Market Snapshot October 29th 2026

The Fed Held 10–2 With Two Votes to Hike, the Bank of Canada Hiked on Energy While New Zealand Surprised With a Hold, and Alphabet Fell 9% After Hours on Negative Free Cash Flow

Data

Theme: “Three Central Banks, Two Answers to Oil” — On Wednesday the Bank of Canada raised rates to 2.50% and said it did so to stop high energy prices from spreading into other prices. The Reserve Bank of New Zealand, facing the same oil shock, held at 2.75% in a 4–3 vote, against market expectations of a hike. The Fed held at 3.75–4.00%, but Governor Lisa Cook and Cleveland Fed President Beth Hammack dissented in favour of a hike. That is the first time since October 1992 that a governor and a regional president dissented together on the same side. Gold was roughly flat through all three decisions. After the US close, Alphabet fell about 9% on a profit miss and negative free cash flow of $9.1 billion, while Microsoft rose 2.5% on a beat with capital expenditure below forecast.

Wednesday, 28 October 2026 was a full trading session. Parts I, II, III and V cover it. Part IV looks to Thursday, 29 October: Q3 GDP, September PCE, the ECB and Amazon.

📌 Corrections and Clarifications

  1. The NZD side of the FX idea in recent editions failed. On Friday and Monday this publication described NZD and CAD as supported by priced hikes against a USD whose December pricing had fallen. On Tuesday it said the USD side of that idea had gone and that NZD and CAD would now need hawkish surprises relative to their own pricing. The RBNZ delivered the opposite: a surprise hold at 2.75%, split 4–3, with markets having priced a hike. NZD/USD fell below 0.5600 (FXStreet headline), from 0.5682 last Thursday. Readers who acted on the NZD part of that view would have lost on it. This publication withdraws it.
  2. The gold test designed for Wednesday could not run cleanly, and its fallback condition has been met. Tuesday’s edition said that if the RBNZ and Bank of Canada hiked as priced and gold did not fall, the claim that oil reaches gold through central banks would be dropped. One hiked and one held, so the test was mixed. But gold barely moved through any of the three decisions, including a hawkish Fed hold. This publication now drops the central-bank channel as it said it would, and keeps only the observed inverse link with oil, cause unexplained.
  3. The dissent preview was partly wrong. Tuesday’s edition passed on Newsquawk’s list of possible dissenters: Hammack, Logan and Kashkari. Hammack dissented. Logan and Kashkari did not. Governor Cook, who was not on the list, did. Analysts quoted by Fortune said Cook’s dissent was the bigger surprise.
  4. The 22 October break in the 10-year has fully reversed. On 22 October the 10-year broke below 5.05%, the floor of a seven-week range, and this publication said the break was flow-driven and the most likely move to retrace. On Wednesday the 10-year closed at 5.05% (Yahoo Finance), back at the old floor. The call held. It is recorded here so the range is read as intact, not broken.

🏛️ 1. The Fed — A Hawkish Hold

Item Detail
Decision Hold at 3.75–4.00%
Vote 10–2
Dissents Governor Lisa Cook and Cleveland Fed President Beth Hammack, both for +25bp
Historical note First time since October 1992 that a governor and a regional president dissented together on the same side (Fortune)
Statement on activity Upgraded to “expanding at a strong pace”
Statement on inflation “Remains elevated”
Projections None at this meeting
December hike odds ~78% before (CME, via Fortune); ~83–85% after (Kitco, Yahoo)

 

Two dissents in favour of a hike, and an upgraded description of growth, make this a hawkish hold. The Fed raised rates in September for the first time in three years, and two voters already want the next increase.

Chair Kevin Warsh’s message at the press conference was to wait and assess. He said “the overwhelming consensus is that what we’ve done in this monetary medicine is starting to work some,” but that there are no firm conclusions (Fortune). He also said: “Getting policy right involves assessing near-term developments at timely, regular, and information-rich intervals” (Yahoo). And he announced that one of the Fed’s eight 2027 meetings will be devoted to longer-term strategy, with no rate vote (StockTitan).

No source found reports Warsh commenting directly on December or on energy. That omission matters: Tuesday’s edition said the statement’s wording on energy would be the most important sentence for oil. On the available reporting, the Fed did not single out energy, which leaves the oil shock inside the general phrase “inflation remains elevated.”

Market reaction was muted, as Tuesday’s edition expected if Warsh confirmed what was priced. December odds ended near 83–85%, roughly where they were before. The dollar index softened slightly after the decision (Kitco). Equities closed modestly higher.

Laura Ullrich of Indeed Hiring Lab, a former Richmond Fed staffer, told Fortune: “I think the probability of a hike in December is very, very high.” Her colleague Aubrey Woessner said holding was hard to reconcile with the Fed’s pledge to restore price stability.

