Tongue twisters aren’t just for fun. As far back as the 18^th century, they were used as a teaching tool. Educators, speech tutors and elocutionists began codifying tongue twisters into formal usage; they were used in schools to teach children clear articulation, proper diction and precise pronunciation of difficult letter combinations. The first published children’s elocution book was Peter Piper’s Practical Principles of Plain and Perfect Pronunciation. Credit the Internet Archive for this twister from the book’s “P-P-P-Preface” that takes the original Peter Piper to another level: “Peter Piper, without Pretension to Precocity or Profoundness, Puts Pen to Paper to Produce these Puzzling Pages, Purposely to Please the Palates of Pretty Prattling Playfellows, Proudly Presuming that with Proper Penetration it will Probably, and Perhaps Positively, Prove a Peculiarly Pleasant and Profitable Path to Proper, Plain and Precise Pronunciation. He Prays Parents to Purchase this Playful Performance, Partly to Pay him for his Patience and Pains; Partly to Provide for the Printers and Publishers, but Principally to Prevent the Pernicious Prevalence of Perverse Pronunciation.”

The pervasive “low hire, low fire” narrative qualifies as a greatly abbreviated tongue twister to this mic drop of mic drops (say it five times fast and you’ll hear what we mean). Refocusing, Thursday’s early U.S. economic data from Challenger, Gray & Christmas and the U.S. Department of Labor reiterated the relevance of the repetitive remark from the reports’ raw results. Challenger hiring plans perfectly and plainly proclaimed:

“Employers announced plans to hire 90,787 workers in September, up from the 12,325 plans announced in August, as seasonal hiring announcements have begun. However, it is down 23% from the 117,313 announced in September 2025, and the lowest September total since 2011, when 76,551 hiring plans were recorded. Announced seasonal hiring is muted so far this year.”

End-of-year seasonal hiring patterns make for a rather lumpy picture. Applying a seasonal adjustment shows the sequencing of the last few months has taken a turn for the worse. Seasonally adjusted hiring plans fell sharply, to 18,000 from July’s 109,000 and August’s 106,000 (light blue bars). September’s showing was the lowest since August 2025’s 11,000.

“Low fire” couldn’t have been more obvious from the initial jobless claims headline. The 197,000 print didn’t just come in below the 200,000 consensus estimate, it was the third straight week under the 200,000 mark. The last time a streak of these lows occurred was in the three weeks ended October 4, 1969, as in 2,973 weeks ago.

When it comes to end-of-year hiring plans, there’s the retail sector and then there’s everybody else. As Challenger outlined, “Spirit Halloween and Michaels are the only Retailers who have announced hiring plans this year, a combined 62,000 seasonal hires. This is compared with seasonal employers announcing a combined 100,800 last September.” To best judge the direction of influence on retail payrolls, apply a seasonal adjustment. The results are striking. September’s -159,000 literally falls off the bottom of the chart (lilac bars). Last September’s seasonal hiring was also below normal; still, the -129,000 stands in stark contrast to the normal positive cadence.

No doubt, there were winners in the entrails of Challenger’s hiring plans. Four industries posted big numbers relative to their past trends: Aerospace/Defense 5,300, Utilities 4,200, Industrial Goods 3,108, and Construction 1,256. These figures landed 216%, 1,276%, 158% and 417%, respectively, above their prior 12-month averages. Altogether, the 13,864 total was a serious departure from the norm (teal line). The need for additional labor resources reflects the outlier vibes generated from defense, data centers and manufacturing capacity pressures.

The September ISM manufacturing report piled on evidence to the capacity pressure narrative. Starting with Backlogs, the current 56.4 was just the second, alongside February’s 56.6, above the 55 threshold over the last four years (green line). Supplier Deliveries, staying elevated at 59.0, sustained bottleneck risk through the conduit of longer lead times (yellow line). A comment from the machinery industry was telling: “Orders have doubled yet again, and delivery times have also doubled, in the semiconductor, electronics and government sectors, with remaining sectors flat to down. Coupled with supply chain lead times and pricing pressures, the factory backlog has nearly doubled.”

Widening Backlogs underpinned Employment’s increase to 52.7, nearly identical to July’s four-year high of 52.8 (dark blue line). ISM’s rule of thumb relative to manufacturing payrolls equates the 50.3 level to a breakeven, suggesting the 10,000-consensus estimate for today’s employment report leans in the right direction. Alternatively, a read-through to ISM orders metrics supports the notion that capacity pressures are more an internal than an external story. Headline New Orders rose to 55.3, while New Export Orders fell to 50.9 and underperformed New Orders in 23 of the past 25 months.

Price pressures, as manifest in ISM Prices Paid, broadcast the ongoing inflationary narrative across financial markets. The 77.9 beat of the 73.0 market’s expectation kept this litmus test of the global industrial supply chain at heightened heights. The eight consecutive readings above the lofty 70-level mark a rare event — over the last 30 years, a streak of this duration only has occurred 12 other times.

With that as backdrop, long-term bond yields are behaving like common moving average lines that default in Bloomberg charts. In the last five years of the 2010s, the U.S. 10-year Treasury drew a smoother path to ISM Prices (lime line). Today, the uptick through the 5% yield has tracked sympathetically with upstream cost pressures. It’s no coincidence that Germany’s 10-year Bund (aqua line) has moved, tracking ISM Prices higher for nearly two years. And Japan’s 10-year JGB looks like a technical support line for the “higher lows” dating as far back as 2023. As one of the first national price gauges each month, ISM Prices sets a tone. Right or wrong on the fundamentals, the latest installment sides with bond bears. Tongue twisters, not necessary.

