Gold Had Its Best Week Since August. Then Jobs Brought the Rate Debate Back.

Spot gold rose 4.89% and silver 5.88% in the week to 6 November 2026 as oil fell about 13%, but Friday’s report of 165,000 new US jobs revived the case for a December Fed rate rise. Over the week the nominal 10-year Treasury yield fell 8 basis points while the real yield rose 3, and our recorded Tactical Score ended the week at 44, one point below a week earlier and back in the Cautious band.

Published November 8, 2026 · 14 min read

Key Takeaways

• Spot gold closed at US$4,065.04 on 6 November 2026, up 4.89% on the week, and silver rose 5.88% to US$52.0165; December COMEX gold futures settled at US$4,074.90, up 4.51%. Both metals ended three straight weekly falls.
• October payrolls rose 165,000, August and September were revised up a combined 84,000, and unemployment was 4.1%. Reuters had the FedWatch odds of a December rise back towards 80% after the report, from about 67% on Thursday.
• The nominal 10-year Treasury yield fell 8 basis points to 4.86% while the real yield rose 3 to 2.63%, so the inflation compensation between them narrowed by 11, to 2.23 percentage points.
• January Brent fell 13.32% to US$98.32 and December WTI 12.76% to US$90.83, while the S&P 500 lost 2.7% and the Nasdaq Composite 3.4%.
• Our recorded Tactical Score was 44/100, Cautious, on 6 November, from 45 on 30 October: the dollar input cost the core block three points, and momentum and the confirmation basket gained one each. The direction indicator moved from Improving to Stable on a reading that barely changed, +0.10 to +0.11.

Gold’s rebound survived a difficult Friday. Spot gold ended the week to Friday 6 November 2026 at US$4,065.04 on our daily series, 4.89% above its close on 30 October, and spot silver rose 5.88% to US$52.0165. On the futures market, the December COMEX gold contract settled at US$4,074.90, 4.51% above its settlement of US$3,898.90 a week earlier. Both metals ended three straight weekly falls. For gold it was the largest weekly rise since August: on Friday morning Reuters described it as on course for its best week in more than two months, none of the weekly moves since late August came close, and in the week to 7 August gold had gained more than 7%. Prices are in US dollars, and metals are per troy ounce.

That recovery came with an important qualification. The relief rally after the US midterm elections met a stronger US employment report, renewed expectations of a December Federal Reserve rate rise, and a recovering dollar. The question for a metals investor is how much the underlying environment improved, and how much of the week was a rebound within an environment that is still difficult.

+3 basis points Change in the US 10-year real yield in the week to 6 November 2026, while the nominal 10-year yield fell 8 basis points. Gold rose in a week when the inflation-adjusted return on a 10-year Treasury inflation-protected security edged higher. A falling nominal yield alone is an incomplete description of gold’s opportunity cost.

What did the week deliver?

Metals rose, stocks fell and oil fell sharply between 30 October and 6 November 2026. Spot gold gained 4.89% and spot silver 5.88%, all four main US share indexes ended lower, and Brent and WTI crude each lost about 13%.

The spot and futures readings for gold differ because they are different instruments read at different times: the futures figure is the exchange’s settlement price for the December contract, while the spot series is a later reading of the market for immediate delivery. Brent’s comparison uses the January contract on both dates, because the December contract expired on 30 October; a front-month series that rolls from one contract to the other over the week shows a larger fall than the contract itself did.

What did Friday’s jobs report change?

Friday’s jobs report moved attention from the election back to monetary policy. Our 5 November article examined gold’s recovery after Democrats won both chambers of Congress, when a falling dollar accompanied a sharp rise in the metal. Markets were reassessing the prospects for fiscal policy and for the Middle East conflict, but those were still interpretations of an election result, not completed changes in policy.

