates are fine and anxiety for the bottom third is extremely elevated," which is "not a good sign for the midterms for the incumbents."
  • Consumer credit: an operator says the cycle is ending. (Operator.) David Johnson, CEO of the loan servicer Vervent, said on Money School Elite (October 1): "it really feels to me like we're kind of at the end of a consumer credit cycle." He described the late stage as the point "where delinquencies are rising… where these kind of, you know, emergency situations start popping up." His firm issues subprime credit cards, and he said "the top of the K is doing just fine. They make their card payments… at the bottom of the K. These are the subprime group… they've got more issues." He added that he is seeing "a lot of new entrants" and called it "the end of one cycle and at the beginning of another." Reed's 2.5%-heading-to-2.8% interest-burden ratio (Bloomberg Surveillance) is the macro version of the same warning. It "didn't shift lower when the Fed was cutting."
  • Housing takes the long-rate hit. NAB (NAB Morning Call) said 30-year mortgage rates had their "biggest jump since October 2022" and are "now up to 7.28 percent." Because mortgages follow the 10-year, not the Fed, a pause in October gives home buyers little relief.
  • Energy: oil supply is better than feared, but diesel is still the pressure point. Oil analyst Rory Johnston on Facts vs Feelings with Ryan Detrick & Sonu Varghese (September 30) estimated "around 13 and a half million barrels a day flowing through Hormuz relative to a pre-war normal of around 21 million." He calls himself "more of a Hormuz half empty kind of guy," because shipping on that route costs "around twenty five dollars a barrel… normally two dollars." US diesel crack spreads (the refining margin over crude) hit "$115 a barrel" in mid-September, against a pre-2022 normal of about $20, and are "around $90" now. If the US banned diesel exports, he warned, "global ex-U.S. diesel markets would just go haywire." On the geopolitics, NAB (NAB Morning Call) reported the US is adding a third carrier group, Trump reportedly expects "to resume bombing by the end of November," and Iran has floated restoring nuclear inspectors in exchange for sanctions relief.
  • Cross-asset: Europe is now the inflation problem. Masters (NAB Morning Call) noted German inflation at 3.3%, eurozone inflation expected at 3.7% with risk of 3.9%, and markets pricing "57 basis points of tightening by the ECB." That reverses the usual pattern. The US prints softer inflation while Europe, which relies on imported energy, prints hotter. It's one reason Ceresna expects the euro to keep falling.

What changed

Tuesday's recap ("10-Year Yield Hits 5.27% With PCE and Payrolls Due") had October hike odds near 70% and core PCE expected at +0.3% for the month. Here's what moved since:

  • Core PCE came in under forecast. +0.2% for the month and 3.0% for the year, against 0.3% and 3.3% expected. The three-month pace fell to 2%.
  • The Fed's tone flipped. Williams went from last week's "don't look through supply shocks" to "no urgency." Jefferson echoed him. October hike odds fell from about 70% to about 35–37%, and the consensus path is now a pause in October and a hike in December.
  • The 10-year went higher anyway. It went from 5.27% to a 5.30% intraday high. The two-year fell to around 4.83%. The front end and the long end are now moving in opposite directions.
  • Growth was revised up. Q2 GDP went from 1.5% to 2.2%, and August spending rose 0.9%.
  • Labor data held firm. Claims 197,000, ADP 90,000 against 75,000 expected, Challenger cuts at a four-year September low. Job openings were steady.
  • Consumer confidence hit a 12-year low, and the decline spread to upper-middle-income households.
  • Oil eased, from Brent about $107–$108 on Monday to roughly $103 midweek, with WTI just under $91 on Thursday. Mortgage rates still rose to 7.28%.