Weak Jobs, Higher Yields: What a Week of Mixed Signals Means for Gold and Silver

US payrolls rose by 29,000 in September and the priced chance of an October rate rise fell from around 70% to about 22%. Yet the 10-year real yield rose 9 basis points to 2.92%, the dollar gained on the week, and front-month gold and silver futures fell 3.59% and 6.68%. Our Tactical Macro Conditions Score moved one point, to 35/100. What the week’s conflicting signals do and do not mean for a strategic allocation to precious metals.

Published October 5, 2026 · 18 min read

Key Takeaways

• US nonfarm payrolls rose by 29,000 in September 2026, and July and August were revised down by a combined 60,000, leaving July at a loss of 10,000 jobs. The unemployment rate was 4.2%. Second-quarter GDP was revised up to a 2.2% annual rate and real consumer spending rose 0.6% in August: hiring weakened while spending did not.
• Rate futures put the chance of a Federal Reserve increase at the 27–28 October meeting at about 22% on Friday 2 October, against around 70% earlier in the week, on Reuters’ reading of the CME FedWatch Tool. Much of that shift came before the jobs report, and an increase by December was still priced on Wednesday.
• Long-term yields rose anyway. Between 25 September and 2 October the US 10-year Treasury yield rose 11 basis points to 5.28%, the 30-year 14 to 5.63% and the 10-year real yield 9 to 2.92%. The 2-year rose 2, to 4.83%.
• Front-month COMEX gold lost 3.59% on the week to settle at US$4,133.70, and silver lost 6.68% to US$59.977, the largest weekly falls since June. Of the four main US share indexes only the Nasdaq composite gained, by 0.5%, and the high-yield credit spread widened 31 basis points through Thursday.
• Our Tactical Macro Conditions Score read 35/100, Cautious, for 2 October: one point above the previous Friday and three below 16 September, with the real-yield input already at 2 of its 20 points. The Long-Term Structural Score read 64/100. A week’s score does not set a strategic allocation, and diversification cushions a loss without preventing it: in a hypothetical 90/10 portfolio, a 20% fall in other assets and a 10% gain in gold still leaves a 17% loss.

US payrolls grew by 29,000 in September, and by Friday 2 October rate futures put the chance of another Federal Reserve increase this month at about 22%, against around 70% earlier in the week, on Reuters’ reading of the CME FedWatch Tool. Shares rose on the day. Gold and silver still finished the week lower, and the yield on long-dated US government debt finished it higher. This review covers Monday 28 September to Friday 2 October 2026. Index and metal figures are changes over that week unless a day is named, metal figures are settlements of the front-month COMEX futures contract and not retail bullion prices, and yields are the US Treasury’s daily par-curve readings.

+9 basis points Rise in the US 10-year real yield over the week: 2.83% on 25 September, 2.92% on 2 October 2026. In the same week the priced chance of an increase at the Federal Reserve’s 27–28 October meeting fell from around 70% to about 22%. The odds on one meeting and the return on a ten-year inflation-linked bond are different measurements, and they moved in opposite directions.

That gap is the story of the week. Hiring, consumer spending, factory surveys, inflation and financial markets did not send one message, and a reader who took only the indicator that suits an existing view would miss most of what changed. It is also a correction to something we wrote in July, when a weak jobs report arrived with a falling dollar, gold rebounded, and we called the labour market the switch that moves the metal. This time payrolls were weaker still, the dollar gained on the week, long-term yields rose, and gold fell. A jobs number does not set the price. What yields and the dollar do around it matters more.

For our Macro Conditions Score, two questions stay separate: how supportive the near-term environment is for metals, and what part metals should play in a portfolio over a much longer period. A difficult answer to the first does not settle the second, which is the argument of our two-horizon framework.

Did Friday’s rally mean the week went well?

No. All four of the main US share indexes rose on Friday 2 October, but three of them fell over the week. The Associated Press’s closing table has the S&P 500 down 0.3% for the week, the Dow Jones Industrial Average down 1.3% and the Russell 2000 down 0.2%. Only the Nasdaq composite gained, by 0.5%. The AP described shares as near their record on Friday, so these are small moves at a high level and not a sell-off. They are also not a broad weekly gain.