Ask younger generations whom they would associate with the voice of God, and the most popular response might be Morgan Freeman for his portrayals in 2003’s Bruce Almighty and 2007’s Evan Almighty. American football fans who grew up in the 1960s and 1970s would answer differently. Widely known as the “Voice of God” by NFL Films’ fans because of his resonant style, “John Facenda could read a laundry list and make it sound like the Constitution of the United States,” said Steve Sabol, co-founder of NFL Films. His numerous memorable narrations cover a span from 1967’s They Call It Pro Football to 1974’s The Autumn Wind. The poetry of his deliberate baritone delivery was unmatched, especially from this passage: “The fourth quarter. Time is no longer a referee, but an executioner. The shadow of the stadium stretches across the field, a reminder that the sun is setting on someone’s hopes. This is when the long afternoons of summer training pay their final dividends. This is the hour of the veteran. The rookie has played his game; now, it is time for the men.”

Today’s start of the fourth quarter comes on the heels of relief on the inflation front, at least for day traders. August’s core PCE price index rose 0.247% month-over-month (MoM), rounding down to 0.2% vis-à-vis the 0.3% consensus expectation. Also unrounded to three decimal points, 12 of the 14 prior months through July were revised downwardly. Yields on the 2-year and 10-year rallied to as low as 4.82% and 5.20%, respectively, before boomeranging to close higher on the day. The core’s three-month annualized rate fell to 2.05%, the lowest since July 2024’s 2.04% (lilac line). Farther back, August 2023’s 2.03% level helps draw the Fed’s 2% target more like a support line for the short-run core inflation trend. A move through this support to a 1-handle could have created more traction for a long-lived Treasury rally. The smoother six-month and year-over-year (YoY) trends did not make the same progress as the three-month metric — the former eased mildly to 2.74% from 2.93% (green line) and the latter ran in place at 3.01% from 2.98% (purple line).

Truflation’s August core PCE projection of 0.19% MoM landed on the correct side of the consensus, as noted in Wednesday’s Feather. With the fourth quarter upon us and Truflation’s September core calculations just completed, we can see a major difference relative to past Septembers. This year’s -.012% not seasonally adjusted (NSA) MoM drop is a prominent outlier. Since its 2010 inception, Truflation’s core PCE has never posted a September sequential decline. Compared to the .229 average of the prior 16 years, the .241 MoM shortfall relative to trend could lead to a notable downside surprise once September core PCE is released on October 29^th. After the National Accounts’ revisions, Truflation’s average .04 MoM percentage point difference from February to August suggests the next core PCE figure could come in at a (rounded) 0.1% MoM pace or lower.

Cumulative inflation has clearly taken its toll — consumer spending’s fundamental fuel registered the second MoM compression in the last five months. The National Bureau of Economic Research’s coincident cycle indicator, real personal income less transfers (PILT), has transitioned from an unbroken expansion stretch in the 27 months ended January 2025 to uneven performance since. The smoothed six-month annualized rate to benchmark business cycles has run under the 1% threshold for seven straight months through August 2026. While the path treads water, the seven-month string raises caution flags. In economic cycles, the ‘7’-month mark first occurred in July 1974, August 1980, December 1982, February 1991, November 2001 and March 2008, all of which overlapped with recessions. July 2022’s observation landed right in the middle of the post-COVID inflation acceleration. Given September’s energy price surge, inflation-adjusted variables face downward pressure, making a continuation of real PILT’s sub-1% six-month trend a near certainty.

Even more noteworthy, nominal personal income’s largest influence from the benchmark revision came from personal interest income. Personal income was amended upward by $446 billion over the five years ended July 2026. July’s interest income revision, up $461 billion, accounted for more than the entirety of the aggregate adjustment. Other upward moves for personal income components came from employee compensation ($130.7 billion) and rental income ($42.6 billion). For completeness, downward offsets came from personal dividend income (-$105.2 billion), transfer payments (-$57.9 billion) and proprietors’ income (-$39.3 billion).

Interest income is not the backbone of American consumers’ budgets; it doesn’t fund their standards of living. That comes from wages and salaries, underpinned by the labor market. A check-in with ADP’s private payrolls revealed a pick-up in private job growth to September’s 90,000 pace (orange line). The three-month high revealed more than half (55,000) was generated by “recession-proof” education and health services. The goods-producing sector continued to outperform the service side (ex-education and health). Construction (15,000; thank you, data centers) and manufacturing (17,000; supply chain capacity pressures) collectively outpaced financial employment (-16,000) and professional and business services (-11,000). Leisure and hospitality also added jobs (22,000), but we wouldn’t hold our breath on sustained gains with future downside travel risks surfacing as detailed in Wednesday’s Feather.

Under the surface of ADP’s top line is a relative spread between the Private Core and the education and health industry. A vigorous economic expansion is characterized by the former’s job creation outpacing the latter; this was supremely evident in the 2010s expansion (green bars). Over the last three years, the inverted spread is indicative of the opposite and questions the ongoing subdued picture for real PILT. The building blocks of the consumer spending outlook are being driven by traditionally non-cyclical sectors. Time is no longer a referee with the upside down-ness of the private labor market at the fourth-quarter’s outset and long-term interest rates moving up and to the right.