🇨🇦🇳🇿 2. The Bank of Canada and the RBNZ — Same Shock, Opposite Answers

Item Bank of Canada RBNZ
Decision 🟥 +25bp to 2.50% 🟩 Hold at 2.75% (surprise)
Expected ~80% priced for a hike 75–80% priced for a hike
Vote Not disclosed 4–3
On energy Hiked “to reduce the risk that high energy prices spread to other goods and services” Flagged near-term uncertainty from volatile oil and fuel prices
On inflation CPI ~3%, expected ~3.5% by year-end; ex-gasoline 2.2%; core near 2% Wage growth subdued; medium-term expectations stable; administered prices lower than expected
Forward guidance “May need to raise the policy rate further”; will “proceed carefully” “Further tightening would likely be needed”
Currency No level found 🟥 NZD/USD below 0.5600

 

These two decisions are the most useful pair this publication has seen for understanding how central banks treat an oil shock.

The Bank of Canada did what this publication’s oil–gold hypothesis assumed central banks would do. It saw inflation excluding gasoline at only 2.2% and core near 2%, and hiked anyway, because it judged that energy prices in the eighth month of the Middle East conflict, “with no clear end in sight,” would spread. It also said near-term inflation expectations have risen and that cost pressures will be passed on over the next year.

The RBNZ looked at similar facts and decided to wait. Subdued wages, stable medium-term expectations and lower administered prices outweighed the oil shock for four of seven members. It also noted that higher global yields, which it linked to inflation expectations, war-related destruction and AI investment demand, could speed up a global downturn.

The lesson is that central banks are split on whether to tighten against oil, and the split is close: 4–3 in New Zealand, 10–2 at the Fed. That makes the policy response to oil less predictable than the hypothesis assumed, which is one more reason to drop the central-bank channel from it.

One headline conflict: Bloomberg’s headline reads “New Zealand Raises Rate for Third Straight Meeting on Fuel Price,” while its own URL and every other source (TradingView, Nikkei Asia, FXStreet, Babypips) report a hold. This publication treats the hold as correct.

🥇 3. Gold — Flat Through Three Decisions

Instrument Wednesday Change Note
Spot gold (Kitco, late US) ~$4,056.80 −$2.50 (−0.06%) ⬜ Flat
Gold (Yahoo ticker) $4,066.20 +0.36% 🟥 Inconsistent with Tuesday’s settlement; see note
Spot silver (Kitco) ~$53.72 −$0.58 (−1.07%) 🟥 Toward its yearly low
Gold/silver ratio ~75.5 from ~74.7 🟥 Above 75 for the first time in this run
Gold/Brent ratio ~35.3 from ~35.8 Oil up, gold flat

 

The gold quotes do not reconcile. Yahoo’s ticker shows $4,066.20, up $14.60, which implies a prior close of $4,051.60. Tuesday’s December futures settlement was $4,074.60. The ticker is most likely a different contract or reference point. Kitco’s late spot quote, essentially unchanged, is the better guide, and no December futures settlement was found.

What gold did around the three decisions:

Gold ignored a hawkish Fed, a hike explicitly aimed at energy, and a dovish surprise. That is the clearest evidence so far that gold is not trading on central-bank expectations at present. As set out in the corrections above, this publication drops that part of its hypothesis.

What remains is the inverse link with oil. On Wednesday oil rose roughly 1–2% and gold slipped 0.06%. The direction fits for a third session, but a 0.06% move is too small to count as evidence either way.

Session Oil Gold Counted?
Monday 26 Oct Brent −0.78% +0.76% 🟩 Yes
Tuesday 27 Oct Brent −1.67% +0.90% 🟩 Yes
Wednesday 28 Oct Brent ~+1% −0.06% (spot) ⬜ Direction fits; too small to count

 

Silver is still the warning. It fell 1.07% to about $53.72 and the gold/silver ratio rose above 75 (USAGold headline: “Gold-Silver Ratio Tops 75 As Silver Slides Toward Its Yearly Low”). Kitco’s silver support is $53.24, then $52.41. Gold support is $4,039, then $4,015; resistance $4,059, then $4,079.

🛢️ 4. Oil — Higher on a Warning to Gulf States

Contract Wednesday Change Source / time
Brent $114.87 +0.87% The National, 2pm UAE
Brent Above $115 — FXEmpire, ~19:00 UTC
WTI $101.49 +1.59% The National, 2pm UAE
WTI Above $101 — FXEmpire, ~19:00 UTC
WTI (Yahoo ticker) $102.06 −0.70% 🟥 Inconsistent with Tuesday’s $99.90 settlement

 

No settlement prices were found for Wednesday. The intraday quotes agree that both benchmarks rose, with Brent around $115 and WTI back above $101, roughly 1–2% higher. Yahoo’s WTI ticker implies a prior close of $102.78, which does not match Tuesday’s $99.90 settlement and is most likely a different contract. Readers should take Wednesday’s close from their own terminal.

What Moved It

UBS raised its forecasts and now expects Brent near $110 through year-end, citing tight supply, falling inventories and record freight costs. Strategist Giovanni Staunovo wrote: “Middle Eastern flows, while improving, have not yet fully normalised,” and noted that global inventories have fallen sharply now that pledged strategic-reserve releases are largely complete.

That last point adds something new. Until now this publication has described Brent as a threat premium sitting on recovered flows. UBS is saying the buffer has also gone: strategic reserves have been released and inventories drawn down, so less stands between a supply disruption and the price than there was earlier in the conflict. That supports Tuesday’s steeper Brent prompt spread.