Precious metals had a harder week. Dow Jones’s weekly tallies put front-month COMEX gold down US$154.10, or 3.59%, at US$4,133.70, and front-month silver down US$4.294, or 6.68%, at US$59.977. Those were the largest one-week falls since early June for gold and late June for silver, a second weekly decline in a row for both, and the lowest settlements since the first week of August. They are settlements of specific futures contracts. Reuters, quoting other instruments the same afternoon, had spot gold at US$4,140.06, down about 3.4% on the week, US gold futures settling at US$4,162.30 without naming the contract month, and spot silver at US$60.36. None of these is what a coin or bar cost at a dealer, where premiums and spreads sit on top.

Credit added a qualification. The ICE BofA US High Yield Index option-adjusted spread is the extra yield over Treasuries demanded on lower-rated corporate bonds, adjusted for the options embedded in them. It widened from 2.93 percentage points on Friday 25 September to 3.24 on Thursday 1 October, an increase of 31 basis points, and it rose on each of the four days.

Is the US economy weakening, or only hiring?

Hiring weakened; spending and output did not follow it down. The Bureau of Labor Statistics reported on Friday that nonfarm payrolls rose by 29,000 in September, a figure it described as having changed little. July and August were revised down by a combined 60,000, which turned July into a loss of 10,000 jobs. The unemployment rate was 4.2%, up from 4.1% in August and inside the 4.1%–4.3% range the Bureau says it has held since March. Average hourly earnings rose 0.1% on the month and 3.0% over the year.

Slow hiring is not the same condition as widespread job losses. Tuesday’s report on job openings showed layoffs and discharges little changed in August at about 1.6 million, while openings fell to 7.1 million from 7.3 million. Fewer vacancies and steady layoffs describe employers who are slow to add staff, not employers shedding them.

The counterweight was spending and output. Wednesday’s third estimate put second-quarter GDP growth at a 2.2% annualised rate, up from 1.5% in the previous estimate, in a release that also carried the Bureau of Economic Analysis’s annual update of the national accounts. Real final sales to private domestic purchasers, which is consumer spending plus private fixed investment, rose at a 4.6% rate. Real consumer spending then rose 0.6% in August alone. One caution sits in the same August release: real disposable income was flat and the saving rate was 4.1%, so the spending was not matched by income. And neither a second-quarter figure nor an August one says what happened to activity in September.

Our reading is that hiring has become the weak part of an uneven expansion. That can ease the pressure for an immediate rate increase without showing that a recession has begun; what gold has done in past downturns is set out in our review of the recession record. Nor does slower growth mean the same thing for every metal. Gold’s role is largely monetary, while industry took about 58% of silver demand in 2025, a difference we set out in our guide to the four precious metals.

The week, release by release

• 29 Sep A Fed voice, and two labour-market readings: New York Fed President John Williams says he sees “no urgency” to follow September’s increase with another, and traders pull back from October bets. US layoffs are little changed at about 1.6 million in August. Canada’s July GDP is essentially unchanged.
• 30 Sep Softer inflation, stronger growth, the week’s highest yields: August PCE prices are up 3.4% over the year and 3.0% excluding food and energy; second-quarter GDP is revised to 2.2% from 1.5%. Futures price about a one-in-three chance of an October increase. The 10-year yield closes at 5.29% and the 10-year real yield at 2.93%. China’s factory PMI edges up to 50.1.
• 1 Oct Factory costs widen: The ISM manufacturing index reads 54.5 and its Prices Index rises to 77.9 from 71.1. The high-yield spread reaches 3.24 percentage points.
• 2 Oct Weak payrolls, a rally in shares, yields higher on the day: Payrolls rise by 29,000 and the priced chance of an October increase is about 22%. Shares rise. The 2-year yield still closes 5 basis points up on the day at 4.83%. Euro-area inflation is estimated at 3.8%, and the G7 announces a 100-million-barrel release. Front-month gold settles at US$4,133.70.

Is inflation cooling?