October nonfarm payrolls rose by 165,000, the Bureau of Labor Statistics reported. Unemployment was 4.1%, from 4.2% in September, a rate the bureau said “changed little”. Average hourly earnings rose 14 cents, or 0.4%, to US$38.00, and were 3.1% higher than a year earlier. August and September were revised up by a combined 84,000, to gains of 186,000 and 60,000.

The revisions mattered as much as the new month. The previously reported September gain of 29,000 had suggested much weaker hiring. Friday supplied both a stronger October and a less fragile recent history, which weakened the argument that a labour market already losing strength would hold the Fed back.

Reuters reported at 2:43 p.m. Eastern on Friday that market-implied odds of a December rise of at least 25 basis points had moved back up towards 80% after the report, by CME Group’s FedWatch tool, from about 67% on Thursday, when US stocks fell sharply. The dollar index was up 0.77% on the day at that point and about 0.3% for the week, and Reuters noted it had already been rising before the release, on safe-haven demand. Treasury yields rose after the report. These are timestamped readings, not closing values, and none of them is a promise about the Fed’s decision.

The metals gave back part of the week on Friday. December gold futures settled 0.94% lower and December silver 0.83% lower, and on our spot series gold ended the day 0.55% down. The week’s advance stayed intact, but the policy headwind had reasserted itself.

Did the productivity report point the other way?

Thursday’s productivity report did offer a counterweight on inflation. Nonfarm business productivity rose at a seasonally adjusted annual rate of 3.6% in the third quarter, the Bureau of Labor Statistics reported on 5 November. Hourly compensation rose 2.9%, so unit labour costs fell 0.7% at an annual rate, although they were still 2.4% higher than a year earlier. Manufacturing was less reassuring: its unit labour costs rose 2.7% at an annual rate, and 4.0% on the year.

The distinction is practical. A business can pay more per hour without an equal rise in labour cost per unit of output if each worker produces more in that hour. Stronger wages and moderating production costs can therefore coexist.

The two releases also measure different things over different periods. October average hourly earnings are not the same series as third-quarter hourly compensation; a quarterly figure at an annual rate is not a year-on-year change; and one preliminary productivity estimate cannot show that inflation has been beaten.

Why did real yields rise while nominal yields fell?

Real yields rose while nominal yields fell because the inflation compensation built into the nominal yield narrowed over the week. The US Treasury’s own curves show it, and they are the clearest reason to resist a simple “lower yields helped gold” explanation.

A nominal yield is stated in dollars before any adjustment for inflation. The real yield on a Treasury inflation-protected security is the market’s quoted return above the inflation adjustment built into that security. It is not found by subtracting today’s consumer price inflation from a conventional bond’s yield.

The difference between comparable nominal and real yields is commonly used as a measure of inflation compensation. It reflects inflation expectations alongside risk and liquidity premiums, so it should not be read as a precise inflation forecast. The 10-year breakeven series that our dashboard reads from the Federal Reserve Bank of St. Louis gave 2.24% for 5 November, the same figure the Treasury’s rows for that day produce, 4.84% less 2.60%.

This week the difference narrowed by 11, from 2.34 to 2.23 percentage points. Gold recovered despite a slightly higher real yield, rather than on the back of a fall in the inflation-adjusted return available on government bonds. Friday itself reinforced the distinction: the nominal yield rose from Thursday’s 4.84%, and the real yield from 2.60%. The Academy’s Market Context section reads rates, inflation and the dollar together.

Did cheaper oil ease the pressure on inflation?

Cheaper oil eased one source of inflation pressure, but it did not end the risk behind it. Brent and WTI each fell about 13% over the week, and Brent had its steepest weekly percentage fall since the start of the COVID-19 pandemic, according to market reviews from eOption and PFL Petroleum Services. Both linked the fall to hopes, after the election, of a de-escalation in the war with Iran; PFL also noted continued attacks on commercial vessels in the Gulf and the Strait of Hormuz, where a Greek ship manager reported an explosion aboard an oil products tanker on 5 November that was believed to have been caused by a missile.

