The trigger was an American inflation report. Consumer prices rose 3.7% over the year to September against 3.6% expected, and a day later producer prices jumped 1.3% in a single month, the biggest rise since March 2022. Both made another rate rise from the Federal Reserve more likely, and higher rates raise the interest bill of every company that lives on borrowed money.

The selling spread fast. The Nasdaq Composite lost 4.5% on the week, and Korea’s main index, which is mostly two memory-chip makers, lost 14.5%.

America’s government borrowing costs fell anyway. Investors frightened by falling stocks bought Treasuries, and the ten-year rate dropped 15 basis points to 5.09%. Oil went the other way and closed Friday at $109.86, its highest since May.

But now let’s grab a coffee and go through it properly.

The Scoreboard

Nineteen numbers, and the bottom half of the Asia block is the place to look first.

America had a bad week for stocks and a good one for bonds. The S&P 500 fell 1.9% to 7,659.44 and closed Friday at its low for the week. The Nasdaq Composite lost 4.5% to 26,140.03, its worst week since June, and smaller companies in the Russell 2000 dropped 3.2%.

Government borrowing went the opposite way. A basis point is one hundredth of a percentage point, and the thirty-year rate fell 17 of them to 5.43%, the ten-year 15 to 5.09%. The two-year, which mostly tracks what traders expect the Fed to do next, fell only 6 to 4.74%. Chapter two is about why those two numbers moved so differently.

Europe was the calm corner of the world. The Euro Stoxx 50 even rose 0.5% to 6,204.23. German ten-year borrowing slipped about 5 basis points to around 3.42%, and Italy’s extra cost over Germany narrowed 3 to 106.

Asia took the damage. The Nikkei 225 lost 5.3% to 65,324, and the KOSPI 14.5% to 5,668.05. The yen barely moved, still near 158 per dollar.

Brent crude rose 5.2% to $109.86 as attacks on shipping near the Strait of Hormuz continued. Gold slipped 0.7% to $4,166.39, and bitcoin 0.8% to about $81,500.

The largest circle belongs to Korea, almost three times the size of the next one. The only big green circle is oil. Nearly everything else that fell was a stock tied in some way to the AI build-out, and every number in the bond strip is red, which for bonds means borrowing got cheaper.

1. CoreWeave’s lenders asked for 16%

A company that wants to grow fast with little cash of its own borrows. CoreWeave did that on a large scale. It buys Nvidia chips, puts them in data centres and rents the computing power to AI labs and big tech firms. Its debt and similar obligations come to more than $51 billion.

For a borrower like that, two numbers decide almost everything: what customers pay, and what lenders charge. This week the second one moved. Wednesday’s inflation report raised the odds of another Fed rate rise, and every rise makes refinancing dearer. Lenders began asking whether CoreWeave can carry its interest bill out of the cash it actually earns.

Their answer showed up in prices. The yield on CoreWeave’s 2031 bonds, meaning the yearly return a buyer locks in at today’s price, passed 16% on Thursday. Insurance against the company defaulting, which trades as a credit default swap, rose to a record of more than 1,000 basis points. Put simply, protecting $1 million of its debt now costs over $100,000 a year.

The worry travelled. Oracle, also borrowing heavily to build data centres, fell 8.8% on Wednesday, and the yield on its 2032 bonds rose to 7.24%. People who own the stock stand behind lenders in the queue and only get paid once bondholders have been, so the stock tends to move first and furthest when lenders grow nervous.

The other side deserves a fair hearing. CoreWeave’s chief executive told staff that signed customer contracts more than cover the company’s debt. Jensen Huang, who runs Nvidia, said the turmoil will pass. And most of the build-out is paid for by companies with enormous earnings of their own, even if they too have sold well over $200 billion of bonds this year.

My read: lenders now price AI borrowers one at a time, and those without steady cash coming in pay the most. The sign of real spreading would be the bond yields of the biggest technology companies climbing alongside CoreWeave’s. That has not happened yet.

2. Hot inflation, cheaper borrowing

Inflation reports usually push government borrowing costs up. This week both arrived in the same 48 hours, and borrowing costs came down.

