The week ahead — September 13, 2026: The hike is priced; the 2027 dots are not

Wed Sept 16: FOMC decision 2:00pm ET with a new dot plot, presser 2:30. Futures put the quarter point near 85-90% (CME FedWatch). Lennar reports that afternoon. Jobless claims Thu, consensus 315k after 310k. August retail sales consensus -0.8%.

Last week, graded

Scorecard. Last Sunday's note called the week neutral and said a close under 759.48 would mean not defensive enough. Sept 10 closed 757.83. Friday repaired it at 764.29, inside the box. Grade: mixed, two straight. The fix is one falsifier per horizon, not two.

• This week — mixed. We said: Neutral into a stacked Thursday and Friday, with the note saying a daily close under 759.48 would mean neutral was not defensive enough and a finish inside the box would mean it was right. What happened: SPY closed at 757.83 on September 10, under the floor, then repaired it with a 764.29 close on September 11. The week itself gave up 0.77% and finished inside the box.
• This month — too early to call. We said: Neutral at 0.5, with the stance flipping defensive on a weekly SPY close below 756.86 and constructive on a weekly close above 779.37 paired with a positive QQQ weekly MACD histogram, graded by September 27. What happened: The weekly close printed 764.29, inside the band, and QQQ's weekly MACD histogram is still negative at -3.24. Neither trigger has fired and the September 27 date has not arrived.
• This quarter — too early to call. We said: Neutral at 0.5 on the grounds that earnings and the discount rate now point opposite ways, wrong to the upside if IWM reclaimed and held above 305.18 with SPY over 779.37, wrong to the downside on falling 2027 estimates or a loss of the 200-day. What happened: IWM went the other way, losing 2.41% on the week to 288.89 and sitting 2.49% under its 50-day. No trigger fired. The evidence moved against the broadening leg of the argument without reaching the level that would grade it.
• This year — too early to call. We said: Constructive at 0.5, tracked on one variable only: 2027 index EPS. Invalidated by two consecutive months of downward revisions or a monthly close under the 200-day. What happened: No estimate turn is visible and the index stayed far above its 200-day. The multiple side degraded again, with the ten-year at 4.96% after 4.81% the prior week and the thirty-year above 5.2%.

Week horizon: 0 right, 0 wrong, 2 mixed over 2 graded notes. Month, quarter and year: 0 graded, 6 too\_early across those same 2 notes. The archive holds 2 prior notes, dated 2026-08-30 and 2026-09-06.

The setup

Transition. Into Wednesday's decision the index and its internals point different ways. SPY and QQQ hold above their 50-day lines while IWM sits 2.49% under its own. Eight of eleven sectors are down over the five sessions to September 11, and the weekly MACD histogram is negative on all three indexes. Leadership has narrowed to semiconductors and energy. That is what a transition looks like from the inside.

What is actually going on

This week the Federal Reserve is expected to raise rates for the first time in three years, and the reason is an import price. Gasoline did the work in the August inflation report, contributing more than a third of the monthly increase, while the core rate eased to its slowest annual pace since 2021. That combination changes the reflex the equity market has run on since 2024. For two years weak data meant help was coming, because weak data bought a cut. It no longer does. Strong data now buys a hike and weak data buys nothing, so both tails point the same way for the cost of money.

Whether the committee should move at all is the live argument on desks. Mark Zandi of Moody's Analytics has called responding to a supply shock a violation of "monetary policy 101" and said he does not think they should raise rates. Higher rates do not produce barrels. They cannot reopen a shipping lane or unwind a war premium. The answer from the other side is that the committee is not fighting gasoline at all. It is defending the expectations channel, and it has one instrument with which to do that. Three regional bank presidents dissented in July in favour of a hike, the largest such bloc since 2016, and they made that case before the energy move accelerated.

One popular argument against a hike has quietly stopped being true. Much of the commentary written in late August leaned on a July payroll print showing a loss of 23k jobs, the kind of number that normally keeps a central bank still. The September 4 revisions turned that loss into a gain of 21k. August itself added 162k against a consensus near 53k, and the unemployment rate held at 4.1%. Anyone still citing the summer labour collapse is quoting a figure the Bureau of Labor Statistics has already replaced. Check the vintage on every soft-labour claim you read this week.