The EIA Report

Series Change Expected Note
Crude stocks −1.9m bbl ~flat 🟥 Fourth weekly draw
Gasoline stocks +3.9m bbl −0.9m bbl 🟩 Large build
Strategic Petroleum Reserve 282.6m bbl from 282.7m —
US crude production 13.922m bpd from 13.269m 🟩 +653k; hurricane shut-ins reversed

 

The crude draw supports the tight prompt market. The gasoline build supports Tuesday’s fall in product futures, which this publication recorded as a correction to its view that products would stay tighter than crude. US output recovered most of the 410,000 barrels per day lost to Hurricane Isaias, and is now nearly 14 million barrels per day.

📈 5. The Session

Index Close Change % Note
Russell 2000 2,830.51 +11.64 +0.41% 🟩
S&P 500 7,743.98 +21.38 +0.28% 🟩 Third gain
Nasdaq Composite 26,549.85 +72.04 +0.27% 🟩
Dow Jones 52,475.10 +97.53 +0.19% 🟩
VIX 16.83 — −3.9% Lowest since 21 Oct
Philadelphia Semiconductor 11,674 — +1.75% 🟩 22 of 30 up

 

Closes from Yahoo Finance and TradingKey. VIX and breadth from SCS. Breadth was positive: 55.3% of stocks advanced and 41.5% declined. New 52-week lows (251) still outnumbered new highs by nearly two to one, and only 36.9% of stocks are above their 50-day average.

Sectors (SCS): technology +0.8%, financials and energy +0.4%, industrials and healthcare +0.3%; real estate and utilities −0.8% each. Six of eleven sectors rose.

🏛️ 6. Rates

The 10-year rose about 5 basis points to 5.05% (Yahoo Finance), back to the floor of its old range. No reliable 2-year or 30-year closing levels were found for Wednesday. The December hike probability ended at about 83–85%.

With the 10-year back at 5.05%, the range of 5.05–5.35% that held from early September to 21 October should be treated as intact. The 22 October break below it was a one-week event, as this publication said at the time it was likely to be.

💱 7. FX

For an FX book, Wednesday is the clearest lesson so far that an 80% probability is not a certainty. Seventy-five to eighty percent pricing for an RBNZ hike turned into a 4–3 hold. The Bank of Canada delivered as priced and signalled it may go further; with oil higher on the day, CAD had both policy and energy behind it, but no level was found to confirm how it traded.

Companies

Theme: “The Market Is Grading Free Cash Flow” — Oracle rose 10.7% after forecasting $30 billion of free cash flow and lower capital spending from fiscal 2029. Microsoft rose 2.5% after hours on a beat, with capex about $6 billion below forecast. Alphabet fell about 9% after hours after missing on profit and reporting negative free cash flow of $9.1 billion, even though its capex also came in below forecast and Google Cloud revenue doubled. SK Hynix rose 13% on memory prices. Taken together, the three hyperscaler results sharpen the season’s rule: what the market is pricing is whether spending is covered by cash.

🔻 1. Alphabet — A Profit Miss and a Second Quarter of Negative Free Cash Flow

Measure (Q3 2026) Figure vs expectations / prior
Revenue $127.58bn +25% y/y; ~$0.6bn below $128.2bn consensus
Diluted EPS $2.12 🟥 vs ~$3.02 expected; $2.87 a year ago
Net income to common $26.17bn 🟥 vs $34.98bn a year ago
Equity securities −$8.66bn net loss vs +$10.73bn gain a year ago; −$0.55 EPS
Capex $51.36bn +114% y/y; ~7% below $55.2bn FactSet
Operating cash flow $42.29bn Did not cover capex
Free cash flow −$9.07bn 🟥 Second negative quarter; analysts expected −$2.27bn; Q2 −$5.86bn
Google Cloud revenue $30.71bn 🟩 +103% y/y
Google Cloud operating income $11.12bn 🟩 from $3.59bn
Search & other $63.35bn +12% y/y
After hours ~$291.50–298.68 🟥 −8% to −9.4%

 

Part of the miss is non-operating. An $8.66 billion net loss on equity securities, against a $10.73 billion gain a year earlier, cut EPS by about $0.55, according to TradingKey. But adding that back gives EPS of roughly $2.67, still well below the ~$3.02 expected. The operating business missed too. The sources consulted do not say which holdings produced the securities loss, and this publication does not speculate.

Google Cloud is the strongest line in any hyperscaler report this season: revenue doubled and operating income tripled. Alphabet tied it to enterprise AI demand and sales of its TPU systems. It did not offset the cash picture.

💻 2. Microsoft — A Beat With Lower Capex

Measure Figure vs expectations
EPS $4.89 🟩 vs $4.73
Revenue $92.28bn 🟩 vs $90.63bn
Microsoft Cloud revenue $63.9bn 🟩 vs $61.62bn
Azure growth 48% 🟩 vs 45% guide (investingLive headline)
Capex ~$6bn below forecast 🟩 (investingLive headline)
After hours $544.41 🟩 +2.52%

 

Figures from Yahoo Finance; Azure and capex from an investingLive headline whose article could not be retrieved. Forecasts for the quarter’s capex ranged from about $45.9 billion (VistaP) to about $50 billion (FinanceFeeds), so the size of the shortfall depends on which expectation is used.