September’s consumer prices rose 0.6% in the month. The core measure, which leaves out food and energy, rose 0.3% against 0.2% expected. On Thursday came producer prices, which track what businesses get paid for their goods before those goods reach a shop. They jumped 1.3%. Goods alone rose 2.6%, the largest monthly increase since the series began in 2009, with petrol up 20.3%.

Two forces pulled on bond prices at once. Higher inflation makes a Fed rate rise more likely, and that pushes yields up. Falling stock prices send frightened money into government bonds, and that pushes yields down. On Thursday the second force won clearly: the ten-year yield fell more than 10 basis points to 5.09% while traders were still reading the producer price report.

The two-year shows the first force never left. It fell only 6 basis points all week. Futures traders briefly lifted the chance of a rate rise at the 27–28 October meeting from about 22% to 47% after the CPI, then let it drift back to roughly a third. A rise by December is priced at more than 90%.

So the gap between two-year and ten-year borrowing narrowed from 44 basis points to 35. That shape says investors expect the Fed to keep money tight now and growth to soften later. Last week the bond market was worried about government debt. This week it was worried about growth.

Hold that loosely, though. Most of the move came on a single day of fear, and by Friday the ten-year was already creeping back up.

3. Korea lost a seventh of its stock market in a week

The KOSPI closed at 6,625.93 on 8 October, before Seoul shut for a holiday. Five sessions later it stood at 5,668.05. Four of those five days were losses.

Thursday was the worst. The index fell 7.57% to 5,554.22, and trading was paused twice by automatic brakes that kick in when prices fall too fast. Foreign investors sold 3.6 trillion won of Korean stocks that day. Samsung Electronics lost 9.5% and SK hynix 11.1%.

The reason is concentration. Samsung and SK hynix make the memory chips that AI data centres buy, and together they carry roughly half of the index. Once lenders started questioning the companies buying those chips, investors started questioning how long the orders keep coming. Foreign funds own a big slice of both firms and can leave quickly, and they had been net sellers for thirteen sessions in a row before Friday.

Tokyo felt it too. The Nikkei fell 3,318 yen on Thursday, the third-largest point drop in its history, and the Japanese press called it the CoreWeave shock.

The best argument against panic is that nothing inside these companies changed. Samsung reported record results the week before, and on Friday the KOSPI rose 2.05% as foreign investors turned buyers again.

Two things went genuinely well. Europe’s main index rose 0.5% while the rest of the world sold. And Britain’s economy grew 0.1% in August, when economists had expected it to shrink.

Data Check

Six releases mattered, and all six came in above forecast.

The producer price row is the one to read twice. Prices that businesses charge each other usually reach shoppers a few months later, so a 1.3% jump in September points to more inflation in October and November consumer prices. Petrol explains much of it. Even with food and energy left out, though, producer prices still rose 0.7%.

The retail sales row is weaker than it looks. Total sales rose 0.8%. The narrower measure that leaves out cars, petrol, building materials and restaurants is the part closest to what households choose to spend, and it rose only 0.1% against 0.3% expected. Americans spent more mainly because petrol cost more.

China’s factory-gate prices rose 5.0% over the year, the most since mid-2022. For years China sold the world falling prices. It is now selling it rising ones.

The Credit Grind

Chapter one was about one borrower. This section checks whether the rest of the corporate bond market caught the same cold.

Anyone lending to a company is paid for two things. One is the price of money itself, which governments set. The other is compensation for the chance this particular borrower stops paying. The company rows measure that second part, in basis points on top of the government rate.

The riskiest European companies now pay 336 basis points on top, 10 more than a week ago. The last column reads 99%, which means hardly a day in the past twelve months was dearer. The year’s high is 337, a single basis point away. America’s junk-rated companies, those rated least likely to repay, rose 8 to 323. Large, solid American companies barely moved, up 2 to 84.

So the scare spread, but only to the bottom of the ladder. The weakest borrowers got dearer on both continents, and the strong ones were left almost alone. That fits chapter one: lenders sorting borrowers by how much cash they actually bring in.

The government rows explain why it hurt less than it sounds. Euro-area ten-year borrowing fell 13 basis points to 3.39%, the two-year 4. For a European junk borrower, the falling base rate and the rising premium nearly cancelled each other out. The total bill stayed roughly where it was, and the relief from cheaper government money went to the lender as extra compensation.