On one measure, and not on others. The US personal consumption expenditures price index rose 0.3% in August and 3.4% over the year; excluding food and energy it rose 0.2% and 3.0%. Markets took that as softer than they had feared. Reuters reported that futures priced only about a one-in-three chance of an October increase that day.

Softer than feared is not the same as falling. The same release incorporated an annual update that revised the figures back to January 2021. On the revised figures, July’s monthly increases were 0.1% for both the headline and the core index, so August’s monthly increases were larger, and Reuters gave July’s annual rate as the same 3.4%. Setting August beside the July figures first published a month earlier, which were 0.2% on the month for both, mixes two vintages of data.

Factory surveys added a second caution. The Institute for Supply Management’s manufacturing index was 54.5 in September, still expanding and a tenth of a point below August, while its Prices Index rose from 71.1 to 77.9. These are survey indexes: 77.9 does not mean input prices rose 77.9%. A reading above 50 means more purchasing managers reported paying higher prices than lower ones. It measures how widespread the increases were, not how large.

So consumer inflation that was softer than feared sat alongside firm spending and wider cost pressure in factories. Neither “inflation is beaten” nor “inflation is accelerating everywhere” describes that.

Why did long-term yields rise when a rate rise became less likely?

Because the odds on one meeting are not the yield curve. The Federal Reserve raised its target range to 3.75%–4.00% on 16 September, as we set out in our review of that decision, and nothing this week changed the rate in force. What changed was the expected timing of the next move. The shift began on Tuesday with Williams’s remark, continued after Wednesday’s inflation figures and stood at about 22% on Friday. Reuters also reported on Wednesday that traders still priced an increase by December, so on that pricing the next increase had been moved later, not removed. All of these are market estimates at a moment in time. None is a commitment by the Federal Reserve.

These are consistent daily readings from the Treasury’s own curves, not intraday peaks, and Friday was not the week’s high: the 10-year closed at 5.29% and the 10-year real yield at 2.93% on Wednesday. The short end did respond to the change in expectations. The 2-year closed at 4.92% on Monday and 4.78% on Thursday. But on Friday, the day of the weak payrolls report, it closed 5 basis points higher at 4.83%, and the 10-year closed 4 higher at 5.28%. One basis point is 0.01 of a percentage point.

A long-term yield reflects the expected path of short-term rates over many years plus a term premium, which the Federal Reserve Bank of New York defines as the compensation investors require for bearing the risk that interest rates change over the life of the bond. A rise in long yields does not say how much came from each. The New York Fed adds that the term premium cannot be observed directly and has to be estimated. Our bond-market analysis sets out the pressures on long-dated government debt this year.

For gold, the real yield is the part that bears most directly. Inflation-linked Treasuries offered a higher quoted real return at the end of the week than at the start, and bullion pays no interest. That raised the return available elsewhere. It is a tendency, and it does not oblige gold to fall. The case against metals ownership in our Learning Academy works through that opportunity cost, including the periods when gold has disappointed.

The dollar drew the same line between a day and a week. Reuters reported that it edged lower on Friday but was headed for a weekly gain, and titled its Friday gold report “Gold heads for weekly drop as strong dollar, elevated Treasury yields weigh”. A softer Friday did not undo the week.

A related Voice: Brent Johnson on gold and the dollar

Brent Johnson, who runs Santiago Capital, is a useful counterpoint to treating gold and the dollar as permanent opposites. In an interview published on 16 March 2026 he restated the framework he calls the Dollar Milkshake: capital flowing into the United States can lift the dollar against other currencies while gold also rises, with heavy volatility along the way. He argued there that recognising the dollar’s relative strength is no argument against owning gold, and the reverse. Our profile’s one-line summary puts the two in sequence, the dollar first and gold afterwards. His own account, like the 2022 statement recorded on that profile, has them rising together.

Applied here, that separates acknowledging this week’s currency pressure from treating it as a rule. It does not explain why gold fell this week or say when that will change. It is a longer-horizon view, given more than six months before these releases, and it is not an input to the Macro Conditions Score. The interview was published by Monetary Metals, a company that sells yield-bearing gold products and on whose advisory board Johnson sits.