The decision itself is close to fully priced. The projections are not. June's median for the end of 2027 sat inside the current range with individual forecasts spread from about 3.0% to 4.4%, a dispersion wide enough that the median can move without many people changing their minds. A quarter point moves the front end of the curve. The 2027 median moves every discounted cash flow in the index. That asymmetry is why Wednesday afternoon is bigger than the number attached to it, and it is the part of the week no futures contract has settled.

Capacity to absorb a surprise looks thin. BofA's Global Fund Manager Survey had cash at 3.6% of assets in July, under the 4.0% line the bank's own rule treats as a sell signal. Benzinga reported the volatility index near 15.75 in Friday midday trade, which sits at the calm end of its range going into the first tightening in three years. The composition is as interesting as the level. Rocky Fishman of Asym Research puts semiconductors at 60% of the S&P 500's average daily move, and Nomura's highest-Sharpe basket has re-formed as semiconductors and energy together. Two sectors are carrying an index of eleven, and low cash beside low hedging is not the balance sheet you want when that is true.

The common explanation for small-cap weakness does not survive contact with the data. The story is that small companies borrow at floating rates and break when policy tightens. Franklin Templeton, citing Furey Research Partners, calls that intuitive and unsupported by history. About a third of Russell 2000 companies hold more cash than debt, and nearly half the index's debt sits in firms making up only 12% of its market capitalisation. The real constraint is pricing power. A cost shock passes through a mega-cap income statement and stalls in a small one, and that is an earnings story wearing a balance-sheet costume.

Broadening is the thing this tape keeps trying and failing to do. Fundstrat notes the equal-weight S&P 500 has broken its uptrend while consumer discretionary, health care and financials weaken underneath a semiconductor-led headline index. Eight of eleven sectors were down over the five sessions to September 11 while the index itself gave up less than a percent. The index is not the market at the moment. A reader who owns the average stock has had a different September from a reader who owns the average index, and Brean Capital's 0.85 oil-to-yield correlation is the machinery connecting the two. While it holds, every headline out of the Gulf is a rates headline, and every rates headline is a breadth headline.

The week ahead

• 2026-09-14Gulf Cooperation Council and Iranian diplomats meet in Oman on shipping through the strait; no consensus number attached
• 2026-09-15Trip.com reports after the close, consensus 0.91 a share
• 2026-09-16FOMC decision 2:00pm ET with the Summary of Economic Projections; press conference 2:30pm ET
• 2026-09-16Lennar fiscal Q3 after the close, consensus 1.30 a share against 2.00 a year earlier
• 2026-09-17Weekly jobless claims, consensus near 315k after 310k
• 2026-09-23General Mills fiscal Q1 before the open, consensus 0.71 a share
• 2026-09-25FedEx fiscal Q1 after the close, tentative on the broker calendar, consensus 3.91 a share after five straight beats

The next five sessions

This week

Wednesday afternoon carries two releases, and the market has settled only one of them. Futures put the quarter point at roughly 85 to 90% depending on the snapshot, and the odds travelled there in a single session after the August inflation report. What is not priced is the Summary of Economic Projections that lands with the statement. June's version left the median policy rate for the end of 2027 inside the current range, with individual forecasts scattered from about 3.0% to 4.4%. If the committee raises the starting point and leaves that median alone, it is describing a one-off adjustment and a path that still comes back down. If the median rises with the rate, it has told the market the cuts still sitting in the curve are not coming, and that repricing runs through every long-duration asset in the index.

The other thing on the calendar that afternoon is a homebuilder. Lennar reports after the close on September 16, and consensus is 1.30 a share against 2.00 in the same quarter a year earlier. The reported numbers underneath explain the shape of that. Lennar's net margin ran 6.71% in the quarter ended August 2025 and 3.84% in the quarter ended May 2026. D.R. Horton went from 11.11% to 9.81% across the same June comparison, and PulteGroup from 13.82% to 11.85%. Housing has been in an earnings recession for three quarters while the index made highs. The print lands the same afternoon the committee raises the cost of a mortgage, and new orders and incentive spend are the numbers that matter in it.

Two smaller prints fill out the week. Neither claims nor retail sales will move a meeting this thoroughly priced, but both feed the projections the committee has to defend afterwards.

The tape arrives at all this having already broken and repaired its own range. SPY traded and closed under the floor of its August box on September 10 and was back inside by the September 11 close. A range that breaks and mends inside five sessions is a range waiting on a scheduled resolution. This week the resolution has a date on it.