Our interpretation is that cheaper energy offered relief from one source of inflation pressure, while the remaining threat to shipping kept that relief vulnerable. The fall in oil is consistent with the lower inflation compensation in bonds, but the two moving together does not prove that oil alone caused the change in yields.

For gold, this is more complicated than “geopolitical risk goes up, gold goes up”. An energy shock can increase demand for protection while also encouraging higher interest rates. Conversely, a smaller risk premium in oil can ease the expected policy burden even while political uncertainty remains. The balance between those channels can change within a week.

Did metals and stocks move differently?

Metals and stocks moved in opposite directions over the week. All four US share indexes in the table finished lower, despite Friday gains in the S&P 500, the Nasdaq Composite and the Russell 2000, while gold and silver rose.

One week of opposite moves illustrates why investors consider assets that respond differently to economic risks. It does not establish a permanent hedge, or show that any particular allocation would have been appropriate. A portfolio’s result also depends on its starting weights, its other holdings, its currency exposure and the costs of ownership.

The Portfolio Impact calculator tests the effect of a chosen move in gold at different position sizes. It holds everything else in the portfolio unchanged, so it isolates the gold decision rather than forecasting a diversified portfolio. Our guide on when to increase or decrease an allocation covers the same decision from the other side.

Structural demand still needs a price and policy context

Our article on third-quarter demand set out an estimated 321 tonnes of official-sector buying and 211 tonnes of ETF inflows in the third quarter, alongside bar and coin demand that fell 19%. That is evidence about who bought during an earlier period. It is not a measure of Friday’s trading.

The World Gold Council’s commentary of 5 November, “Where floor art thou?”, found two of its three conditions for an interim bottom in place, with a washout in speculative positioning still missing. Its figures ran to 28 October, and to 27 October for positioning. It was not an assessment made after the jobs report.

Friday did not rewrite that demand history. It changed the near-term policy setting in which the next buyers and sellers must act. A strong rebound cannot, by itself, show that the Council’s remaining condition has now been met.

What did our Tactical Score record on Friday?

Our recorded Tactical Score for Friday 6 November was 44/100, in the Cautious band, one point below the 45/100 recorded on Friday 30 October, which was also Cautious. Both are the Macro Conditions dashboard’s stored Engine V2 runs of about 20:00 UTC; Friday’s run came after the jobs report and before the US market close. In between, the score read 50, 45, 49 and 47, so it spent three of the week’s five runs in the Neutral band before returning to Cautious. The record of each run was the same when we read it on 7 November as when we first read it on the afternoon of 6 November; nothing here was recalculated with revised data. The Neutral band begins at 46.

The change came from the recorded components, not from the week’s headlines. The core block fell from 19 to 16 points because the dollar input dropped from 5 to 2: the dollar index’s 13-week change, which the engine reads, rose from about +2.0% to +2.5% and crossed the 2% line between two of its brackets. The real-yield, inflation-expectations and financial-stress inputs scored the same on both Fridays. Momentum rose from 16 to 17 points, and the confirmation basket from 5 to 6, when the credit-spread input gained a point. Fiscal stress was unchanged at 5.

The direction indicator read Improving on 30 October and Stable on 6 November, but its underlying reading barely moved, from +0.10 to +0.11. The engine changes the label only when two runs in a row fall in a new range: on 30 October the reading was already in the Stable range, with the change pending, and it was confirmed only after the readings of 5 and 6 November were both in that range. The level and the direction say the same thing in different ways: conditions were slightly less favourable than a week earlier, and the inputs were not moving decisively either way.

The input dates matter as much as the scores. Friday’s run read the real yield as of 4 November and inflation expectations as of 5 November, because the published daily series it uses run a day or two behind; it did not see the Treasury’s Friday real yield of 2.63%. The weekly financial-stress and term-premium readings were as of 30 October, and the futures-positioning reading was for positions on 27 October on both Fridays. Only the market inputs, such as the dollar and gold, were read on the day, during trading. A fresh dashboard timestamp does not mean every input reflects Friday’s close.