The gap between the two junk markets widened to 13 basis points, with Europe the dearer one for a second week. France’s budget trouble sits behind much of that, as it did last week.

So what does this mean for anyone holding a corporate bond fund? You are now paid close to the most all year for lending to weak companies. That extra pay arrived because lenders grew more worried, and the bonds you already held lost a little value on the way.

The Onchain Pour

Bitcoin held up better than the stocks tied to it. It finished near $81,500, down 0.8% on the week. On Thursday morning, right after the producer price report and Korea’s worst day, it fell below $80,000 for about two hours before buyers stepped in near $79,800 for the third time in a week.

The companies built around bitcoin had a far rougher Friday. Coinbase fell 5.8% and Strategy 9.6% as the Nasdaq slid. Ether lost 2.7% to about $2,420.

The funds stayed cautious. US spot bitcoin funds, which hold actual coins and trade like stocks on an exchange, saw roughly $229 million leave over the week. Ether funds lost about $320 million.

October has been unusually poor. Bitcoin opened the month at $83,576 and sits a couple of per cent below it. It has not finished two Octobers in a row lower since at least 2013, and last October was one of the red ones, so a weak finish this month would break that pattern.

The Week Ahead

The Fed meets on 27 and 28 October, and from this weekend its officials stop speaking in public about policy. That leaves the data to speak for them, and there is no American inflation report this week. The question running through the week is whether this one’s two shocks keep going: lenders backing away from AI borrowers, and oil above $100 feeding prices everywhere.

Britain and Japan publish inflation figures, and both central banks are under pressure to raise rates into an oil shock. The flash business surveys on Friday carry a measure of what companies pay for their inputs, which is the earliest place an oil price shows up.

Korea’s central bank meets on Thursday, a week after the KOSPI’s worst day in months. Raising rates into a falling market is unusual, and some forecasts expect exactly that.

China GDP, third quarter — Monday, 04:00.
Consensus 4.4% after 4.3%. Factory output and retail sales land at the same time, the latter barely growing at 0.4%.

Canadian inflation — Monday, 14:30.
Previous 3.0%. A first read on how quickly North American fuel costs reach the shops.

German ZEW survey — Tuesday, 11:00.
Consensus 31.6 after 34.7. It asks financial analysts how they expect the German economy to do over the next six months.

UK inflation — Wednesday, 08:00.
Consensus 3.4% after 3.1%. A jump of that size would put a Bank of England rate rise firmly back in play. Christine Lagarde speaks at 18:00.

Bank of Korea decision — Thursday, 03:00.
The rate sits at 3.00%. Trading Economics expects a rise to 3.25%.

US jobless claims — Thursday, 14:30.
Consensus 200,000. New claims for unemployment support have stayed near their lows all autumn.

Japanese inflation — Friday, 01:30.
Core consensus 2.1% after 1.7%. The Bank of Japan watches this number closely before each rise.

Flash business surveys — Friday, euro area 10:00, US 15:45.
Euro area services consensus 52.8 after 53.0, US services 57.9 after 58.8. Anything above 50 means growth.

University of Michigan sentiment, final — Friday, 16:00.
Consensus 46.3, unchanged from the flash and close to the weakest readings in the survey’s history.

☕ Last Sip

This was the week lenders asked the AI build-out a direct question: can the borrowers pay? For most of the industry the answer is plainly yes. For the companies that borrowed most and earn least, the price of that answer went up sharply, and markets as far away as Seoul paid for the asking.

The bond market’s reaction is the part to keep. Inflation came in hot and borrowing still got cheaper, because fear briefly mattered more than prices did. Fear fades faster than oil at $110 does, and the Fed meets in ten days.

Enjoy the autumn weekend. See you next Sunday.

Sources

Market levels from CNBC, Yahoo Finance, Trading Economics and the US Treasury; ANSA for the Italian–German spread; Chunichi Shimbun and Kyodo for the Nikkei, the yen and Japanese yields; Asia Economy and Money Today for the KOSPI. Economic releases from the US Bureau of Labor Statistics, the Census Bureau, the National Bureau of Statistics of China and the ONS. Credit and euro curve data from ICE BofA indices via FRED and the ECB Data Portal, as of 15 October, own pipeline. CoreWeave and Oracle from CNBC and Reuters. Weekly changes are measured against the levels published in edition #12. Analysis and personal opinion — never investment advice.

