← All notes
SUNDAY RESEARCH

The week ahead — September 27, 2026: Growth hot enough to buy an October hike

Published September 27, 2026 · 23 min read · RSS

Forward-looking scenarios, archived as published. This note states no forecast, target or trade, and is not updated after the fact.

The week ahead has two prints that set how much of an October hike gets locked in. August PCE Wed 9/30, core consensus +0.3% on the month; Micron reports the same day. September payrolls Fri 10/2, previews 85k to 100k, unemployment 4.1%. Futures price October near 75%.

COIL MACRO, September 27, 2026: Growth hot enough to buy an October hike. Standing view — week neutral at 55%; month neutral at 55%; quarter neutral at 55%; year neutral at 50%. Wrong if SPY's Friday 2026-10-02 closing price prints outside 761.57 to 779.37.

The house view

Into a week with August PCE and September payrolls, futures already price an October hike near 75%, so the misses carry more information than the hits. Last week showed where the tightening lands: on bonds and small caps, while the AI-linked large caps kept rising on growth the same data confirmed.

  • CME FedWatch priced an October 27-28 hike at 75.8% as of September 25, up from 57.6% a week earlier, after a Fed governor said further increases are likely (CNBC, September 23).
  • The ten-year hit its highest yield since June 2007 on Thursday and ended the week near 5.16%; the two-year rose more than 13 basis points to 4.9% on September 23 alone (CNBC).
  • September's flash composite PMI printed 58.4, the fastest growth in over five years, with input costs rising at the fastest pace since October 2022 (S&P Global, September 23).
  • On our bars, QQQ is up 3.87% in September and IWM down 4.07%, a 7.94-point gap between the two.
  • Core capital goods orders rose 1.6% in August against forecasts near 0.6% and are up 14.1% on the year, the strongest since August 2021 (Bloomberg, Wolf Street, September 25).

Week horizon: 2 right, 0 wrong, 2 mixed over 4 graded notes. Month, quarter and year: 0 graded, 12 too\_early across those same 4 notes. The archive holds 4 prior notes, dated 2026-08-30, 2026-09-06, 2026-09-13 and 2026-09-20.

The setup

Range. SPY enters the week above its 10, 20 and 50-day averages and a little over half a percent under its sixty-session ceiling at 779.37, so the range holds with the index pressing its top. Underneath, the split has widened into the widest monthly gap of the year on our bars: QQQ carries every average with a daily RSI near 65, while IWM sits under its 10, 20 and 50-day and trades within 1% of its sixty-session low at 279.05. Weekly MACD histograms are negative on all three. A range whose top is being tested by one leg while another leg tests the bottom is the shape to read from here.

What is actually going on

Into the week ahead, the thing worth carrying is not the hawkish turn itself. It is who paid for it. A Fed governor said on Wednesday that further increases are likely needed, per CNBC, and the same morning's flash PMI showed input costs rising at the fastest pace since October 2022. October hike odds jumped to 73% within hours, per CME FedWatch, and the two-year rose more than 13 basis points to 4.9% in a day. The ten-year printed its highest yield since June 2007 on Thursday, helped along by a weak five-year auction. On every textbook reading that is a bad week for stocks. The large-cap index rose, the Nasdaq closed at a record, and small caps fell for a third straight week.

That split is the story, and it has a mechanism. The growth behind the rate move is not evenly spread. S&P Global's flash composite printed 58.4, the fastest growth in more than five years, and core capital goods orders rose 14.1% on the year, per Wolf Street, the strongest pace since August 2021. That spending is on AI infrastructure. It shows up as revenue at chipmakers, equipment makers and the companies that build data centres. Those companies sit in the large-cap and Nasdaq indices, they carry little floating-rate debt, and their earnings are rising faster than their discount rate. The borrowers who feel the rate increase first are smaller companies with bank loans that reprice inside a quarter. CNBC reported the correlation between small caps and the ten-year note price touched a one-year high above 0.97 last week. Small caps are trading as a bond.

The inflation underneath has changed character too, and that is what makes the committee's job harder. For most of the summer the problem was fuel. September's flash PMI tied the jump in services input costs to fuel, freight and wages together. A fuel shock fades when the fuel price does. A wage-backed services trend does not, and it is the one a policy rate can reach. So the committee is now tightening against something it can influence, which makes a sequence of increases more credible, not less. That is why Wednesday's core PCE carries more weight than its headline.