The Macro Conditions Score reflects macro conditions historically associated with precious metals performance. It did not predict the week’s rise, and a reading is not an instruction to change an allocation. The Two-Horizon framework separates the purpose of a strategic holding from decisions about additional tactical exposure. That distinction remains useful when market signals disagree. Neither reading substitutes for position limits or an investor’s circumstances.

Jeffrey Christian: waiting buyers and incomplete confirmation

In a CPM Group presentation published on 3 November, Jeffrey Christian, the firm’s managing partner, described investors waiting for prices to stabilise and begin rising before buying again. CPM’s summary of it says the firm considered a move towards US$3,800 possible but not certain, and did not see further short-term weakness as necessarily ending the longer-term bull market. December gold futures were trading at about US$3,956 at the recording. Those were conditional views offered before the election result and Friday’s jobs report.

Our reading of the week is that a price recovery supplied one kind of encouragement, while payrolls supplied a reason to remain careful about the macro outlook. Christian’s commentary is useful for the distinction it draws between buying interest and certainty about timing; the rally that followed is not a verdict on every part of his view.

What deserves attention next?

The next scheduled tests are October consumer prices on Tuesday 10 November and October producer prices on Friday 13 November, both at 8:30 a.m. Eastern, on the Bureau of Labor Statistics calendar. Neither will capture this week’s fall in oil, because both cover October.

The useful questions are whether inflation supports the renewed case for tightening, whether real yields and the dollar sustain Friday’s strength, and whether gold can hold its recovery under those conditions.

For a portfolio owner, the immediate task is to revisit the purpose and size of the holding. The Academy’s Physical vs Paper section covers the choices of ownership, and The Case Against helps test the reasons for committing capital. A futures return does not measure the net proceeds of selling a coin after spreads, storage, taxes or currency conversion.

Gold’s strong week and Friday’s renewed rate pressure can both be true. A durable portfolio process needs room for both: a clear reason for owning metals, a defined tolerance for losses, and evidence strong enough to justify changing the plan.

Sources and Methodology

This review compares Friday 30 October with Friday 6 November 2026; every figure keeps its own reference date. Spot gold and silver are Investing.com daily-history rows, as read for our review of the same week; the three earlier weekly falls are from the same series in our reviews of 17, 24 and 31 October. Gold futures are COMEX settlements of the December contract, for 6 November as given by eOption and FXEmpire’s contract table and for 30 October as reported by Reuters; silver futures are the December contract’s Friday settlement only, because we did not read its 30 October settlement. Oil settlements for 6 November are as given by eOption and PFL Petroleum Services, and for 30 October as reported by Reuters, comparing the same contract on both dates. Share indexes are the Associated Press’s closing levels and weekly changes. Yields are the Treasury’s par and real par curves; percentage changes are (later ÷ earlier − 1) × 100, basis-point changes are (later − earlier) × 100, and the nominal-less-real gap is our subtraction. Payrolls, earnings, revisions and productivity are the bureau’s own figures as first published and may be revised.

The Tactical Score readings are the dashboard’s recorded Engine V2 runs for 30 October and 6 November 2026, with their blocks, component signals, direction readings and input dates, read from the record on 6 and 7 November and compared; nothing was recalculated or adjusted for this article. The band edges, the dollar input’s brackets and the two-run rule for the direction label are from the engine’s code. Proposed explanations and portfolio implications are editorial analysis, not a causal audit, and no rate-rise probability was calculated by us.

Conflicts of interest, stated plainly. CPM Group sells research, consulting and trading recommendations on the metals it comments on, and the World Gold Council is funded by gold-mining companies. eOption is a division of Regal Securities, a broker-dealer, and PFL Petroleum Services a petroleum services firm; their market reviews are cited only for settlement prices and the reasons they reported. GoldSilverPortfolio has no affiliate or commission arrangement with any firm named here. It may earn a commission through two links elsewhere on this site: an affiliate link to SilverGoldBull, a bullion dealer, and a referral link to Wealthsimple, a brokerage. It also sells TaxStack, a paid record-keeping tool, and publishes the Macro Conditions Score discussed here. No fund, dealer, exchange, trading signal or analyst’s view is presented here as a recommendation to buy or sell.