Microsoft beat on every line reported and spent less than expected. Yahoo’s headline described the shares as jumping 4%, but its own after-hours figure was 2.52%. This publication uses the 2.52%.

🟩 3. Oracle — Rewarded for Promising to Spend Less

Oracle rose 10.6–10.7% to about $144.5 after setting out a cash-flow plan (TradingKey):

This is the cleanest single confirmation of the returns-versus-spending rule this season. Oracle committed to spending less, borrowing less and generating cash, and the market paid more than 10% for it. It also answers part of the credit question this publication has followed since 19 October, when CNBC described Oracle’s bonds as “dangerously close to junk.” A $20 billion cap on new financing is a direct response to that concern. The shares slipped 2.66% after hours.

⚖️ 4. The Rule, Refined Again: Free Cash Flow

Until now this publication has stated the rule as: a firmer commitment to future returns is paid for; a firmer commitment to future spending is not. Alphabet does not fit that neatly. Its capex came in 7% below forecast, which is the opposite of a raised spending commitment, and it still fell about 9%.

What separates the winners from the losers across the hyperscalers is free cash flow:

Company Cash picture Capex vs forecast Reaction
Oracle $30bn FCF projected for FY2029; financing capped Lower from FY2029 🟩 +10.7%
Microsoft Beat; capex below forecast ~$6bn below 🟩 +2.5% AH
Alphabet FCF −$9.07bn; operating cash flow did not cover capex ~7% below 🟥 ~−9% AH
Meta (21 Oct) FCF −$2.1bn; capex floor raised; $95bn equity raise followed Raised 🟥 −6.98%
CoreWeave Debt-funded; CDS ~1,000bp — 🟥 Repeated falls

 

The refined rule: the market is paying for spending that is covered by cash, and punishing spending that is not, whether it is funded by debt (CoreWeave), equity (Meta) or a drawdown in cash (Alphabet). Lower capex alone did not save Alphabet, because its operating cash flow of $42.3 billion still fell $9.1 billion short of what it spent.

This is one session and three companies, and Alphabet’s profit miss means the FCF reading is not the only explanation for its fall. But it is a more precise version of the rule, and it gives readers one number to check first in each remaining AI-related report: operating cash flow minus capex. Amazon reports after Thursday’s close.

💾 5. SK Hynix and Memory

SK Hynix ADRs rose 13.3% to $164.70 after third-quarter revenue of ₩104.62 trillion and operating profit of ₩83.33 trillion, up 32% and 38% from the second quarter. The operating margin was 79.7%. JPMorgan said both beat consensus, by 7% and 9% (TradingKey).

The driver was price. Average DRAM prices rose almost 30% from the previous quarter and NAND about 25%. A DRAM-focused ETF rose 6.97% (SCS), and the Philadelphia Semiconductor Index rose 1.75%.

This is strong support for the “own the constraint” framework this publication set out on 26 August: memory is the bottleneck in AI hardware, and whoever owns it captures the pricing. A 79.7% operating margin is the result. The open risk remains the one raised on 12–13 October, when CXMT said it would have a server-memory product by year-end. Prices rising 30% in a quarter give new entrants every reason to come in.

📋 6. The Rest of the Tape

General

Theme: “Central Banks Disagree About Oil, and Gold Is Not Listening to Any of Them” — Wednesday gave a direct look at how three central banks are handling the same energy shock, and they disagreed. Gold did not react to any of them. Oil rose on a reported US warning to Gulf states, and UBS said the inventory buffer that cushioned earlier disruptions has largely been used up. In equities, the hyperscaler results pointed to free cash flow as the measure the market is now using.

  1. How Central Banks Are Treating Oil

The question every central bank faces is whether an oil shock is a one-off rise in the price level that will fade, or the start of broader inflation. Wednesday showed three different answers.

Central bank Action Reasoning on energy
Bank of Canada 🟥 Hiked to 2.50% Energy prices are persistent; hiked to stop them spreading, even with core inflation near 2%
Fed ⬜ Held 10–2; two votes to hike “Inflation remains elevated”; no specific energy language reported
RBNZ 🟩 Held 4–3 (surprise) Oil volatility noted; subdued wages and stable expectations outweighed it

 

The Bank of Canada was the most explicit. It hiked even though inflation excluding gasoline was only 2.2% and core measures were near 2%. That is a central bank tightening against the risk that energy spreads, rather than against evidence that it already has. Its reasons: the conflict is in its eighth month “with no clear end in sight,” near-term inflation expectations have risen, and businesses are expected to pass on more costs over the next year.

The RBNZ came to the opposite view on similar evidence, by one vote. Its reasons point to how a central bank decides that an oil shock will fade: wages not responding, and medium-term expectations stable.

The Fed sits between them, closer to Canada. Two dissenters already want to hike, the statement upgraded growth to “expanding at a strong pace,” and US one-year household inflation expectations (Michigan) are at 4.7%. That last figure is the US equivalent of the risk the Bank of Canada was acting on, and it is higher than anything the RBNZ cited.