On Friday, lenders charged France about 4.79% a year to hand over money for ten years. Italy got the same ten years for 4.57%. Lenders now treat France as the riskier bet.

The corporate side is stranger still. About 38% of French investment-grade company bonds, which is roughly €215 billion worth, trade at lower yields than French government debt of the same maturity. In January that figure was €12 billion.

America had a quieter week on the surface. The S&P 500 added 1.1% and set another record on Tuesday. Underneath it, the average company in that index is trading about 16% below its own one-year high, and Goldman’s measure of how many companies are actually taking part is back near where it last sat in 2000.

The Federal Reserve’s September minutes arrived on Wednesday and said most of the committee still expects one more rate rise before the year is out. Korea took that badly and lost 5.4% in four sessions. Crypto took it worse, with about $1.2 billion of positions force-closed on Thursday, nine out of ten of them bets on higher prices.

But now let’s grab yourself a coffee and dive into it properly.

The Scoreboard

Nineteen numbers, and the American rows are the calmest thing on the board. Everything that moved this week moved somewhere else.

Watch the Asia block and the Italy–Bund row. Those two carry the information.

America spent the week doing very little twice over. The S&P 500 closed at 7,811.51, up 1.1%, within a whisker of Tuesday’s record of 7,818.93. The Nasdaq Composite added 0.6% to 27,366.17. Smaller companies in the Russell 2000 went the other way again, down 0.9% to 2,808.57, which is the eighth week out of eleven that they have lagged the big index.

Government borrowing costs came down three or four basis points across the whole American curve — a basis point is one hundredth of a percentage point — leaving the ten-year at 5.24% and the thirty-year at 5.60%. That is the bond market declining to react to minutes that told it another rate rise is coming. Last week’s panic did not continue; it also did not reverse.

Europe is where the week happened, and it happened to France rather than Italy for the first time. German ten-year borrowing barely moved, up a single basis point to 3.47%. Italy’s extra cost over Germany actually fell 6 basis points to 109, the narrowest it has been in a month. France, which does not sit on this board, pays over 130 basis points above Germany and came within five basis points of a round 5% on Thursday. The Euro Stoxx 50 lost 1.1% to 6,176.15, with banks doing most of the damage.

Asia split down the middle. The Nikkei 225 added 1.0% to 68,994, a fourth winning week, while Japanese ten-year borrowing fell 9 basis points to 3.02%. Korea had the worst week of any major market on the board. The KOSPI dropped 5.4% to 6,625.93 in four trading sessions — Seoul was shut on Friday for Hangul Day — with foreign investors selling about 2 trillion won of Korean shares on Thursday alone and domestic institutions another 1.7 trillion. Samsung and SK hynix both fell more than 2% that day. The reason given was oil feeding inflation and American rates staying high, which is a long chain of causation for the best-performing market in Asia this year.

Oil is the number doing the quiet work in this whole letter. Brent closed at $104.43, up 2.1% on the week and up about 66% over twelve months. Gold rose 1.3% to $4,193.87, recovering part of last week’s fall. The dollar index barely moved at 102.21.

The two biggest moves on the board are the two assets people most often buy with borrowed money. Ether fell 7.9% to $2,488 and bitcoin 4.5% to $82,123. That pattern — the leveraged end of the market taking the hit while the index sits at a record — repeats in every chapter below.

1. France is now Europe’s expensive borrower

Size of debt is not what changed. Italy owes about 138.5% of its annual economic output; France owes about 118.1%. What investors are pricing is direction. Italy’s deficit — the gap between what the state spends in a year and what it collects — is 2.9% of output, and the European Commission expects it to stay there through 2027. France’s is 5.1% and is forecast to widen to 5.7%. Interest payments alone are set to climb from 2.6% to 2.8% of French output over the same two years.

Add a parliament that cannot agree a budget and a presidential election in 2027 with no obvious outcome, and lenders have decided that French promises are the ones most likely to be rewritten. Italy, for all its debt, has had the same prime minister and the same fiscal line for four years.