I’m singin’ in the rain, just singin’ in the rain

What a glorious feelin’, I’m happy again

I’m laughin’ at clouds so dark up above

The sun’s in my heart and I’m ready for love

Let the stormy clouds chase everyone from the place

Come on with the rain, I’ve a smile on my face

I walk down the lane with a happy refrain

Just singin’, singin’ in the rain

Moviebuffsforever.com ranks 1952’s Singin’ in the Rain as the No. 4 best feel-good movie of all time: “Stanley Donen and Gene Kelly’s musical masterpiece is the most purely joyful film ever made. Gene Kelly’s title number – dancing through a downpour with an umbrella and a grin that seems to contain the entire history of human happiness – is the single most life-affirming image in the history of cinema. The film is funny, romantic, and technically dazzling, and it leaves every viewer feeling that the world is a brighter, more musical place than they had previously supposed. There is no film on this [top 10] list that more reliably produces a smile.”

American consumers could use a little Singin’ in the Rain joy about now. According to Conference Board, the September consumer confidence index fell to 81.9, worse than the 89.0 consensus estimate and below all 51 estimates in the Bloomberg survey. Both components – present situation and consumer expectations – weakened over the month. More importantly, September marked a meaningful milestone, taking out the identical April 2020 COVID low of 85.7 and April 2025 Trade War 2.0 low of 85.7 in one shot, rewinding confidence back 12 years. Conference Board added:

“Consumers’ write-in responses regarding factors affecting the economy were mostly pessimistic in September. References to prices, the high cost of goods and services, and oil and gas prices in particular, rose to new heights, reflecting September’s surge in fuel costs. Comments about war/conflict eased this month but remained elevated. Consumers also frequently cited politics, trade, and employment in their write-in responses, though to a lesser extent.”

U.S. households are not feeling the vibe that the Job Openings and Labor Turnover survey (JOLTS) conveyed. No doubt, total nonfarm job openings disappointed market expectations, falling to August’s 7.079 million level from July’s upwardly revised 7.335 million (previously 7.271 million). Private job openings, which drive underlying labor demand, also eased to 6.348 million, down from the prior 6.562 million pace and April’s 6.792 million local high (teal line). Despite monthly gyrations, all remain above December’s 5.828 million cycle low.

Since then, Conference Board’s Jobs Plentiful index, a steadfast job openings’ proxy, has yet to bottom, questioning whether “the labor side of the Fed’s congressional remit is in good shape” to use Fed Chair Warsh’s words. September’s 23.6 level was the lowest since February 2021 (orange line). Moreover, the -.54 correlation since December (red box) ran in the opposite direction of the .87 relationship over the entirety of the overlapping history since December 2000. Jobs Plentiful does not get revised, but the correlation breakdown flags downward revisions to job openings in train.

As expected, the Dallas Fed services Future General Business Activity weakened; it was down 7.9 points to 10.7, in line with guidance from regional bankers and manufacturers, as noted in Tuesday’s Feather. Eleventh District inflation expectations flared up: Future Input Prices jumped 13.2 points to 61.3 (yellow line), the second biggest month-over-month (MoM) advance ever; Future Selling Prices rose 6.7 points to 34.3 (green line), extending the MoM-gain streak to four months; that duration has only been matched four times since the survey’s 2007 inception.

Price pressures contrasted with future profits and current revenues. The former’s calculation comes from the Future Selling-Input Price spread tightening to September’s -27.0, the worst since June 2022’s -29.0 and markedly below its -19.8 long-run average. Current Revenues’ net -0.9 figure was the first negative print in 10 months (light blue line). The Dallas Fed has only documented falling service revenue 12% of the time. With Current Selling Prices expanding more quickly, to 14.9 from August’s 8.9, the top-line result speaks to softer volume performance driving activity into the red. Add Household Income Expectations’ downshift to a 17-month low of 2.5 to the mix (red line), and there’s no surprise in top-line compression in the Texas service sector.

The broad-based nature of September’s consumer confidence drop was evident across the income distribution. Of the eight buckets surveyed by Conference Board, seven registered MoM declines, something that’s happened just 9% of the time. Most notably, earners making $125,000 or more posted a 13.6-point fell to 93.1 (fuchsia line). Given the zeroes in their bank accounts, this group is most likely to travel. And while the top-end’s confidence drop is not yet reflected in vacation intentions, which ticked up five tenths to 42.6 (dark blue line), the directional implication is clear. Over the 13-year period shown, the only major disconnect (red box) followed a global pandemic.

It’s curious that two of the biggest vacation destination states, California and Florida, both reported outsized decreases in September consumer expectations: the former down 24.0 points (lilac line); the latter down 17.5 points (aqua line). The combined point loss of -41.5 was an outlier, the sixth worst on record. Of the other seven occurrences when California + Florida saw declines of 40 points or more, four accompanied recessions.

This observation is noteworthy despite the noisiness of the two series. Are these states already feeling the pinch of cancelled travel plans in the context of lower purchasing power and downward pressures on future income? And when will curtailed discretionary travel yield disinflationary impulses? For Federal Reserve hawks with above-target inflation in their sights, today’s PCE price gauges take precedence over all else. Truflation’s Core PCE projects a 0.19% MoM August increase, below the 0.3% consensus estimate. Even if Truflation proves accurate, year-over-year (YoY) Core PCE inflation would still come in at 3.3%, leaving only hawks Singin’ in the Rain.