For the December decision, the Canadian reasoning is the one to watch. If the Fed comes to view energy as Canada does, a December hike is close to certain. What would move it toward New Zealand’s view is softer wage growth and stable long-run expectations. Michigan long-run expectations held at 3.5% in October after two rises, and Friday’s Q3 Employment Cost Index is the next wage reading.

  1. Gold: Dropping the Central-Bank Channel

Over four sessions this publication has tested a specific claim: that oil reaches gold through expected central-bank tightening. The evidence is now against it.

Event What the channel predicts What gold did
Tuesday: Fed December odds 73% → 83% Gold falls 🟥 Rose 0.90%
Wednesday: RBNZ dovish surprise Gold rises ⬜ No reaction reported
Wednesday: BoC hike citing energy Gold falls ⬜ No reaction reported
Wednesday: Fed hawkish hold, two hike dissents Gold falls ⬜ Flat; dollar softer

 

Gold rose when Fed hike odds rose, and ignored three central-bank decisions in one day, one of which was aimed specifically at energy. As this publication said it would on Tuesday, it drops the central-bank part of the hypothesis.

What survives is narrower: gold has moved opposite to oil in the last three sessions, two of them by enough to count. This publication does not have an explanation for that link, and the candidates set out on Tuesday (direct inflation-expectations effects, physical and ETF demand, or a common driver behind both) remain untested. Silver continues to fall and the gold/silver ratio has passed 75, so whatever is supporting gold is not lifting the wider precious-metals complex.

The practical stance for readers is unchanged: watch gold against oil, do not trade on the link. Two counted sessions are not enough to rely on, and the explanation that made the link seem logical has now failed.

  1. Oil: A Dated Warning, and a Thinner Buffer

Two things changed in oil on Wednesday.

The first is that the post-midterm window has become more specific. Until Wednesday the evidence for US action after 3 November was Trump’s own comments. The Wall Street Journal’s report that Washington warned Gulf Arab states to prepare for a possible renewed military campaign after the midterms is a different kind of evidence: it describes preparation, not rhetoric. At the same time, Qatar delivered a new draft deal to Tehran, with a response expected by the end of the week.

Date Event Why it matters for oil
By ~30 Oct Iran’s response to Qatar’s draft The last diplomatic marker before the midterms
3 Nov US midterm elections The date the US has linked to a decision
4 Nov Official Israeli election results Coalition talks begin; incumbent stays meanwhile

 

The second change is UBS’s point about buffers. Earlier in the conflict, strategic reserve releases and high inventories absorbed supply losses. UBS says pledged reserve releases are now largely complete and global inventories have fallen sharply, while “Middle Eastern flows, while improving, have not yet fully normalised.” The US SPR is at 282.6 million barrels, and US commercial crude stocks have drawn for four weeks.

This changes how this publication describes the oil market. It has called Brent a threat premium sitting on recovered flows. That is still right about flows, but the second half of the picture is that there is now less stored oil to fall back on if the threat is realised. A disruption after 3 November would hit a market with thinner inventory cover than at any earlier point in the conflict. That is the most likely reason the Brent prompt spread has steepened even as diplomacy has advanced.

UBS’s revised view of Brent near $110 through year-end sits a little below the current price, around $115. It implies some of the premium fades without a deal or a disruption.

  1. The Yield-Leadership Rule — A Partial Confirmation

The 10-year rose about 5 basis points, enough to score the rule for the first time since Friday.

Sectors fit: technology led at +0.8%, and the rate-sensitive sectors, real estate and utilities, were worst at −0.8% each. Small caps did not fit: the Russell (+0.41%) beat the S&P 500 (+0.28%), which the rule would not predict when yields rise. As on 23 October, the result is a partial confirmation.

The tally is now six clean confirmations and two partials in eight scored sessions. In both partials the sector pattern held and the small-cap pattern did not. That suggests the rule describes sector leadership more reliably than the large-versus-small split, and this publication will score the two separately from here.

  1. Where the Frameworks Stand
Framework Status Note
Yield-leadership rule — sectors 🟩 Holding Fit in every session scored on sectors, including Wednesday
Yield-leadership rule — small vs large 🟨 Weaker Failed in both partials
Returns vs spending → free cash flow 🟩 Refined Oracle, Microsoft up; Alphabet, Meta down
Gold inverse to oil 🟩 2 counted of 3 Cause unexplained
Gold via central banks 🟥 Dropped Ignored Fed odds and three decisions
Oil as threat premium on recovered flows 🟩 Holding, with thinner buffers UBS: reserves released, inventories down
10-year range 5.05–5.35% 🟩 Intact 22 Oct break reversed; back at 5.05%
NZD/CAD relative policy 🟥 Withdrawn RBNZ surprise hold
“Own the constraint” (memory) 🟩 Strongly supported SK Hynix 79.7% margin; DRAM +~30% q/q

 