Energy and the international data kept the picture unsettled

On Friday the leaders of the G7, meeting by video link, announced “a coordinated release through the IEA of 100 million barrels” over four months, to begin immediately, with a substantial diesel release brought forward into the first 20 days. They undertook to coordinate maintenance schedules across G7 refineries so that capacity is not shut at the same time, and reaffirmed a commitment to refrain from export restrictions on energy between G7 countries.

This is an announced programme and not barrels already delivered, and it is not clearly new supply. The statement presents the release as carrying out the commitments of March 2026, when the International Energy Agency’s 32 members agreed to make 400 million barrels available, and says it takes into account what has already been fulfilled. It gives no figure for diesel.

For metals the effects could run in opposite directions. If relief in fuel markets arrives, it could ease inflation pressure and with it the case for further tightening. It could also reduce demand for protection against an energy shock. Neither channel settles the net effect on gold in advance, and both have to be observed.

Europe showed why the inflation concern has not gone away, and how concentrated it is. Eurostat’s flash estimate put euro-area annual inflation at 3.8% in September, up from 3.2% in August. Energy inflation was 18.8%, up from 14.3%, and services 3.2%, up from 3.0%. Excluding energy, food, alcohol and tobacco, the rate moved from 2.4% to 2.5%. The jump in the headline was an energy story far more than a broad one.

Growth evidence varied by country. Statistics Canada reported that real GDP was essentially unchanged in July, with an advance estimate of 0.2% growth in August. China’s official manufacturing PMI rose from 49.8 to 50.1 in September, just above the line between contraction and expansion, with large firms at 50.6 and medium-sized and small ones at 49.7 and 48.9. Together these describe uneven conditions, not a synchronised boom or collapse.

Did the Macro Conditions Score change?

Barely. The dashboard’s published Tactical Score was 35/100 for Friday 2 October, in the Cautious band, with its direction indicator at Deteriorating Rapidly. That is one point above the 34 of the previous Friday, 25 September, and three below the 38 of 16 September recorded in our September review. The Long-Term Structural Score, which updates weekly, read 64/100, Bullish, in its 28 September update, two points below the 66 of 14 September. The Macro Conditions Score reflects macro conditions historically associated with precious metals performance. It is a systematic framework, not investment advice.

A week in which front-month gold lost 3.59% and the real yield rose 9 basis points moved the total by one point, and the reason is in how the score is built. Headlines earn no points. Only measured inputs do, and each is scored in bands. The real-yield input earns 2 of its 20 points anywhere between 2.5% and 3.5%, so a rise inside that band costs nothing more. The move showed up in momentum, which was already at 3 of 30. The one point gained came from the confirmation basket, where the volatility and futures-positioning components each scored a point higher than a week earlier. A component near its floor cannot keep losing points, and a total that hardly moves is not a statement that nothing happened.

Input dates matter as much as bands. The dashboard is scheduled to run each weekday at 19:30 UTC, mid-afternoon in New York and before the Treasury publishes that day’s curves, and it reads real yields from FRED, which posts each day’s figure on the following business day. Friday’s run therefore scored a real yield of 2.93%, which is Wednesday’s reading; the Treasury’s figure for Friday itself was 2.92%. Our dashboard shows the date of the run beside each input, so that 2.93% appears under 2 October. When we retrieved FRED’s series on 3 October it had been updated on 2 October and ended on 1 October. A refresh date is not an observation date, on our dashboard or anyone else’s. The Learning Academy’s market context section gives the background to reading these measures together.

The table below is our reading of the week’s evidence. It is not how the score is calculated.

Earlier demand strength is context, not a price floor

The World Gold Council’s August report showed global gold ETF holdings rising by about 121 tonnes to a record 4,189 tonnes, as we set out in our review of those flows. That is evidence of substantial investment demand in August. It is not evidence that the same buying continued through this week, and September’s figures had not been published when this review was written.

Our article on China’s gold imports drew the same line between customs imports, official reserves, fund holdings and exchange withdrawals: they cover different activities and periods and cannot be added into one larger demand total. The discipline for a weekly review is to label older evidence as older. Strong purchases in one month can sit alongside selling pressure in the next.