The next four weeks

This month

The month ahead is an energy question wearing a monetary costume. Crude is what put the hike on the table, and crude is what could take it back off. West Texas Intermediate traded above 100 a barrel midweek and settled near 99.36 on Friday, up 8.6% across five sessions and more than 19% over the month. Brent ended near 104.41 on the same pattern. AAA put the national diesel average at a record 6.05 a gallon on September 11, and diesel is the number that reaches every freight invoice in the country about six weeks later.

The supply picture behind it is specific and checkable. The Strait of Hormuz, which once carried about a fifth of the world's seaborne oil, remains severely disrupted: seven vessels crossed on Wednesday. Iran-backed Houthi forces have taken the Red Sea port of Mocha and moved toward the Hanish Islands, extending their reach over Bab el-Mandeb. Saudi Arabia shut its East-West crude pipeline as a precaution after attacks. Gulf Cooperation Council diplomats meet their Iranian counterpart in Oman this week to discuss a temporary arrangement for shipping through the strait. That meeting is the highest-leverage event on the month's calendar and the only one with no consensus number attached to it.

Both tails are already published. Goldman Sachs has warned Brent could run above 120 in 2027 if Gulf output stays four million barrels a day under prewar levels. The International Energy Agency has cut its demand outlook and models a contraction of 2.5 million barrels a day in 2026. A committee that tightens on Wednesday is acting on a price whose direction two credible forecasters cannot agree on.

The transmission into equities does not run through energy shares. Brean Capital measures the daily correlation between oil and Treasury yields since August 31 at 0.85, the tightest stretch since the conflict began. While that holds, a tanker headline is a duration headline, and duration is where the index's leadership sits. That is how a shipping lane ends up setting the multiple on a semiconductor company.

The next quarter

This quarter

The quarter ahead sets a durable earnings line against a market that keeps failing to widen, and both of those got more extreme in the five sessions to September 11. On the earnings side, Oracle's fiscal first quarter showed revenue of 19.35 billion against a 19.14 billion consensus, cloud infrastructure revenue up 121% to 7.4 billion, and remaining performance obligations higher by 209 billion over the year. Management guided full-year capital spending to 90 to 95 billion. The read-through moved the suppliers harder than it moved Oracle, with Dell up 11.3% on the day and Hewlett Packard Enterprise 9.1%. Contracted AI capital spending is the most visible earnings stream in the index and it did not weaken this quarter.

The demand side of the same tape said something else. Adobe guided fourth-quarter revenue below consensus at the midpoint. Kroger cut its full-year identical-sales guidance excluding fuel on the morning the inflation report landed. Those are two ordinary companies reporting that the consumer and the software budget are tighter than the index implies, and they said it in the same week the energy bill went up again.

Underneath the index, the dispersion is severe. Over those five sessions energy shares added 1.69% and technology 0.21%, while health care gave up 3.55% and materials 2.84%. Eight of eleven sectors finished lower while the index itself gave up 0.77%. That is not a market absorbing a shock evenly. It is a market whose average stock is having a worse quarter than its benchmark, and the benchmark's behaviour from here depends on two sectors more than it usually would.

Semiconductors are the sector that decides. The group sits roughly 11.5% above its July lows and more than 21% under its June high, which leaves a lot of unresolved positioning in the part of the index that carries it. If the fourth quarter's question is whether the market broadens, the answer shows up first in whether small caps can reclaim 296.28 while the index defends 779.37. Neither has happened. The quarter has three months in which to say so.

The next twelve months

This year

The twelve-month case ahead rests on one variable and one assumption, and only one of the two is still intact. The variable is forward earnings, and it has not turned. Contracted AI capital spending is visible years out, and nothing in this reporting season cut it. The assumption is a roughly flat multiple, and that is the half under pressure. The ten-year Treasury ended September 11 at 4.96% and the two-year at 4.63%, with the thirty-year above 5.2%. A year ago the argument for paying twenty-two times forward earnings was that the policy rate was coming down. This week the policy rate is going the other way, and the discount rate has moved against the multiple two weeks in a row.

Published year-end index forecasts show how wide the disagreement runs. Across 19 firms the median sits near 7,850, with Bank of America at 7,100 citing what it calls too many red flags and RBC at 8,150. The gap between low and high is about 15% of the index, and almost none of it is a disagreement about profits. Consensus has index earnings per share near 306 for the year, and both the most bearish and the most bullish house can live with that number. What separates them is the multiple a market pays when cash yields near 4% and the central bank has resumed tightening.