The oil market split in two last week. Saudi Arabia restarted its East-West pipeline on September 22, per The National, and US crude fell almost 8% on the week. Brent barely moved and held its triple-digit price. The pipeline fixed a Saudi routing problem. Brent's premium is about the Strait of Hormuz, and that has become a negotiation. Iran offered to reopen the Strait within a week in exchange for lifting the US blockade and oil sanctions, per CNBC. The US president said on Saturday he had rejected it, per NBC News, while Iran International reported talks on a phased deal. The widest gap between the two crude benchmarks in weeks is the market pricing exactly that uncertainty.

The argument on desks is whether good news is still good news. One camp says strong growth is the whole case: earnings are compounding, Street estimates keep rising, and a committee raising rates into a boom is what 1997 looked like. The other camp says strong growth is now the risk, because each strong print extends the cycle and the multiple is what pays. Last week both camps got what they wanted in different assets. The equity index voted for the first camp, and the bond market and small caps voted for the second. That cannot both run for long. Either growth cools enough to let yields settle, or yields rise far enough that the large-cap multiple starts to feel it.

Positioning says the crowd is braced for the second camp. AAII bears reached a one-year high of 55.6% in the week to September 24, and the bull-bear spread has been under its average for nine straight weeks. Options traders bought about 480k IWM puts against 371k calls in one session, per CNBC. Yet the VIX futures proxy sits well under its 50-day, so the index-level options market is not paying for protection. Fear is concentrated in the leg already falling. That leaves a tape that can take bad news in small caps without flinching, and could be caught flat-footed by bad news in the leg carrying it.

For sectors, the dividing line is whether a company sells into the capital-spending cycle or borrows to fund itself. Semiconductors and hardware took the top five gainer slots on Thursday, per 24/7 Wall St., and memory is the bid into Micron. A new AI assistant from Meta hit brokerage and cybersecurity names on fears of disruption, per CNBC, which is a reminder that the AI trade now has losers inside large caps too. Refiners still have the diesel spread working for them. Banks get a steeper curve. Utilities, real estate and the floating-rate borrowers carry the rate bill every day the long end holds its level.

Where this sits against consensus

On rates this issue sits with the futures market, which prices October near 75% (CME FedWatch, September 25), and reads the week's prints as asymmetric around that. On the index we sit below most of the Street. Barclays' Venu Krishna (7,950, September 9), HSBC (8,100) and Ed Yardeni (8,400) have raised year-end numbers on earnings, and we agree with them on earnings. We part company on the multiple: this issue moves the twelve-month view to neutral because a ten-year at 2007 highs makes a flat multiple hard to hold. BofA's Savita Subramanian at 7,400 is the published view closest to ours. We would be late if the long end reverses under 4.80% with estimates intact.

Scenarios

Central — 50%

PCE and payrolls land near consensus, October stays about three-quarters priced, and SPY works inside 761.57 to 779.37 while the split between tech and small caps persists without resolving.

Trigger. Core PCE at 0.3% on Wednesday and payrolls between 50k and 150k on Friday.

Invalidated by. A Friday 2026-10-02 SPY close outside 761.57 to 779.37.

Upside — 25%

A soft core print pulls October odds lower, yields ease from 2007 highs, and the index breaks its sixty-session high with small caps finally joining.

Trigger. Core PCE under 0.3% or payrolls under 50k, with a Micron guide that clears consensus; or a phased Hormuz deal with Brent under 95.

Invalidated by. A breakout led only by semiconductors while IWM stays under its 10-day at 284.48.

Downside — 25%

Core PCE at 0.4% and strong payrolls move the market to pricing a sequence, the two-year crosses 5%, and the small-cap break spreads into the leg carrying the index.

Trigger. Core PCE of 0.4% or more with payrolls above 100k, or a Micron guide miss into a ten-year above 5.2%.

Invalidated by. The ten-year back under 5.0% by Friday with IWM holding 279.05.

The week ahead

  • 2026-09-30August PCE price index and personal income, ADP employment and Chicago PMI; Micron fiscal Q4 results
  • 2026-10-02September employment report: payrolls, the unemployment rate and average hourly earnings
  • 2026-10-28FOMC decision at the end of the October 27-28 meeting, no projections
  • 2026-10-30Grading date for the month horizon
  • 2026-12-09FOMC decision with the next Summary of Economic Projections
The next five sessions

This week

The week ahead hangs on two prints, and the market meets both with most of the hawkish outcome already paid for. CME FedWatch had an October increase at 75.8% on Friday. That changes how each print works. A number in line with consensus confirms a repricing that has already happened. The live risk is in the misses, and it is lopsided: a soft number can pull those odds down a long way, while a hot one can only add a few points to October and then starts pricing December.