How everything turns away

Quite leisurely from the disaster; the ploughman may

Have heard the splash, the forsaken cry,

But for him it was not an important failure; the sun shone

As it had to on the white legs disappearing into the green

Water; and the expensive delicate ship that must have seen

Something amazing, a boy falling out of the sky,

Had somewhere to get to and sailed calmly on.

The passage from W.H. Auden’s “Musée des Beaux Arts” was inspired by a December 1938 visit to the Royal Museums of Fine Arts of Belgium. Auden spent considerable time in Brussels reflecting on paintings by Flemish Renaissance master Pieter Bruegel the Elder. Bruegel was unique for placing miraculous or catastrophic events in the periphery of everyday, rural life. In Landscape with the Fall of Icarus, a farmer guides a horse-drawn plow, a shepherd stares aimlessly, and a fisherman sits on the shore. Tucked away in the bottom corner near a tall merchant ship is a pair of thrashing legs splashing in the water. Icarus is reduced to art’s equivalent of a footnote.

Monday’s release of the Dallas Fed Manufacturing survey was relegated à la Icarus in Bruegel’s painting. The opening paragraph of Bloomberg’s Markets Wrap column was absent any mention of the Lone Star State’s September industrial installment: “A standoff between the U.S. and Iran spurred oil-market volatility, dragging down stocks and bonds on concerns that potential inflationary pressures could trigger Federal Reserve rate hikes.”

Granted, September’s headline easing to 9.8 from August’s 11.6 hadn’t the heft to press the global risk-free rate to an intraday high of 5.27%. And yet, Icarus managed a handful of cameos. Like other Federal Reserve regions, capacity constraints were on display in Texas. Using the Empire survey as a template, four of five current metrics – Backlogs, Delivery Time, Prices Paid and Prices Received – all cleared the +1 z-score hurdle, while, at a +.2 z-score, the Average Workweek was on trend.

Unfinished Business, as in Backlogs, was the clear standout – it vaulted to 22.7, the highest since March 2021, multiples of its -2.6 long-run average, and sufficiently substantial to generate a +2.5 z-score (lime line). Though a barrage of multi-billion contracts may feel to workers at Fort Worth-based Lockheed Martin like a hurricane, Backlogs following Hurricane Katrina, at 35.1, were even higher than today’s.

Dallas’ Growth Rate of New Orders also spiked by 11 points to 19.0, well north of its -1.0 long-term trend; with a +1.4 z-score, it also intersected with 2017’s rush to source goods ahead of the following year’s Trade War 1.0 (orange line). The 11.1-point jump in Operating rates, a.k.a. Capacity Utilization, to 23.9, also put this gauge at three times its long-run average of 7.5 (aqua line).

These strains on manufacturing capacity don’t come cheap. Texas’s current margin proxy fell to a nine-month low of -24.6 (teal line), well below its -18.8 long-run trend (dashed teal line). As with other regions, the upstream profits squeeze catalyzes downstream pass-through risks, which are manifested in Future Prices Received — September’s 14-month high of 42.8 scaled to the top 9% of all observations since the survey’s 2004 inception (pink line).

The yawning divide between capacity and pricing power’s potential amidst demand destruction for most U.S. consumers drew Danielle’s eye to an old saw from her days at the Dallas Fed: “Always follow the Future Growth Rate of Orders to see where the factory sector is headed.” On that count, one might not blink at September’s 29.2-level, which is comfortably above its six-month average of 26.7. That it slipped from August’s blistering 41.0, though, puts it on our radar for signs of burnout.

Part of our broader concerns about the future stem from a turn in the lending cycle. Referencing the same source, Monday’s Dallas Fed September Banking Conditions Survey was not pretty, especially in its outlook. The current backdrop for credit standards and loan pricing set the table. The former fell to a net -10.4, the tightest in 11 months as 13.8% of Eleventh District senior loan officers reported more stringent lending terms relative to the 3.4% who indicated easing (dark blue line). Loan pricing also swung hard — September’s net 25.0 marked a material shift from the prior 24-month period of a more favorable – and decreasing – pricing environment; the 28.2-point swing from August’s -3.2 only rivaled the start of the Fed’s 2022-23 tightening regime (yellow line).

What’s to come in the lending landscape was even more jolting. Future loan demand collapsed in September to a net -1.6 from strong and steady readings in the 40s in the previous five surveys covering last December to August (green line). Not surprisingly, expectations for nonperforming loans rose to a net 16.6, the worst in 17 months (red line).

The lending cycle turn also shows Texas bankers turning downbeat on the region’s broader prospects. Future General business activity for banks dropped to a net -14.8, the first negative print since August 2024 (light blue line). For perspective, financial firms’ assessments are a higher beta proxy for those of the manufacturing and service sectors, with respective .79 and .75 correlations since the Banking Survey’s 2017 inception. Banker pessimism suggests the drop in the manufacturing outlook, from August’s five-year high of 37.2 to September’s 20.8, may just be getting started (fuchsia line).

Today’s Dallas Fed Future Service Activity could also be headed lower vis-à-vis August’s 19-month high of 18.6 (purple line). Freightwaves’ September 22^nd story hit too close to home for the epicenter of data center buildouts. Texas construction hauler, Jett Transport & Materials LLC filed for Chapter 11 protection September 14^th. Two days later, Xoco Transport, a Hidalgo, Texas-based carrier entered Chapter 11. Smaller Texas trucking firms filing Chapter 7 (liquidation) included: “T Yorkman Trucking LLC of Midland, which specialized in oil field water transportation, vacuum truck operations and hazardous-material hauling; Blue Star Transports LLC of Garland, listed with two trucks and two drivers; and Jackdollars Transport LLC of McKinney, a one-truck carrier.” These are definitely not Icarus footnotes.