That is why the falsifier for a twelve-month view is an estimate revision and not a price level. Price can fall 15% on a multiple re-rating and leave the earnings case untouched. Estimates falling two months in a row is a different event, because it removes the only leg the case stands on. The specific line to watch is 2027 index earnings: it is the number that has to keep rising for a flat multiple to be enough, and it is the number that would show a capital spending cycle slowing long before any price chart did.

One more asymmetry belongs in a year view. An energy-driven inflation shock is temporary at some horizon by construction, because a price shock either passes through and fades or destroys the demand that sustains it. A policy mistake made in response to one is not temporary. The twelve-month risk that matters here is not the oil price. It is a tightening cycle that outlives the reason for it, and the September projections are where the first evidence either way shows up.

The levels

Into Wednesday, SPY's box runs 756.64 to 779.37 with the 50-day at 758.62 just inside the floor. QQQ has 702.70 to 734.58, 50-day 710.41. IWM is the odd one, 288.89 against a 296.28 50-day it has already lost. Small caps break first here.

• SPYThe August floor broke on a closing basis September 10 and was repaired the next session, and the 50-day at 758.62 sits just above it, so the two levels now defend each other. 712.05 is the 200-day and the line the twelve-month case gets graded against.
• QQQ702.70 to 734.58 is the sixty-session envelope. The 50-day at 710.41 is the line separating a pullback from a change of character in the part of the index doing the carrying.
• IWM288.89 against a 50-day it has already lost, with 287.18 the low that has held twice. Small caps break first in this tape, so this is the early-warning line for the broadening question.

What would change it

Watch three things this week: whether the 2027 dot median moves with the policy rate, whether IWM can reclaim its 50-day at 296.28, and whether the Oman talks drain the war premium out of crude. A Friday close outside the August box settles the range question either way.

Leadership

• Leading: Energy, Semiconductors, Communication services
• Lagging: Health care, Materials, Consumer discretionary, Small caps

Leadership is two sectors deep. Energy leads on the supply shock and semiconductors on contracted capital spending, and those are the only engines the index has this month. The laggards are the cyclical middle, which is where a broadening market would have to show up first.

Cross-asset

• USO (WTI proxy, NOT spot crude). Up 15.86% in September and 41.95% over its 200-day. This is the input that put the meeting in play.
• GLD (gold proxy, NOT spot gold). Down 1.97% on the week and 4.13% under its 200-day. Gold is not behaving like a hedge against this particular shock.
• UUP (DXY proxy, NOT the dollar index). Flat on the week and 1.58% over its 200-day. The dollar has not yet priced the tightening the rates market has.
• TLT (20y+ Treasuries). Down 1.63% on the week and the same on the month. The long end is where the projections land first.
• IEF (7-10y Treasuries). Down 1.34% on the week and 2.59% under its 200-day. The belly of the curve is carrying the repricing.
• VIXY (VIX FUTURES ETF, NOT spot VIX; we carry no spot vol). Up 1.76% on the week and 32.53% under its 200-day. Forward vol is cheap going into a dated, two-sided event.

Sources

• CNBC, August 2026 CPI report
• CNBC, August 2026 jobs report
• CNBC, July 2026 FOMC rate decision
• Benzinga, stock market today September 11 2026
• Benzinga, As the Rotation Dies (Fundstrat, Brean Capital, Nomura, Asym Research, Franklin Templeton, Furey Research)
• CNBC, oil prices and the Strait of Hormuz
• Trading Economics, Brent crude
• Spokesman-Review, what a Fed rate increase means (Mark Zandi, Moody's Analytics)
• Yahoo Finance, where will the S&P 500 end 2026 (BofA, Citi, RBC)
• Gate, BofA Global Fund Manager Survey cash levels
• Kiplinger, economic calendar September 14-18 2026
• ETF Trends, Treasury yields snapshot September 11 2026
• TradingKey, September FOMC preview and dot plot
• Robinhood broker data: earnings dates, EPS estimates and reported quarterly financials for LEN, DHI, PHM, FDX, GIS

Prices, ranges, moving averages and open interest in this note come from our own market-data bars through 2026-09-11. 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.