🌐 Global Macro Sentiment Summary

Theme Direction Evidence Read
Fed 🟥 Hawkish hold 10–2; Cook and Hammack for a hike December ~83–85%
Bank of Canada 🟥 Hiked 2.50%; aimed at energy spreading May go further
RBNZ 🟩 Surprise hold 4–3; NZD below 0.5600 Close call
Oil 🟥 Higher Brent ~$115; WTI ~$101 WSJ: Gulf states told to prepare
Oil buffers 🟥 Thinner UBS: reserve releases done; inventories down Less cover if disrupted
Gold ⬜ Flat ~$4,057; no reaction to three decisions Central-bank channel dropped
Silver 🟥 Lower ~$53.72; ratio above 75 Not following gold
10-year 🟥 Back at 5.05% +~5bp Range intact
Hyperscalers ⬜ Split Microsoft +2.5% AH; Alphabet −9% AH; Oracle +10.7% Graded on free cash flow
Memory 🟩 Strong SK Hynix +13.3%; DRAM +~30% q/q Own the constraint
Equities 🟩 Modestly higher S&P +0.28%; VIX 16.83 Fed confirmed pricing

 

Compliance note: No Wednesday settlement prices for oil or gold were found; levels are intraday or late-session quotes, and Yahoo ticker figures for both are inconsistent with Tuesday’s settlements. No closing 2-year or 30-year yields were found. The NZD level is from a headline. Microsoft’s Azure and capex figures are from a headline whose article could not be retrieved. The WSJ report on US warnings to Gulf states is cited via The National. Bloomberg’s headline on the RBNZ conflicts with all other sources and is not used. The free-cash-flow refinement rests on three companies in one session.

Upcoming News

Thursday, October 29th, 2026 — Theme: “The Data Behind December” — Third-quarter GDP and September PCE inflation arrive together at 19:30 Hanoi time, the day after a Fed hold with two votes to hike. The ECB decides at 20:15, Amazon reports after the US close, and Iran is expected to answer Qatar’s draft deal by the end of the week.

🔴 Calendar — Thursday, October 29th, 2026

Times in ICT (Hanoi). ET is ICT minus 11 hours. Times marked  are standard release times not confirmed in the sources consulted.*

Time (ICT) Currency Event / Indicator Consensus Impact
06:30* JPY Tokyo CPI (Oct) Sep 2.7% y/y 🟠 Med — BoJ Friday
Morning* SEK Swedish GDP (Q3) — 🟡 Low
Before London open* — Shell Q3 results — 🔴 High for oil
16:00* EUR Eurozone economic sentiment (Oct) — 🟡 Low
19:30 USD Q3 GDP, advance 3.0% q/q SAAR (GDPNow 3.5%) 🔴 Highest
19:30 USD September PCE, headline +0.5% m/m; 3.6% y/y 🔴 Highest
19:30 USD September core PCE +0.3% m/m; 3.1% y/y 🔴 Highest
19:30 USD Personal income and spending (Sep) — 🟠 Med
19:30 USD Initial jobless claims — 🟠 Med
20:15* EUR ECB decision No hike expected 🔴 High
20:45* EUR Lagarde press conference — 🔴 High
21:30* USD EIA natural gas storage — 🟡 Low
00:00 Fri USD $44bn 7-year note auction — 🟠 Med
~03:00 Fri* — Amazon Q3 results — 🔴 High
By end of week — Iran’s response to Qatar’s draft deal — 🔴 High for oil

 

Consensus for GDP, PCE and the ECB from Newsquawk’s week-ahead preview. Friday: BoJ (hold at 1.25% expected), ExxonMobil, Chevron and BP before the open, the Q3 Employment Cost Index, Chicago PMI, eurozone GDP and flash CPIs.

  1. PCE and GDP — The Case for December, Tested

The Fed’s statement on Wednesday described growth as “expanding at a strong pace” and inflation as “elevated.” Thursday’s data will show whether both descriptions are right.

Release Consensus What would change December pricing (~83–85%)
Q3 GDP 3.0% (GDPNow 3.5%) At or above 3.5%: supports the “strong pace” upgrade
Headline PCE 3.6% y/y (from 3.4%) Mostly energy; already expected
Core PCE 3.1% y/y (from 3.0%); +0.3% m/m The key line. +0.4% or more strengthens December; +0.2% weakens it

 

With December priced at 83–85%, the market has less room to rise than to fall. A core PCE print below consensus would do the most to move pricing. An in-line print mostly confirms what is priced.

One data note to resolve. This publication has cited a core CPI figure of 2.6%. Consensus for core PCE is 3.1%. Core PCE normally runs below core CPI, so one of those figures is likely wrong or not comparable. Last Friday’s edition flagged this; Thursday’s release will give the PCE side directly.

Timing matters too. September PCE predates October’s move in oil. Brent rose about 7% in the two sessions to 23 October, so the full energy effect will show in the October data, published after the next FOMC meeting.

  1. The ECB

Newsquawk expects no hike and a statement that commits to nothing in particular, with the political situation in France a focus.

Wednesday’s split between the Bank of Canada and the RBNZ makes the ECB’s language on energy more informative than usual. If the ECB describes energy as a risk to medium-term inflation, as Canada did, it is leaning toward tightening. If it describes oil as volatile and temporary, as New Zealand did, it is not. No EUR/USD level was found for Wednesday; it was 1.1136 on Tuesday.

  1. Oil — Shell, and Tehran’s Answer

Shell is the first major to report this week, ahead of ExxonMobil, Chevron and BP on Friday.