‘The primary aim of modern warfare (in accordance with the principles of doublethink, this aim is simultaneously recognized and not recognized by the directing brains of the Inner Party) is to use up the products of the machine without raising the general standard of living.”

George Orwell’s 1984 is one of the few books we were forced to read in high school English that we actually remember. Wuthering Heights, Tom Sawyer, and A Tale of Two Cities have disappeared from our memories, but the images in 1984 — telescreens, Big Brother, the mandatory “Two Minutes Hate” each morning, and “Room 101” where the Party tortures dissidents with their deepest fears — have stayed. In its critique of totalitarianism, Orwell’s masterpiece astutely exposes governments’ use of geopolitical conflict as an excuse to tighten their grip on power. In 1984, the fictional country of Oceania is perpetually at war with one country or another. The novel’s protagonist, Winston, is given a book written by an underground resistance leader, from which the above quote originates.

The Iran War, which spectacularly began seven months ago with the United States and Israel’s joint Operation Epic Fury, exhaustingly drags on. Oscillations between peace talks and renewed aggression have become trite, with the latest barbs traded at last week’s United Nations summit. With no end in sight, markets anticipate higher fuel and energy costs for the foreseeable future. In the University of Michigan’s (UMich) final data for September, the Median Year-Ahead Expected Change in Gasoline Prices rose from 19.8 to 29.7 cents costlier per gallon. (For comparison, in February that number was just 0.2 cents!)

If there’s one sector that’s benefitted from the outbreak of conflict, it’s Defense. Friday’s Durable Goods report revealed that Defense Capital Goods Shipments (defined as the sum of small arms and ordnance, communications equipment, aircraft, missiles, space vehicles & parts, ships & boats, and search & navigation equipment) climbed to a record high $19.698 billion (red line). This is unsurprising after Defense Capital Goods New Orders hit a cycle high in June of $23.38 billion, and were only slightly off that peak in August, at $21.549 billion (blue line). In fact, the current level of Orders has just one historical precedent: June 2000’s outlier $25.893 billion result, which was skewed by a flurry of contracts being finalized at the same time by the Pentagon. Today’s six-month streak north of $20 billion is unparalleled. Furthermore, that Orders have run above Shipments since February suggests the latter has yet to top out this cycle.

Shipments of artillery, weapons, vehicles, and equipment require massive raw material inputs like earth metals to produce. Unsurprisingly, New Orders for Primary Metals and Fabricated Metals have historically trended with Defense Shipments, posting tight correlations of 0.81 and 0.86 since 1992, respectively. In August, Primary Metals New Orders rose to a record $32.6 billion, a 14% increase since February (green line); the six-month increase of $3.92 billion equates to a +1.58 z-score in data back to 1992. Meanwhile, at $44.972 billion, Fabricated Metals, were just shy of July’s record high of $45.572 billion (yellow line). Last week’s barn-burner S&P Global Mfg PMI surely benefited from a juiced Defense sector, but also laid bare firms’ concerns about input costs and needing to pass those through to their customers. Chief hawk on the FOMC, Beth Hammack, alluded to her concerns on that front this past Friday. In a roundtable on “Inflation Drivers and Dynamics,” she stated her concern that an “inflationary mindset” is setting in among consumers and businesses who’ve come to expect higher prices as the rule, rather than the exception.

Fear of prices rising into the future has contributed to pull-ahead purchasing activity. Core Non-Defense Capital Goods Orders, which proxy business investment, far exceeded Bloomberg’s 0.5% month-over-month (MoM) consensus and rose 1.6% in August. Meanwhile, Non-Defense Shipments rose 0.6% MoM and are now up 11.4% year-over-year (YoY), appreciably higher than their long-run average of just 2.7% (purple line). While the Non-Defense numbers look impressive on their own, they pale in comparison to Core Defense Shipments, which are running at a 31.6% YoY pace (pink line). Though down from May’s 48.3% YoY record high, the current pace is leaps and bounds above the long-run trend of 3.5%. In fact, they’ve been running above 25% YoY for nine straight months. For comparison, there are just six other months since 1993 that broke this threshold, all of which were one-off outliers.

The last chart in today’s quad illustrates the historical relationship between Defense and broader Industrial Production. During the Korean War, the ratio between the two peaked at 2.26, and then hovered near 2.0 during the Vietnam War and 1.9 during the Reagan Administration ahead of the collapse of the Soviet Union. The current 1.21 is nowhere near those historical heights, but it has steadily risen from a cycle low of 1 in June 2022 (aqua line). Manufacturing is traditionally a cyclical sector more attuned to changes in interest rate policy, but Defense-related production is less exposed to this vis-à-vis non-Defense. (After all, why angst about passing through higher input costs when your customer is Uncle Sam?) Unless an end to this conflict magically presents itself, the upward trend in this ratio has further to run.

While last week’s S&P Flash PMI was red hot, and helped spur Treasury yields to multi-decade highs, a more mixed bag has emanated from recent regional Philly, New York, and Richmond Federal Reserve factory surveys. This said, the margin squeeze narrative has been consistent across these releases. For now, as the Iran War segues to status quo, we’re seeing Defense’s outsized strength meet up with activity pulled forward earlier this year as companies scrambled to stockpile. Will further Fed tightening do anything to slow the red-hot Defense sector? No, but it will far harder yet on rate-sensitive commercial manufacturing. Anyone who’s read 1984 knows that Orwell would call that “doublethink”.