August PCE comes first, on Wednesday. Published previews put headline at +0.4% on the month and core at +0.3%, or 3.2% on the year, per TradingKey and CMC Markets. Headline will carry August's fuel prices, and nobody on the committee will be surprised by it. Core is the number. A 0.3% core says the energy shock is still mostly inside the energy complex, even with September's flash PMI showing services input costs at their highest since November 2022. A 0.4% says the pass-through has started, and that is the reading that moves December as well as October.

Payrolls on Friday arrive after a week of labor data that gave no sign of cracking. Initial claims fell to 197k, one of the lowest readings since 1969, per the Labor Department via Bloomberg. Previews range from 85k to 100k against August's 162k, with unemployment at 4.1% and hourly earnings up 0.3%. A labor market this tight tells the committee that tightening has not yet bitten. The wage line matters more than the headline, because wages are the channel that turns a fuel shock into a services problem.

Micron reports Wednesday into both prints. Consensus sits near 31 a share in earnings against 3.03 a year ago, with the company's own guide at 31, plus or minus one, on revenue of about 50 billion. The options market priced a move near 10%. The number to read is the next-quarter revenue guide. The memory trade carried the Nasdaq to a record last week, and Micron's guide is the first test of whether that bid is still pricing future orders or has already priced them.

IfCore PCE prints 0.3% or less on Wednesday and Friday's hourly earnings come in at or below 0.3%Thenthe pass-through argument loses its best evidence of the week, October odds have room to fall, and the index gets a clean look at its sixty-session high.
IfCore PCE prints 0.4% or more and payrolls clear 100kThenthe market moves from pricing one more increase to pricing a sequence, and the two-year near 5% becomes the level every other asset keys off.
IfIWM closes a day under its sixty-session low while the ten-year holds above 5.1%Thenthe transmission that has run through small caps for three weeks has not stopped, and the index is defending a top its weakest leg has abandoned.
The next four weeks

This month

Over the month ahead the question shifts from October to December. With October about three-quarters priced, the October 27-28 meeting becomes a formality unless the next four weeks of data soften. What matters is whether the committee signals a sequence. The next set of projections is not until December, so the statement and the press conference carry the whole message.

The oil story changed shape last week. Saudi Arabia restarted the East-West pipeline on September 22, per The National, far ahead of the repair window reported a week earlier, though Aramco is still working flows back toward 4 million barrels a day. WTI fell 7.9% on the week. Brent barely moved. The gap between the two is the story: the pipeline fixed a Saudi routing problem, while Brent's premium is about the Strait, and that one is in a negotiation. The seven-day reopening offer and its rejection on Saturday, covered in the narrative above, leave a phased deal as the open question. The month's oil price is a bet on that process.

Earnings open mid-October with the banks, into a curve whose long end is at its highest since 2007. That is good for net interest margin. The question is whether credit costs show up alongside it, and where. The small-cap stress in the equity market is a reason to read the regional lenders' provisioning lines before the headline beats.

The honest tension over the month is that the same data are bullish and bearish at once. Core capital goods orders rose 14.1% on the year, the strongest since August 2021, per Wolf Street. That is an earnings tailwind for anyone selling into the build. It is also the demand the committee is now leaning against.

IfA phased Hormuz deal is announced and Brent trades under 95Thenthe inflation impulse loses its largest input without policy doing anything, and December stops being a live debate.
IfPCE and payrolls both run hot and the October statement adds language about further adjustmentsThenthe market prices a sequence, and the gap between the index and its rate-sensitive leg widens again.
IfQ3 regional bank results show provisioning rising faster than marginThenthe small-cap weakness is a credit story as well as a rate story, and it stops being containable inside one leg.
The next quarter

This quarter

Through the quarter ahead the argument is whether growth this strong is an earnings story or a policy story, and the Q3 season is where it gets numbers.

The growth side keeps winning its own argument. S&P Global's September flash composite printed 58.4, the fastest growth in over five years. Core capital goods orders rose 1.6% in August against forecasts near 0.6%, per Bloomberg, beating every economist in its survey. Strategists have been chasing that. Barclays raised its 2026 index EPS estimate to 365 from 337 on September 9, per CNBC. HSBC and Ed Yardeni lifted their year-end numbers too.