The test is the one this publication set out on Monday: at Brent around $115, does a major raise capital spending, which means it expects high prices to last, or raise buybacks, which means it is taking the cash and not betting on the price? UBS’s forecast of Brent near $110 through year-end, and the annual forecasts in the September Reuters poll, which topped out at $97.60, suggest the industry itself does not expect current prices to hold. Shell’s capital plan is the first direct read.

Iran’s response to Qatar’s draft deal, expected by the end of the week, is the last diplomatic marker before the 3 November midterms. A rejection, set against the WSJ’s report that the US has told Gulf states to prepare for a renewed campaign after that date, would leave the escalation path as the main one. An engagement would give the market its first reason in over a week to price out some of the premium on diplomacy alone.

  1. Amazon — The Free-Cash-Flow Test

Amazon reports after the US close on Thursday, early Friday in Hanoi. On Wednesday’s evidence, the line that will matter most is free cash flow: operating cash flow minus capex.

Oracle was rewarded for a credible path to positive free cash flow. Microsoft was rewarded for beating with capex below forecast. Alphabet was punished for spending $9.1 billion more than it generated, despite cloud revenue doubling. Amazon’s AWS growth will get the headlines, but on this season’s pattern the share reaction will follow the cash line. No consensus figures were found in the sources consulted.

  1. Gold Into the Data

Gold has ignored three central-bank decisions and a 10-point move in Fed hike odds. Thursday’s PCE and GDP will show whether it also ignores US inflation and growth data.

Kitco levels: support $4,039, then $4,015; resistance $4,059, then $4,079. Silver support $53.24, then $52.41; resistance $54.46. The gold/silver ratio above 75 is the measure to watch for whether metals are recovering as a group. It needs to fall back below about 74 for that.

  1. Carry-Over Into Thursday
  1. Three Questions for Thursday

Does core PCE come in at or below 0.3% month-on-month? That is the single most important number for whether December pricing holds at 83–85%.

Does the ECB describe energy the way Canada did, or the way New Zealand did? That tells you which side of the split the second-largest central bank is on.

Does Amazon spend within its cash flow? After Oracle, Microsoft and Alphabet, that is the line the market is grading.

Snapshot

Wednesday, 28 October 2026: a hawkish Fed hold with two votes to hike, a Bank of Canada hike aimed at energy, a surprise RBNZ hold, oil higher on a reported US warning to Gulf states, gold unmoved through all of it, and hyperscalers graded on free cash flow.

📌 The Session in Eight Lines

  1. The Fed held at 3.75–4.00% by 10–2; Governor Cook and Cleveland’s Hammack voted for a hike, the first governor-and-president dissent on the same side since 1992.
  2. The Bank of Canada hiked to 2.50% to stop energy prices spreading; the RBNZ held at 2.75% by 4–3 against expectations, and NZD/USD fell below 0.5600.
  3. Brent traded around $115 and WTI around $101 after the WSJ reported the US told Gulf states to prepare for a possible renewed campaign after 3 November.
  4. UBS said strategic reserve releases are largely complete and inventories have fallen, and now sees Brent near $110 through year-end.
  5. Gold spot was flat at about $4,057 through all three decisions; silver fell about 1% and the gold/silver ratio rose above 75.
  6. The 10-year rose about 5 basis points to 5.05%, back at the floor of its old range.
  7. Oracle rose about 10.7% on a $30 billion free-cash-flow forecast; SK Hynix ADRs rose 13.3% on memory prices.
  8. After hours, Microsoft rose 2.5% on a beat with capex below forecast; Alphabet fell about 9% on a profit miss and free cash flow of −$9.1 billion.

📊 Reference Levels

Instrument Level Change Note
Fed funds target 3.75–4.00% Hold 10–2; two for +25bp
December Fed hike odds ~83–85% ~78% before Kitco, Yahoo; Fortune
Bank of Canada 2.50% +25bp 🟥 May go further
RBNZ OCR 2.75% Hold 🟩 4–3 surprise
NZD/USD <0.5600 Lower 🟥 Headline only
Brent ~$115 ~+1% No settlement found
WTI ~$101–102 ~+1.5–2% No settlement found
EIA crude stocks −1.9m bbl vs ~flat 🟥 Fourth draw
EIA gasoline stocks +3.9m bbl vs −0.9m 🟩
US crude output 13.922m bpd +653k 🟩 Hurricane recovery
Spot gold ~$4,056.80 −0.06% ⬜ Flat
Spot silver ~$53.72 −1.07% 🟥
Gold/silver ratio ~75.5 from ~74.7 🟥 Above 75
S&P 500 7,743.98 +0.28% 🟩
Nasdaq Composite 26,549.85 +0.27% 🟩
Dow Jones 52,475.10 +0.19% 🟩
Russell 2000 2,830.51 +0.41% 🟩
VIX 16.83 −3.9% —
Philadelphia Semiconductor 11,674 +1.75% 🟩
10-year Treasury 5.05% +~5bp Back at range floor
Oracle ~$144.5 +10.7% 🟩 FCF plan
Microsoft (AH) $544.41 +2.52% 🟩
Alphabet (AH) ~$291.50–298.68 −8% to −9.4% 🟥 FCF −$9.07bn
SK Hynix ADR $164.70 +13.3% 🟩

 

🥇 Gold — Standing Section

Part of the hypothesis Status Evidence
Gold moves inversely to oil 🟩 2 counted of 3 Wednesday: oil up, gold −0.06% (too small to count)
Link runs through central banks 🟥 Dropped Rose on higher Fed odds Tue; ignored three decisions Wed
Silver confirming 🟥 No Ratio above 75

 

Gold did not react to a hawkish Fed, a Bank of Canada hike aimed at energy, or a dovish RBNZ surprise. This publication drops the claim that oil reaches gold through central-bank expectations. What remains is an inverse link with oil over three sessions, two of them large enough to count, with no established cause.