Uber was founded in 2009 by computer programmer Garret Camp and his friend Travis Kalanick. When they started the business in San Francisco, the original name was ‘Ubercab.’ Because of legalities raised by local cab drivers, Ubercab was a short-lived moniker. The city’s cabbies claimed Uber was operating as an unlicensed taxi company, so the Camp/Kalanick duo dropped ‘cab’ and rebranded to ‘Uber’ in 2011. DoorDash was born in 2013 by four Stanford University students — Tony Xu, Andy Fang, Stanley Tang and Evan Moore. Originally launched as a prototype under the generic name, PaloAltoDelivery.com and an outgrowth of a student project, its goal was to help local small businesses deliver food without hiring their own staff. The founders went so far as to go the last mile, personally delivering the first orders around Stanford’s campus. Both disruptors have come a long way from their early days. As of yesterday’s trading, Uber boasted a $141 billion market capitalization, while DoorDash’s was more than half that, at $81 billion.

Uber and DoorDash didn’t just disrupt taxi services and local delivery; in the process, they distorted a critical U.S. economic metric and varsity leading indicator – jobless claims. We’ve written extensively about the opportunity cost of driving over collecting an unemployment check. Today, we attempt to account for Uber’s and DoorDash’s employment rise in the context of jobless claims’ fall.

Our starting point (we always have one) is the Quarterly Census of Employment and Wages (QCEW). Industry granularity drills down to the two most specific designations under the Transportation and Warehousing sector. The standard six-digit classification (i.e., NAICS code) of 485310 – Taxi and Ridesharing Services — covers the employment of app-based passenger transportation, ride-hailing arrangement platforms and traditional taxicabs. Meanwhile, NAICS 492210 – Local Messengers and Local Delivery — compiles headcount from businesses providing local, intra-city delivery services of light cargo, food and small parcels. Since QCEW measures W-2 employees, it doesn’t capture 1099 independent contractors. Our best stab at counting these workers comes from the Household Survey’s self-employed in the Transportation and Warehousing industry.

The combined raw, not seasonally adjusted (NSA) proxy for Rideshare/Delivery employment showed a sustained structural shift up and to the right from the middle of the 2010s expansion. Before Uber and DoorDash – call it the decade starting 2000 – this job count ran between 400,000 and 500,000 on a 12-month moving average basis. The most recent tally through 2026’s first quarter QCEW saw the figure rise to more than 1 million (aqua line). A closer inspection shows a count of roughly 700,000 immediately before the pandemic hit. The subsequent ~300,000 add is one factor that helps explain the lack of a further increase in NSA continuing claims (purple line) after the post-COVID normalization.

Pay differences for rideshare and delivery vis-à-vis an unemployment check have widened significantly over time. Presently, on a four-quarter rolling basis, rideshare’s $1,366 average weekly wage (dark blue line) and delivery’s $1,178 comparable figure (orange line), are multiples of the $479 average weekly jobless benefit payment (fuchsia line). To our dismay, it was more profitable to rideshare right after the pandemic, when drivers opted to stay home and collect government stimulus checks. Even after the normalization, the “earnings gap” between unemployment insurance and Uber ($900 more) and DoorDash ($700 more) remained significant over a monthly budgeting period. Before the disruptors, taxi drivers made somewhere between $200 and $300 more vis-a-vis jobless benefits, while delivery only proffered a $100 to $200 differential. Today’s math though is plain — it doesn’t pay to be unemployed.

Taking a turn to heed the other entries on yesterday’s economic calendar, September’s Kansas City (KC) Fed manufacturing survey added evidence to the widening capacity-pressures’ theme touched on in recent Feathers. Applying the Empire Manufacturing Capacity/Cost/Price Pressure hurdle for +1 z-scores for Current Backlogs, Delivery Times, Prices Paid, Average Workweek and Prices Received, KC Fed data cleared the hurdle for all five metrics for just the 14^th time in 303 months, a 5% hit rate. The profit squeeze in the 10^th District echoed other regions too. The KC Fed Manufacturing Current Margin Proxy (Prices Received – Prices Paid) fell to this month’s -31, down from August’s -19 which is also in line with its long-run average (red line).

KC’s fusion of higher capacity pressures and an escalating profit squeeze presents higher pass-through risks from upstream to downstream. At September’s four-year high of 59, Future Prices Received is one of the highest levels on record, only comparable to 2008’s oil shock during the Great Recession and COVID’s supply chain bulge (green line). These two comments from regional executives speak to pass-through from separate angles:

• “Business continues to weaken and costs continue to rise. Our workers deserve higher wages, but the profit isn’t available to support more.” (Translation: Raise prices to pay higher wages)
• “Grave concerns about energy costs and the impact on cost/pricing. Prices to customers are going to go up and go up significantly if something doesn’t change quickly. We are looking at a huge new round of inflation. Perhaps even scarcity of supply. Very tenuous at the moment.” (Translation: Raise prices to cover higher costs.)

As was illustrated in Thursday’s daily, another option is to increase leverage to pay for either higher wages or higher costs. But as the yield curve attested again yesterday, borrowing is getting pricier too.