Why hold metals when the short-term signals are difficult?

Because a strategic holding is sized for a purpose that one week’s data does not change. This site’s framework treats precious metals as a deliberately sized part of a diversified portfolio. That is a view about structure. It is not a promise that metals will outperform in any year, rise after any inflation release or protect against every loss.

The size, the mix of metals and the way they are held are the investor’s decisions. Our 5%, 10%, 15% and 20% illustrations, set out in our guide to building a core, are ways to compare outcomes, not targets. The level that fits depends on circumstances, time horizon, existing exposures, income needs, tolerance for risk and the purpose of the holding. The allocation section of our Learning Academy works through that decision. The percentage follows from the investor’s plan, not from a week’s score.

What the framework looks for is less dependence on a single market outcome. The World Gold Council’s strategic-asset research describes gold as “a clear complement to stocks and bonds”, with demand from investment, central-bank reserves, jewellery and technology. The Council is funded by gold-mining companies, and the same report concedes that gold provides no regular income and has had years of close to 30% gains and close to 30% losses. FINRA, the US securities industry’s self-regulator, gives the plain version for physical metal: prices can fluctuate and are sometimes quite volatile, metal an owner stores can be lost or stolen, and firms may charge commissions, storage, management and other fees.

A related Voice: Ray Dalio on preparing for different outcomes

The All Weather approach that Ray Dalio developed with colleagues at Bridgewater Associates addresses a different question: how to balance a portfolio against surprises in growth and inflation without depending on one forecast. Bridgewater’s January 2012 account of the strategy describes four environments, with inflation rising or falling and growth rising or falling, each measured against what markets expected, and a portfolio built to carry equal risk in each. The paper balances risk, not money, uses leverage to do it, and names no weight for gold.

The lesson we take from it is an all-weather portfolio, not an all-weather asset. It is not a suggestion to copy an institutional portfolio or to adopt a particular gold percentage; the 5–10% and 10–15% on our profile of Dalio are his own statements of 2017 and August 2026, not this paper’s. A role through different markets is not a positive return in every market.

What a metals holding can and cannot do

During monetary or currency stress, gold may give an exposure that behaves differently from conventional financial assets. In a growth scare, demand for protection and changing rate expectations may support it. But high real yields make interest-bearing assets more competitive, and gold itself can fall sharply. Silver’s industrial uses add a further source of cyclicality, and it should not be expected to give the same protection as gold. The record, crisis by crisis, is in the Learning Academy’s section on how metals behave in a crisis.

Even diversification that works does not remove losses. Take a purely hypothetical portfolio with 90% in other assets and 10% in gold. If the other assets fall 20% and gold gains 10%, the portfolio still loses 17% before costs. The gold cushions the loss. It does not prevent it.

The practical approach differs from one investor to the next. One might hold a target allocation and rebalance when it moves outside a chosen range. Another might build gradually through scheduled purchases, the approach we tested in our comparison of regular and lump-sum buying. A third might keep a strategic core and treat any additional, tactical exposure separately. Each needs a purpose and rules set before a volatile week, so that it does not depend on reading every release correctly.

None removes trade-offs. Rebalancing can trigger costs and taxes. Gradual buying leaves part of the intended exposure uninvested during a rally and does not make a falling asset safe. Tactical changes add the possibility of being wrong about both the exit and the re-entry. A strategic holding is still reviewed when circumstances or its rationale change: “core” does not mean untouchable.

The ownership decision matters as much as the percentage

Someone seeking personal control of metal weighs different risks from someone seeking convenient price exposure in a brokerage account. Physical possession, allocated custody, funds and exchange-traded receipts differ in ownership rights, counterparty dependencies, risk and cost, not only in ease of trading. The Learning Academy’s section on physical versus paper introduces those alternatives.

Our article on the Royal Canadian Mint’s gold receipts is a worked example: legal ownership, segregation of the metal, permission for the custodian to use it and the terms of redemption are separate questions, and a reassuring label answers none of them. Our article on bullion lending shows what changes when metal is lent.