Levels (Kitco): resistance $4,059, $4,079; support $4,039, $4,015. Silver: resistance $54.46; support $53.24, $52.41. No December futures settlement was found for Wednesday.

🛢️ Oil — Standing Section

Element Evidence Direction
Escalation path WSJ: US told Gulf states to prepare for renewed campaign after 3 Nov 🟥 More concrete
Diplomatic path Qatar’s new draft; Iran to respond by end of week ⬜ Pending
Buffers UBS: reserve releases largely done; inventories down sharply 🟥 Thinner
Flows “Improving, not yet fully normalised” (UBS) 🟩 Recovering
US stocks Fourth weekly crude draw; gasoline build ⬜ Mixed
Forecasts UBS: Brent near $110 through year-end ⬜ Slightly below spot

 

Brent is still a threat premium on recovering flows, but there is now less stored oil behind it. A disruption after 3 November would meet thinner inventory cover than at any earlier point in the conflict, which explains why the prompt spread has steepened while diplomacy has advanced.

🧭 Sentiment & Bias

Asset / Theme Bias Rationale
Fed path 🟥 Hawkish Two hike dissents; growth upgraded; December ~83–85%
Front end ⬜ Data-dependent Most risk now in a soft core PCE print
10-year ⬜ Range 5.05–5.35% 22 Oct break reversed
USD ⬜ Neutral Confirmed, not surprised, by the Fed
NZD 🟥 Weak Surprise hold; relative-policy view withdrawn
CAD 🟩 Supported Hike plus guidance; oil higher on the day
Brent 🟥 Upside risk after 3 Nov Thinner buffers; concrete escalation preparation
Gold ⬜ No directional view Not responding to policy; oil link unexplained
Silver 🟥 Weak Ratio above 75
Hyperscalers ⬜ Graded on FCF Positive FCF rewarded; negative punished
Memory 🟩 Strong DRAM +~30% q/q; 79.7% margin at SK Hynix

 

🎯 Top Trade Takeaway

The highest-conviction observation from Wednesday is the free-cash-flow split among the AI spenders. Oracle (+10.7%) promised positive free cash flow and capped borrowing. Microsoft (+2.5% after hours) beat while spending less than expected. Alphabet (about −9% after hours) spent $9.1 billion more than its operating cash flow, despite doubling cloud revenue. Meta, which fell 6.98% on 22 October, had negative free cash flow and then announced a $95 billion equity raise.

The practical test for every remaining AI-related report, starting with Amazon on Thursday, is operating cash flow minus capex. This is based on three companies in one session plus Meta last week, so it is a pattern, not a rule. But it explains more of the season’s reactions than any other single measure this publication has tried.

For an FX book, the lesson is about probabilities. A 75–80% priced RBNZ hike turned into a 4–3 hold, and NZD fell below 0.5600. With December Fed pricing at 83–85%, the same risk applies in reverse to the USD: a core PCE print below consensus on Thursday would be the outcome least priced.

For oil, the risk now leans to the upside after 3 November, because inventories that cushioned earlier disruptions have been drawn down. For gold, there is still no basis for a directional view.

📚 Full Source Citations

Source Used for Note
Fortune — Fed holds rates steady; governor and president dissent together Vote, dissenters, 1992 precedent, Warsh “monetary medicine” quote, statement language, Ullrich and Woessner quotes, pre-decision CME odds Reports no Warsh comment on December or energy
Yahoo Finance — Stock market today, 28 October 2026 Index closes, 10-year, December odds after the decision, Warsh quote, Microsoft and Alphabet after-hours 🟥 Gold and WTI ticker figures inconsistent with Tuesday’s settlements; Microsoft headline says +4%, figure says +2.52%
StockTitan — Stock market today, 28 October 2026 Decision details, 2027 strategy meeting, after-hours reporters, Fortinet headline Data to 3pm ET only
TradingKey — US stocks close; Oracle, SK Hynix Index closes, SOX, Oracle plan, SK Hynix results —
TradingKey — Alphabet Q3 results Full Alphabet figures, securities loss, capex, FCF, Cloud, Search —
investingLive — Microsoft Q1 headline Azure +48%; capex ~$6bn below forecast 🟥 Article could not be retrieved; headline only
BNN Bloomberg — Text of the Bank of Canada decision Rate, reasoning, inflation, energy, labour market, guidance Primary text
TradingView (Trading Economics) — RBNZ holds at 2.75% Decision, 4–3 vote, reasoning, guidance —

 

The report belongs to The Concept Trading and Van Hung Nguyen

Tagged , ,