Elsewhere on Thursday, and inconveniently for Chair Warsh, the Law of Supply and Demand remained conspicuous in the new home market. Historically, high months’ supply catalyzes year-over-year (YoY) declines in both average (-8.8% YoY, yellow line) and median (-5.8% YoY, light blue line) new home prices. August’s damage marked the deepest since August 2024 and July 2025, respectively, levels that rivaled the 2000s housing bust episode’s discounting. Housing gurus at Zelman & Associates observed that “the pipeline of soon-to-be started specs ticked higher to another all-time high this month, suggesting builders appear poised to reaccelerate spec production if demand improves.” Builders have shovels up, but mortgage rates north of 7% and long-end Treasury (and global) yields still on the rise are disruptors in their own right.

We know that disorderly conduct erupts when sports teams win championships. But the modern-day can’t hold a candle to the Nika Riots. Procopius of Caesarea’s History of the Wars describes: “The news was brought to the emperor that the people were gathered in the Hippodrome…Thereupon Belisarius, drawing his sword, charged into the midst of the vast crowd, cutting down those in his path. Mundus also, who was standing near the entrance, rushed in…The result was that more than thirty thousand of the populace perished on that day.” This account from January 532 AD described the most destructive civil uprising in Byzantine history. The violence began at Constantinople’s Hippodrome, where rival chariot-racing factions, the Blues and Greens, united in fury against heavy taxation and corruption under Emperor Justinian I. The mobs launched a week-long rampage across the imperial capital. Rioters burned down state buildings, including the Senate House and the original Hagia Sophia, and declared a rival nobleman emperor. As Justinian prepared to flee, Empress Theodora convinced him to hold his ground. Dispatching Generals Belisarius and Mundus, their troops trapped and slaughtered an estimated 30,000 rioters inside the Hippodrome.

The entire U.S. yield curve shifting up between 10 and 17 basis points (bps) across tenors doesn’t convey Nika Riot mayhem. One trading session, a trend does not make. String several of these days together, though, and we may be facing a rate tantrum. Investors were already vexing a disorderly rise in bond yields, assigning it the top tail risk in September’s Bank of America Global Fund Manager Survey. Now, the MOVE index is testing the 100 level for the first time since Spring 2026’s Trade War 2.0 episode.

S&P Global’s U.S. manufacturing and services PMIs induced the one-day disarray. September’s manufacturing PMI came in at 57.0, 3.3 points above the 53.7 market expectations, and the services PMI advanced to 58.7, 2.9 points north of the 55.8 consensus call. Both deltas were record upside deviations, rendering the combined 6.2-point beat also unprecedented. The better-than-expected growth impulses were accompanied by other details via S&P Global that could have made even Emperor Justinian’s central banker angst about being behind the curve:

“Employment also rose sharply, with jobs added at a pace not seen for over four years, as firms sought to meet rising demand. However, backlogs of work continued to rise at an increased rate and supply chain delays intensified, pointing to a lack of operating capacity which fed through to higher prices. Input costs meanwhile surged higher on the back of the recent spike in energy prices, adding to a worsening inflation picture.”

Widening Delivery Times on the manufacturing side rose to September’s 61.5 on an ISM scale (aqua line), the highest since July 2022 and well north of the 55.2 long-run average. Readings at 60 or higher qualify as outliers accounting for 12% of instances since the 2007 inception. Longer lead time has created urgency to source products in the here and now. In turn, manufacturing’s Quantity of Purchases jumped to 57.2 (orange line), taking out the previous 2026 high in June, at 56.3 and leaving the sourcing guide at its strongest since April 2022. This level intersected the post-COVID bulge and ran near the high points of the 2010s expansion. Composite Backlogs rising to 54.9 punctuate the capacity pressure theme, they’ve had no rival before the pandemic (purple line).

Costs continue to run above prices, keeping pass-through risk alive. Manufacturing Input Prices increased to 68.1, the highest in three months (green line). Meanwhile, service Input Prices leapt 7.2 points to 66.1, the largest one-month increase in service PMI history not affected by pandemic disruptions (yellow line). Output Prices were relatively tame in terms of monthly volatility: manufacturing down two tenths to September’s 59.2 (dark blue line) and services up 1.2 points to 57.0 (red line). Both though stood above their respective 54.9 and 54.2 long-run averages.

Higher selling prices should be in train, otherwise manufacturers and service providers will confront acute margin compression. Margin proxies, defined by the spreads between output and input prices, for each sector fell sharply in September – manufacturing to -8.9, a four-month low, well below its -4.8 long-run trend, and services to -9.2, a three-year low nearly three times its -3.4 long-run average. In sum, z-scores for manufacturing and services were -.8 and -1.8, respectively (lilac and teal lines).

The reflex for any business crippled by a profit squeeze is to raise prices. But not all pricing power is created equal. Those with it can pass through higher costs to preserve margins; those that lack it must resort to cost cuts. Upstream production channel manufacturers fall in the first bucket, while downstream distributors in wholesale, retail and consumer-facing areas land in the second.

What’s a pricing-powerless company to do? Borrowing more is a viable option. At least that’s what the trends in commercial and industrial (C&I) loans portray. Against the sudden plunge in the composite margin proxy (lime line), nominal and real C&I loan growth is expanding at year-over-year rates of 10.4% and 6.8%, respectively. Several years ago, American businesses were faced with a similar dilemma (first red arrow), but back then it was early in the Fed’s tightening cycle. Accelerated loan growth built a bridge until the composite margin proxy normalized.

Today’s markets risk disorderly rate conduct making it more expensive to fund more borrowing. More evidence of heightened capacity pressures could cause additional spikes in rate volatility which ultimately would spill into the credit markets. Keep a close eye on the U.S. Treasury 5s30s spread, which appears to be in freefall and is just 40 basis points shy of inversion.