A holding can serve an ownership purpose during a price decline. Bullion held personally does not depend on a fund’s redemption process, but it leaves the owner responsible for secure storage, insurance, proof of authenticity and eventual resale. Removing one dependency does not remove investment risk.

Mining shares, royalty and streaming companies and exploration stocks bring business and equity-market risks of their own. In our framework they belong to whatever growth or speculative portion an investor chooses, not to the strategic bullion core.

For Canadian readers the currency matters too. Before costs, a US-dollar gold price multiplied by the number of Canadian dollars per US dollar gives its Canadian-dollar price, so a move in the exchange rate can cushion or amplify a move in the metal. The US-dollar change for the week is not automatically the change in a Canadian investor’s holding.

Whatever the approach, keep the purchase history. Our cost-basis guide explains why acquisition dates, quantities and full purchase costs belong in the record before a sale makes them urgent.

The Bigger Picture

The week did not show that the case for gold has failed, and it did not show that a weak jobs report guarantees a rebound. It showed three pairs of things that are easy to confuse: an expected policy decision and an actual yield, a Friday rally and a weekly return, and this week’s evidence and last month’s demand statistics.

The observations that would clarify the picture are whether real yields and the dollar ease together, whether credit spreads stop widening, whether softer hiring spreads to spending, and whether the announced energy release shows up in delivered supply and prices. The Federal Reserve next meets on 27–28 October. New fund-flow and official-reserve figures will update the demand side, so that August no longer has to stand in for October.

The Macro Conditions Score describes the environment. A portfolio plan decides how much exposure belongs in it. Our part is to set out the options, the strategies and the trade-offs so that the decision stays with the investor.

An expectation about one meeting is not a yield. A Friday is not a week. Last month’s demand is not this week’s price.

Sources and Methodology

This review covers releases and market moves from 28 September to 2 October 2026, and every figure keeps its own reference period. Share-index changes are the Associated Press’s closing table for 2 October. Metal figures are Dow Jones Newswires’ weekly tallies of the front-month COMEX contracts, both for October delivery, which Dow Jones calculates from settlement prices; the spot quotations and the second futures price are from Reuters’ gold report of the same day and are not interchangeable with them. Treasury yields are the Treasury’s daily par nominal and par real curves, and changes in basis points are our arithmetic. The high-yield spread is the ICE BofA index as carried by FRED, which ran to 1 October when retrieved on 3 October. Rate-pricing figures are Reuters’ reports of the CME FedWatch Tool on 30 September and 2 October and are approximate by their own wording. Payrolls, unemployment, revisions and earnings are the Bureau of Labor Statistics’ Employment Situation for September; layoffs and openings are its Job Openings and Labor Turnover release for August. GDP and consumer-price figures are the Bureau of Economic Analysis’s releases of 30 September, which incorporated its annual update; July’s first-published monthly rates are from its release of 26 August. The manufacturing indexes are the Institute for Supply Management’s September report. The G7 wording is the leaders’ statement published by the French presidency, and the March commitment is the International Energy Agency’s announcement of 11 March 2026. Euro-area figures are Eurostat’s flash estimate and may be revised; the Canadian August figure is advance information; the Chinese figures are the National Bureau of Statistics’ release, read in Chinese. The share of silver demand taken by industry is the figure in our guide to the four metals. The hypothetical portfolio is our own arithmetic. We do not attribute the week’s price moves to any single release.

All score readings are our own. The Tactical Macro Conditions Score of 35/100 for 2 October 2026, its four blocks and its component values, and the readings of 34 for 25 September and 38 for 16 September, are the dashboard’s published Engine V2 snapshots, read from its record on 3 October 2026. Nothing was recalculated for this article. The Long-Term Structural Score of 64/100 is the 28 September update and the 66 is the 14 September update cited in our September review; the two-point difference is the dollar-index component, which fell from 4 to 2 of its 8 points as the index it reads rose from 99.6 to 101.2. Because a run uses the latest observation FRED has posted at that moment and not that day’s close, the 2 October run used 2.93% for the real yield, the Treasury’s reading for 30 September. The run of Monday 28 September read 39 because three FRED inputs could not be fetched and fell back to stored values, which the dashboard flags as stale; we do not use that reading as a comparison. The Macro Conditions Score reflects macro conditions historically associated with precious metals performance. It is a systematic framework, not investment advice.