The policy side keeps winning its own argument as well. Every one of those data points is also a reason for the committee to keep going. The same flash PMI showed services input-cost inflation at its highest since November 2022, and S&P Global tied it to fuel, freight and wages together. A fuel shock alone fades. A wage-backed services price trend does not.

Small caps are where the quarter keeps its score. September took the Russell 2000 from up 20% on the year to up 14%, per CNBC, while the large-cap index held its gains. Goldman Sachs has about 32% of Russell 2000 debt on floating rates against 6% for the S&P 500, which is why every increase lands on small-cap interest costs within a quarter. The broadening trade needs the cycle to end. Nothing in the quarter's calendar before December can tell it that.

IfIWM reclaims its 50-day at 294.00 on two consecutive Friday closes while the large-cap index holds its range topThenthe market is looking through the cycle to its end, and the broadening trade has a working mechanism for the first time this year.
IfQ3 guidance from capital-goods and semiconductor suppliers steps down while rates stay near highsThenthe growth half of the argument weakens without the policy half easing, which is the combination that takes the multiple lower.
IfA Friday SPY close under the 200-day at 718.45Thenthe range has failed and the index-level question replaces the leadership question.
The next twelve months

This year

The twelve-month view moves this week, and the reason is arithmetic.

The case for owning the index over a year has rested on earnings doing the work at a roughly flat multiple. The earnings half is in better shape than it was a month ago. Street numbers keep rising: Barclays' Venu Krishna lifted the year-end index level to 7,950 on September 9 on higher earnings estimates, per CNBC, and HSBC and Ed Yardeni moved to 8,100 and 8,400. Micron's quarter is priced near ten times last year's earnings, and capital goods orders are growing at a 14% annual pace.

The multiple half has now degraded for four straight weeks. A month ago the ten-year sat near 4.8%. It printed its highest yield since June 2007 on Thursday. Over the same month the committee went from a debated increase to a delivered one, with a governor saying more are likely and futures pricing October at about three in four. A flat multiple under that path is an assumption we can no longer defend at conviction.

The outcome that fits both halves is a flat year on rising profits. Nobody publishes a target for it. BofA's Savita Subramanian, at 7,400, sits closest to it among the published numbers we found. That is the path this horizon now weights most.

What would move it back is the long end. Profits are not the constraint. If the ten-year returns under 4.80% while 2027 estimates hold, the constructive case is intact again.

IfThe ten-year returns under 4.80% while 2027 index EPS estimates holdThenthe multiple stops being the constraint and the earnings case carries the year.
IfThe committee delivers two or more further increases and the long end stays above 5%Thenthe year resolves as rising profits against a falling multiple, a flat index in a strong economy.
IfTwo consecutive months of downward revisions to 2027 index EPSThenthe one part of the case still working has turned, and the question becomes how far the multiple and the estimates fall together.

The levels

Into the week SPY sits at 771.35, between its 50-day at 761.57 and the sixty-session high at 779.37. QQQ sits at 744.50, under its 748.35 high. IWM is at 281.97, above a 279.05 sixty-session low and under every short average. The index tests its top as small caps test a floor.

  • SPYThe 50-day at 761.57 is the week's floor and the line a hot core print would test first. The sixty-session high at 779.37 is the cap, with call open interest clustered at 780 and 787 as of September 24. A close through it with small caps still falling would be a narrow breakout. 718.45 is the 200-day and the year horizon's graded line.
  • QQQQQQ is 0.5% under its 748.35 high after a 3.19% week, with daily stochastics above 90. Micron's report lands on the leg that carried it. The 10-day at 727.25 is the first test of whether a guide disappointment is a pause, and the 50-day at 712.65 separates a pullback from a change of character.
  • IWM279.05 is the sixty-session low and the week's most informative level. A close under it with yields still rising says the rate transmission is still running. The 200-day at 275.68 sits under it. The 10-day at 284.48 is the first reclaim, and the 50-day at 294.00 is the quarter's broadening test.

What would change it

Watch three things. Core PCE on Wednesday: 0.3% keeps the energy shock contained, 0.4% says it is spreading. Whether SPY's Friday close stays inside 761.57 to 779.37. And whether IWM holds 279.05 while the ten-year sits above 5.1%.

Leadership

  • Leading: Semiconductors and AI hardware, taking the top gainer slots on Thursday and carrying the Nasdaq to a record into Micron, Capital-goods suppliers to the AI build, on core orders up 14.1% year on year, Refiners, still on the diesel margin
  • Lagging: Small caps, whose price now tracks the ten-year note almost one for one, Long-duration bonds and the bond-proxy sectors, utilities and real estate, Gold, against real yields near two-decade highs, Brokerage and cybersecurity names, hit on AI disruption fears

The line runs between companies that sell into the capital-spending cycle and those that borrow to fund themselves. The first group sits mostly in the large-cap indices, which is why the index and the small-cap benchmark moved in opposite directions through a hawkish week.