“The first little pig was very lazy. He didn’t want to work at all, and he built his house out of straw. The second little pig worked a little bit harder, but he was somewhat lazy too and he built his house out of sticks…The third little pig worked hard all day and built his house with bricks…It looked like it could withstand the strongest winds. The next day, a wolf happened to pass by the lane where the three little pigs lived; and he saw the straw house, and he smelled the pig inside. He thought the pig would make a mighty fine meal, and his mouth began to water.”

Americanliterature.com’s excerpt from “The Three Little Pigs” was adapted from Flora Annie Steel’s 1922 English Fairy Tales. It’s a given that each pig knew to build the house’s most essential part: the roof. Even cavemen knew this hard rule. Accidentally discovered in 1965 when a farmer was expanding his cellar, Ukraine’s Mezhyrich archeological site unearthed dwellings made of mammoth bones covered with animal-skin dome-shaped roofs dating back approximately 15,000 to 18,000 years.

Roofs need not be built over the New York Fed’s thrice-a-year Survey of Consumer Expectations (SCE) Household Spending Survey. But yesterday’s August 2026 installment revealed future spending performance foresees the need to prepare for “essentials over housing.” Consumers believe they’ll be spending more on three household budget staples – transportation (that includes gasoline), food, and utilities – and less on housing.

For perspective, all three essentials categories saw the median expected change in spending over the next 12 months reach respective cycle lows of 3.2%, 4.8% and 3.7% in December 2024 (orange, fuchsia and aqua lines). Over that same interval, future housing spending also reached a low of 2.7% (dark blue line). Some 20 months on, August 2026 showed a sharp advance in expected transportation spending (to 4.7% from December 2025’s 4.1%) and similarly elevated levels of funds to be diverted to food (5.4%) and utilities costs (4.6%). The need to shuffle spending built upon April’s record one-point drop for future housing spending, falling another two-tenths to 2.9%. This was the set up going into the Fed’s rate hike and piles on to the housing sector’s future woes.

Housing spending isn’t the only thing expected to get a haircut. Fewer travelers are flying according to the Transportation Security Administration (TSA). Passenger volume measured by TSA checkpoint travel numbers has fallen on a year-over-year (YoY) basis in the five months ended September (green bars). Not since 2020’s global pandemic shutdown have we witnessed a losing streak of such duration. Higher jet fuel prices are ramping up operating costs and to compensate, airlines have raised checked bag fees. But don’t expect that to be the end of travelers’ price squeeze. Both American and United are cutting capacity. These excerpts were telling, per Fortune.com:

American Airlines CFO Devon May: “‘You’re just going to want to pull a little capacity out when we see a rise in fuel like we’re seeing right now’…the fuel spike has added $1 billion to the company’s projected fourth-quarter expenses, prompting it to cut some December flights and plan for less growth next year.”

United CFO Mike Leskinen: “‘There’s some marginal routes that don’t make sense in a higher fuel environment, so we cut them. We’re flying to maximize profitability and free cash generation, so we’ll make those adjustments…Jet fuel price gets passed through with a lag’…there’s room to pass on higher fuel costs to consumers eventually.”

Higher airfares should temper future hotel rate inflation. Thus far in September, Smith Travel Research’s U.S. hotel average daily rate (ADR) concurs. After the World Cup and America’s 250^th lifted the ADR YoY trend to June’s 6.9% peak, it’s since cooled to September’s month-to-date 2.4% (light blue line). Isolating this month’s second week, though, ADR fell 1.3% YoY. This nods to an earlier 2026 Labor Day, which pulled activity into the first week, to 6.5% YoY, at the expense of its second week. Demand showed even greater noise via Occupancy, which went from a 9.2% YoY gain in the week ended September 5^th to -4.7% YoY in the week ended September 12^th.

Should ADR continue to print in negative territory, after expanding in every week since the beginning of March, it would imply something more fundamental is at work. The SCE Housing Spending Survey’s upper-income vacation expectations had already pointed in that direction, but we’re only now seeing it. The seasonal nature of vacation plans means past comparisons must be made vis-à-vis the same reference period. Earlier this spring, we saw April 2026’s 47.7% four-month forward probability run below both that of the 50.4% 2021-25 average and the 49.1% average since the 2014 inception, when the artificially low 24.7% April 2020 COVID-infected number is taken out of the calculation. August did see some improvement among higher-income earners, who saw expectations rise to 48.9%, their highest August tally since 2023’s 49.3%. But, to United CFO Mike Leskinen’s point, jet fuel prices do get passed through with a lag. The well-to-do may be less exposed than those flying in the back of the plane, but a prolonged regime of higher airfares from higher jet fuel costs and capacity cuts suggests additional declines in travel plans that manifest as further hotel rate inflation compression

Any downward adjustments to discretionary demand would translate faster into overall consumer pricing than from other nondiscretionary goods and services. Moreover, higher airfares and lower hotel rates would create tension under the core PCE umbrella. For now, Truflation’s core PCE trend, at 2.13% YoY thus far in September, is already on its way to 2% (yellow line). This real-time alternative to the official core PCE inflation rate (3.34% in July) is out ahead in the inflation reporting department (Can we get a task force for that?). And Truflation’s version has a stout .89 correlation to the actual core over the time span we’ve illustrated and suggests “policy error” could become a new market narrative.