The two Voices passages are paraphrases, not quotations, and neither is a comment on this week’s releases. Brent Johnson’s is drawn from an interview published by Monetary Metals on 16 March 2026, read in its machine-generated transcript; the passage on the dollar and gold rising together is his restatement of a framework he first presented publicly in 2018. The All Weather passage rests on Bridgewater Associates’ own account of the strategy, dated January 2012, which carries no byline, credits Dalio with Bob Prince, Greg Jensen, Dan Bernstein and others, and is not a current allocation disclosure. Its framework is on pages 6 and 9. It names no gold weight, and the 7.5–12% range that earlier articles on this site attach to All Weather does not appear in it. Applying either view to this week is our interpretation, and neither person has endorsed GoldSilverPortfolio, its score or its allocation illustrations.

Conflicts of interest, stated plainly. The World Gold Council is a membership organisation of gold-mining companies, and its research is that of an interested party. The Silver Institute is an industry association. Monetary Metals sells yield-bearing gold products and has Brent Johnson on its advisory board, so the interview is a commercial publisher hosting one of its own advisers. Bridgewater’s paper describes its own product. Two Reuters reports were read on the website of Kitco, a bullion dealer and news publisher with which GoldSilverPortfolio has no affiliate or commission arrangement. GoldSilverPortfolio 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 paid record-keeping tools. No dealer, fund, broker or product is recommended here, and the Macro Conditions Score is our own framework, not a signal to buy or sell anything.