The strange part is in the corporate market. Air Liquide sold €2 billion of bonds on Tuesday, took €12.5 billion of orders, and priced both of its fixed-rate tranches below the French government curve. Bloomberg counts €215 billion of French high-grade corporate debt now trading through the state, against €12 billion in January — an eighteen-fold rise in nine months.

A lender doing that is saying that L’Oréal’s cash from Asia and America looks more dependable over ten years than the French Treasury’s ability to collect taxes. Where a company earns its revenue matters more than where its head office is registered. French banks are the exception, because their balance sheets are full of French government bonds, and insuring their debt now costs more than insuring other European lenders’.

Keep it in proportion. France has no funding problem: the auctions clear, the buyers turn up, and 4.8% for ten years when America pays 5.24% is an expensive price rather than a distressed one. Spreads like this have lasted years elsewhere without anything breaking. The thing worth watching is not the yield but whether that €215 billion keeps growing.

2. Sixty-five billion dollars of loans nobody wants

A leveraged loan is money lent to a company that already carries a lot of debt, then sliced up and sold on to funds. JPMorgan counted the American ones trading below 60 cents on the dollar, which is the market’s way of saying it expects to get back considerably less than it lent. The total is $65 billion, up from $40 billion a year ago and the highest since March 2020.

Widen the definition to loans at or below 80 cents and the pile is $139.8 billion, nearly double twelve months ago and about $4 billion short of the peak reached in May 2020. The number of borrowers in that category has gone from 106 to 141.

The mechanism is plain. These loans pay a floating rate: a base that tracks what the central bank does, plus a fixed margin on top. A company that borrowed in 2021 when the base was near zero now pays a base near 4%, and Wednesday’s minutes said another rise is likely. Nothing about the underlying business has to get worse for the interest bill to double.

Technology is 39% of the distressed pile, or $54.4 billion, and software companies face more than $100 billion of debt coming due. That is the same sector the stock market is paying record multiples for. CCC-rated loans, the lowest rung of junk, are down 1.97% this year while every other junk category has made money, which is the market sorting borrowers one by one rather than selling the asset class.

Jamie Dimon spent the week warning that corporate credit will get dearer. JPMorgan’s own forecast has loan defaults rising from 2.25% this year to 4.50% next.

Against that: the troubled share of the American loan market is still around 5%, most of these companies are privately owned by funds with money to inject, and American corporate borrowing actually got cheaper this week. This is a slow problem concentrated in one sector.

3. A record index in an unhappy country

Tuesday’s S&P 500 close of 7,818.93 was an all-time high. The median stock inside that index is about 16% below its own one-year high. Goldman Sachs says its breadth indicator — how many shares are rising together — is close to the lowest since the dot-com peak.

Across the wider Russell 3000, 51% of companies are down more than 20% from their June highs, against just over 40% two weeks earlier. Semiconductors are worst at 96% of the sector in that state, software next at 75%. The ten largest companies in the S&P 500 now account for about 40% of its market value and 37% of its expected earnings. At the 2000 peak the top ten were 27% of value and 15% of earnings — today’s concentration is bigger, and the earnings behind it are real.

Hedge funds have been piling into exactly that trade. Goldman’s prime brokerage desk puts their net exposure to the seven big technology names at a record 22% of their American holdings as of 5 October, up from 15% in July.

Away from the screens the mood is worse than the index suggests. The University of Michigan’s first October reading of consumer sentiment came in at 46.3, the second-lowest in the survey’s history. Its measure of how households see conditions right now fell to 44.7, the worst ever recorded. Joanne Hsu, who runs the survey, put it down to prices and borrowing costs making big purchases feel out of reach, and called households’ expectations “largely stagflationary”.

The Federal Reserve’s three-yearly Survey of Consumer Finances, published Friday, says the same thing with harder numbers. Nearly 20% of American families reported falling behind on a loan payment in the three years to 2025, up from about 12% in the previous survey and the highest since 2010. More than 8% were two months late or worse. The share spending over 40% of their income on debt payments reached 8.6%, the most since 2013. Real median family income rose 7% across the same period.

Two things went genuinely well. Nobody is being laid off: initial jobless claims fell to 197,000, a fifth straight weekly decline and the lowest since July. And American companies had a good week in the debt market, with high-yield borrowing costs falling nine basis points and investment-grade four.