Cross-asset

  • oil (oil; USO is a WTI futures proxy and not WTI itself). WTI fell 7.9% on the week to about 92.61 while Brent held near 104, per EnergyNow and The National. USO lost 3.57% on the week and is still up 10.9% on the month. The Brent premium over WTI is the price of the Strait talks.
  • long\_treasuries (20y+ Treasuries, TLT as the vehicle). TLT fell 2.38% on the week and sits 3.6% under its 50-day. The ten-year ended the week near 5.16% after a 2007 high on Thursday, per CNBC. The long end is where the week's hawkish repricing was paid for.
  • dollar (US dollar; UUP is a DXY proxy and not DXY itself). UUP rose 0.81% on the week and 1.78% on the month, sitting above both its 50 and 200-day. A firm dollar alongside a rising US yield is the rate differential doing the work.
  • gold (gold; GLD is the ETF proxy). GLD fell 1.93% on the week and 3.68% on the month, now under its 50-day. Gold struggling against real yields near two-decade highs is the opportunity cost of holding it.
  • volatility (VIX futures ETF, not spot VIX). VIXY sits 11.4% under its 50-day. Index-level protection is cheap while small-cap put buying is heavy, so the fear sits in one leg.

The record

Last Sunday's note called the week neutral on one test: SPY's Friday close inside 749.60 to 775.30. It closed at 771.35. Grade: right. But the hawkish triggers we mapped as the downside path both fired and the index rose anyway. Week horizon: 2 right, 0 wrong, 2 mixed over 4.

  • This week — right. We said: Neutral at 0.60 on one test: SPY's Friday 2026-09-25 close inside 749.60 to 775.30, with the flash PMI prices-paid line and the count of Fed speakers pointing at October named as the swing variables. What happened: SPY closed the week at 771.35, inside the band, up 1.27% on the week. Both downside triggers fired anyway: input costs hit their highest since October 2022 in the flash PMI, a Fed governor said further increases are likely, and October hike odds rose to 73%. The index rose through all of it, while IWM fell 0.75% and TLT 2.38%.
  • This month — too early to call. We said: Neutral at 0.60, wrong if any Friday close through 2026-10-23 printed SPY under 749.60 or over 779.37, with August PCE and the East-West pipeline as the dated swing variables. What happened: The first Friday in the window closed at 771.35, inside the band. The pipeline restarted on September 22 at a low rate, well inside the reported five-to-six-week repair window, and WTI fell 7.9% on the week. PCE has not printed.
  • This quarter — too early to call. We said: Neutral at 0.55, wrong to the upside on two consecutive IWM Friday closes above 295.14 with SPY over 775.30, wrong to the downside on a Friday SPY close under 716.28 or two months of falling 2027 estimates. Graded 2026-12-31. What happened: IWM fell 0.75% to 281.97, printing a new sixty-session low at 279.05 inside the week, and now sits 4.1% under its 50-day. SPY closed far above its 200-day. No trigger fired.
  • This year — too early to call. We said: Constructive at 0.40 on the 2027 index EPS line, with the multiple assumption degrading three weeks running and a pre-committed downgrade at 0.35. What happened: No estimate turn is visible; if anything the profit side strengthened, with core capital goods orders up 14.1% on the year. The multiple side degraded for a fourth week: the ten-year printed its highest yield since June 2007 on Thursday.

Week horizon: 2 right, 0 wrong, 2 mixed over 4 graded notes. Month, quarter and year: 0 graded, 12 too\_early across those same 4 notes. The archive holds 4 prior notes, dated 2026-08-30, 2026-09-06, 2026-09-13 and 2026-09-20.

Sources

Prices, ranges, moving averages and open interest in this note come from our own market-data bars through 2026-09-25. Every external claim names its source above. This note is forward-looking market research: it describes scenarios and the conditions that would confirm or invalidate them. It is not investment advice, not a recommendation, and states no forecast, price target or trade.

These notes are forward-looking market commentary and describe scenarios only. They are not forecasts, price targets, base cases, or trade recommendations, and they are not investment advice or tailored to anyone’s circumstances. Prices come from our own market-data bars; every external claim names its source in the note.