Primary sources

• US Bureau of Labor Statistics — “The Employment Situation — September 2026” (USDL-26-1549), released 2 October 2026: nonfarm payrolls up 29,000 and the unemployment rate at 4.2%, both described as having changed little; 4.1% in July and August; July payrolls revised from +21,000 to −10,000 and August from +162,000 to +133,000; the unemployment rate in a range of 4.1% to 4.3% since March; average hourly earnings up 5 cents, or 0.1%, to US$37.81 and 3.0% over 12 months
• US Bureau of Labor Statistics — “Job Openings and Labor Turnover — August 2026” (USDL-26-1547), released 29 September 2026: layoffs and discharges little changed at 1.6 million; job openings at 7.1 million, from 7.3 million in July
• US Bureau of Economic Analysis — “GDP (Third Estimate), Industries, Corporate Profits, State GDP, and State Personal Income, 2nd Quarter 2026; State PCE, 2025”, 30 September 2026: real GDP up at a 2.2% annual rate, revised from 1.5%; real final sales to private domestic purchasers up 4.6%; results of the annual update covering the first quarter of 2021 to the first quarter of 2026
• US Bureau of Economic Analysis — “Personal Income and Outlays, August 2026”, 30 September 2026: the PCE price index up 0.3% on the month and 3.4% over the year, and 0.2% and 3.0% excluding food and energy; July’s monthly increases 0.1% for both, as revised; real PCE up 0.6%; real disposable personal income unchanged; a saving rate of 4.1%; revisions beginning with January 2021. July’s first-published monthly rates are from “Personal Income and Outlays, July 2026”, 26 August 2026
• Institute for Supply Management — “Manufacturing PMI at 54.5%; September 2026 ISM Manufacturing PMI Report”, 1 October 2026: Manufacturing PMI 54.5, from 54.6 in August; Prices Index 77.9, from 71.1
• US Department of the Treasury — Daily Treasury Par Yield Curve Rates and Daily Treasury Par Real Yield Curve Rates, 2026: the 2-year, 10-year and 30-year nominal yields and the 10-year real yield on each business day from 25 September to 2 October 2026, and the 10-year real yield on 4, 11 and 18 September
• Federal Reserve Bank of St. Louis, FRED — ICE BofA US High Yield Index Option-Adjusted Spread (BAMLH0A0HYM2), data from ICE Data Indices, retrieved 3 October 2026: 2.93 on 25 September, then 3.02, 3.08 and 3.12, and 3.24 on 1 October, the latest observation; and the 10-year inflation-indexed Treasury yield (DFII10), updated 2 October with observations to 1 October
• Dow Jones Newswires, carried by Morningstar — “Comex Gold Ends the Week 3.59% Lower at $4133.70 — Data Talk” and “Comex Silver Ends the Week 6.68% Lower at $59.977 — Data Talk”, 2 October 2026: front-month contracts for October delivery; the weekly and daily changes and the comparisons with June and August, calculated on settlement prices
• The Associated Press — “How major US stock indexes fared Friday 10/2/2026”, 2 October 2026: closing levels and the daily and weekly changes of the S&P 500, the Dow Jones Industrial Average, the Nasdaq composite and the Russell 2000
• Reuters, carried by Kitco News — “Gold heads for weekly drop as strong dollar, elevated Treasury yields weigh”, 2 October 2026: about a 22% chance of an increase this month against around 70% earlier in the week, on the CME FedWatch Tool; spot gold at US$4,140.06 at 2:33 pm Eastern, down about 3.4% on the week; US gold futures settled at US$4,162.30; spot silver at US$60.36; the dollar lower on the day and headed for a weekly gain
• Reuters, carried by Kitco News — “Fed seen on pause in October after inflation data”, 30 September 2026: about a one-in-three chance of an October increase, with an increase by December still priced; John Williams’s “no urgency” remark; the PCE price index up 3.4% over 12 months in August and, as revised, in July
• Board of Governors of the Federal Reserve System — FOMC statement of 16 September 2026 and the 2026 meeting calendar: the target range raised to 3-3/4 to 4 percent; the next meeting on 27–28 October 2026
• G7 — “G7 Leaders’ statement on global energy security and market stability”, published by the French presidency, 2 October 2026: the 100-million-barrel release through the IEA over four months, the diesel release in the first 20 days, refinery maintenance schedules, export restrictions, and the March 2026 commitments
• International Energy Agency — “IEA Member countries to carry out largest ever oil stock release amid market disruptions from Middle East conflict”, 11 March 2026: 32 member countries agree to make 400 million barrels of emergency reserves available
• Eurostat — “Euro area annual inflation up to 3.8%”, flash estimate for September 2026, 2 October 2026: all items 3.8%, from 3.2% in August; energy 18.8%, from 14.3%; services 3.2%, from 3.0%; all items excluding energy, food, alcohol and tobacco 2.5%, from 2.4%
• Statistics Canada — “Gross domestic product by industry, July 2026”, The Daily, 29 September 2026: real GDP essentially unchanged in July; advance information indicating a 0.2% increase in August
• National Bureau of Statistics of China — Purchasing Managers’ Index for September 2026 (in Chinese), 30 September 2026: manufacturing PMI 50.1, up 0.3 points on August; large, medium-sized and small enterprises at 50.6, 49.7 and 48.9
• World Gold Council — “Gold as a strategic asset: 2026 edition”, 4 February 2026, including its section “Potential risks and challenges”; and “Gold ETF Flows: August 2026”, 9 September 2026: holdings up 121 tonnes to a record 4,189 tonnes
• FINRA — “4 Tips to Know Before Buying Physical Precious Metals”, 24 March 2026: price volatility, the risk of loss or theft for metal stored by its owner, and the fees firms may charge
• The Silver Institute — “Silver in Industry”, an undated reference page read on 3 October 2026: silver’s industrial applications
• Federal Reserve Bank of New York — “Treasury Term Premia”, an undated data page read on 3 October 2026: the two components of a Treasury yield, the definition of the term premium and the note that it must be estimated
• Monetary Metals — “Brent Johnson: The dollar milkshake is unfolding as expected”, The Gold Exchange Podcast, 16 March 2026: the interview transcript
• Bridgewater Associates — “The All Weather Story”, January 2012: the four environments relative to expectations (pages 6 and 9), the equal-risk design (page 7) and the people credited (page 2)
• GoldSilverPortfolio — Macro Conditions Score dashboard, Engine V2: the published tactical readings for 16 September, 25 September and 2 October 2026 with their blocks and component signals, and the Long-Term Structural Score updates of 14 and 28 September 2026, as recorded on 3 October 2026