Data Check

Seven releases mattered, and five of them came in short of what economists expected.

The German row needs disarming before anyone panics. Factory orders fell 10.6% in a single month, the sharpest drop since January, and almost all of it came from one line: orders for aircraft, ships and trains collapsed 61.5% after more than doubling in July on a handful of enormous contracts. Strip out the large orders and German manufacturing was down 0.1%. Nothing happened.

The American service survey is the row that matters. The headline slipped to 54.9, still comfortably in growth, but the sub-index tracking what service companies pay for their inputs rose to 74.0, the highest since 2022. Oil is the obvious culprit. Energy costs reach manufacturers and hauliers first and service firms a month or three later, and service prices tend to stay where they land, because they sit inside wage agreements and leases and fee schedules that reprice slowly.

That single number explains why the Fed minutes read as they did, why Michigan’s one-year inflation expectation climbed to 4.7%, and why the market still gives December a high chance of another rate rise.

The Credit Grind

For the first time since this letter started tracking it, European junk borrowers pay more than American ones. The gap between the two high-yield indices flipped from −7 basis points a week ago to +11 now, as Europe widened nine and America narrowed nine.

Anyone lending to a company is paid for two things. One is the price of money itself, which governments set. The other is compensation for the chance this particular borrower stops paying, and that second part is what the company rows measure, in basis points on top of the government rate.

Europe’s high-yield borrowers now pay 326 basis points on top, which is 3.26 percentage points and sits at the 97th percentile of the past year — higher than on all but a handful of days since last October. America’s pay 315, and that sits at the 90th. The quietest number on the board is American investment grade at 82 basis points, back down at the 75th percentile after last week’s jump, which is to say large solid American companies are borrowing at ordinary prices again.

The flip has a tidy explanation. France is Europe’s second-largest corporate bond market, and when the French state’s own borrowing cost rises and its budget stalls, European credit gets marked down alongside it. America has the reverse: an expensive government and companies that looked relatively good against it all week.

The government rows underneath moved together for once. Euro-area two-year and ten-year borrowing both fell seven basis points, leaving the gap between them at 49 basis points, unchanged on the week. A curve that keeps its shape while everything above it is repricing tells you the European Central Bank is not the story here. What companies pay on top of it is.

The Onchain Pour

Friday was the first anniversary of 10 October 2025, the day a record $19 billion of crypto positions were force-closed in a single session. The market marked it in character. About $1.2 billion of positions were liquidated in the twenty-four hours to Thursday evening, roughly 90% of them bets that prices would rise, with 191,736 accounts caught.

Ether took the heavier beating. Measured against the size of each market, ether positions were wiped out at roughly six times bitcoin’s rate — about $1.2 million of liquidations per billion dollars of market value, against $180,000 for bitcoin. Ether ended the week at $2,488, down 7.9%. Bitcoin closed at $82,123, down 4.5% after touching $80,400 on Thursday.

The useful part is what did not happen. CoinDesk Research measured how much money sits in the order book ready to absorb a sale and found bitcoin’s depth within 1% of the price at about $11.7 million, roughly 75% above where it stood on crash day — and that with bitcoin itself about a third cheaper, so those dollars represent considerably more committed capital. Ether’s depth has more than doubled. Market makers came back to the two big coins and stayed through Thursday.

They have not come back to anything else. Depth across the broader basket of alternative coins is down about a third since early 2025, and weekly spot volumes across exchanges average $279 billion, well under half the level of the crash week. The market got sturdier at the top and thinner everywhere below it. Thursday was a reminder of what the second half of that sentence costs.

The Week Ahead

Wednesday is the week. American consumer prices for September land at 14:30, into a market that has spent a fortnight arguing about whether the Federal Reserve moves again this month, in December, or not at all. September already delivered a quarter-point rise to 3.75–4.00%, unanimously, and the minutes published this week said most of the committee expects another before the year ends. Futures give October about one chance in four and December roughly five in six.

What makes this reading awkward is the gap between its two halves. The core measure, which strips out food and energy, is expected to behave. The headline measure, which does not strip them out, has oil running 66% higher than a year ago sitting underneath it. A central bank watching the core can afford to wait. A central bank watching households’ inflation expectations climb to 4.7% has less room.