Primary sources

• Investing.com — XAU/USD and XAG/USD historical data, daily-history rows for 30 October and 5 and 6 November 2026, read on 6 November 2026: gold US$3,875.44, 4,087.71 and 4,065.04; silver US$49.1277, 51.7634 and 52.0165
• eOption — “Market Review November 06, 2026”, 6 November 2026: December gold settled at US$4,074.90 (−0.94%) and December silver at US$51.94 (−0.83%); WTI settled at US$90.83 and Brent at US$98.32
• FXEmpire — Gold futures contracts and Silver futures contracts, read on 7 November 2026: December 2026 gold 4,074.90 and silver 51.942, the 6 November settlements
• PFL Petroleum Services — “The PFL Petroleum Markets Report for November 6, 2026”, 6 November 2026: Brent US$98.32, WTI US$90.83, the attacks on shipping
• Reuters, carried by MarketScreener — Anna Hirtenstein, “Oil jumps 2%, settles at 6-month highs on fears of escalation in Iran”, 30 October 2026: WTI US$104.12 and January Brent US$113.43
• Reuters, carried by MarketScreener — Noel John and Polina Devitt, “Tech rout drives gold to one-year low, white metals sink deep into oversold zone”, 30 October 2026: December gold futures settled at US$3,898.90
• The Associated Press — “How major US stock indexes fared Friday 11/6/2026”, 6 November 2026: closing levels and changes for the week
• Reuters, carried by MarketScreener — Chuck Mikolajczak, “Dollar jumps after strong US payrolls report”, 6 November 2026, 2:43 p.m. ET: FedWatch towards 80%, about 67% on Thursday, the dollar index up 0.77% on the day and 0.3% on the week
• Reuters, carried by Kitco — “Gold drops but stays on track for best week in over two months”, 6 November 2026, before the jobs report
• Reuters, carried by Kitco — “Gold set for best week since January as inflation fears ebb”, 7 August 2026: prices up more than 7% on the week; and Kitco News, “Gold’s week in review: From seven-session rally to Jackson Hole rout”, 28 August 2026, weekly gold survey of 4 September (spot down 0.36% on the week) and of 18 September (spot from about US$4,340 to US$4,377)
• US Bureau of Labor Statistics — “The Employment Situation — October 2026”, USDL-26-1632, 6 November 2026, 8:30 a.m. ET
• US Bureau of Labor Statistics — “Productivity and Costs, Third Quarter 2026, Preliminary”, USDL 26-1631, 5 November 2026, 8:30 a.m. ET
• US Bureau of Labor Statistics — “Schedule of Selected Releases for November 2026”: Consumer Price Index for October on 10 November and Producer Price Index for October on 13 November, 8:30 a.m.
• US Department of the Treasury — Daily Treasury Par Yield Curve Rates and Daily Treasury Par Real Yield Curve Rates, 2026: 10-year 4.94%, 4.84% and 4.86%; 10-year real 2.60%, 2.60% and 2.63%, for 30 October and 5 and 6 November
• Federal Reserve Bank of St. Louis, FRED — 10-Year Breakeven Inflation Rate (T10YIE): 2.24% for 5 November 2026, as read by the dashboard’s 6 November run
• World Gold Council — Johan Palmberg, “Gold Market Commentary: Where floor art thou?”, 5 November 2026
• CPM Group — “Gold and Silver Alert: Buyers Are Waiting for One Thing”, video presentation and written summary, 3 November 2026
• GoldSilverPortfolio — Macro Conditions Score dashboard (Engine V2), recorded tactical runs of 30 October and 6 November 2026 with their blocks, component signals, direction readings and input dates, read from the record on 6 and 7 November 2026