Earnings season starts the day before, which gives the week a second axis. The big American banks report first, and what they set aside for bad loans will be read as a verdict on everything in chapter two of this letter.

US bank earnings — Tuesday, from 12:00.
JPMorgan, Goldman Sachs, Citigroup and Wells Fargo on Tuesday, Morgan Stanley and Bank of America on Wednesday. Analysts expect financial-sector earnings up about 3% on the year. The lines worth reading are loan-loss provisions and anything said about leveraged lending.

China consumer prices — Wednesday, 03:30.
Consensus 0.9–1.0% after 0.8%. Factory-gate prices expected unchanged at 3.8%, which matters for everything the rest of the world imports.

US consumer price inflation — Wednesday, 14:30.
Consensus +0.6% on the month, lifting the annual rate to 3.6% from 3.4%. Core expected at +0.2% and 2.5%. A core print of 0.3% or higher puts an October rise back on the table; 0.1% or lower pushes the whole argument to December.

Federal Reserve Beige Book — Wednesday, 20:00.
The anecdotal survey of the twelve regional banks. Worth reading for what companies say about passing costs on, which is the question the ISM prices index raised on Monday.

UK GDP, August — Thursday, 08:00.
Consensus a fall of 0.3% to 0.4% after two readings of +0.4%.

US retail sales — Thursday, 14:30.
Consensus +0.3% after +1.1%. The Chicago Fed reckons that once inflation is removed the real figure is nearer −0.7%. With sentiment where it is, this is the test of whether American households are still spending anyway.

US producer prices — Thursday, 14:30.
Consensus +0.5% on the month, annual rate easing to 5.3%. Producer prices are where the oil pass-through shows up before it reaches consumers. Jobless claims at the same time, consensus 197,000.

Euro area final inflation, September — Friday, 11:00.
A confirmation rather than news, though the country detail matters for the French story above. Euro area industrial production for August lands Thursday at 11:00.

☕ Last Sip

Two facts from this week belong in the same sentence. A stock index at a record high, and the worst reading on how Americans see their own finances since that survey began. Both are true at once because they describe different populations. The ten companies carrying the index and the twenty per cent of families who missed a loan payment do not overlap much.

France found itself in the same shape in miniature. Its blue-chip exporters borrow more cheaply than its treasury, because they sell to the world and the treasury can only tax France.

What connects them is sitting at $104 a barrel and up two thirds in a year, feeding service-sector costs in America, inflation expectations everywhere, and the rate path that decides what both kinds of borrower pay next. Wednesday’s number tells us how far along that chain we are.

Enjoy the rest of the weekend. See you next Sunday.

Sources

Market levels from Trading Economics, the Washington Post’s index summary, ANSA for the Italian–German spread, and Seoul Economic Daily for the KOSPI, whose close is Thursday’s because Seoul was shut on Friday. Economic releases from the Institute for Supply Management, Destatis, Eurostat, the US Labor Department and the University of Michigan. Credit and euro curve data from ICE BofA indices via FRED and the ECB Data Portal, as of 8 October. The French, leveraged loan and market breadth figures come from Bloomberg, JPMorgan and Goldman Sachs respectively; onchain data from CoinDesk Research and CoinGlass. Weekly changes are measured against the levels published in edition #11. Analysis and personal opinion — never investment advice.

Then the jobs figures landed at 14:30 CET on Friday, and they were bad. 29,000 new jobs in September against the 84,000 economists had estimated. Unemployment went up to 4.2%.

But now the funny part. Markets absolutely loved these news. Why is that you ask? The chance of another rate hike by the fed this month fell from 64% to 16% in a few hours. Equities rallied, yields came back down, and the week closed looking much calmer than it had been for most of it.

Euro area inflation on the other hand came in at 3.8% for September, the highest in three years, with energy up 18.8% over twelve months.

But now let’s grab yourself a coffee and dive into it properly.

The Scoreboard

Nineteen numbers, and what stands out is how little the stock market rows moved.

Watch the gap between the equity rows and everything that measures the price of borrowing. Most weeks those travel together. This week they went separate ways, and the borrowing side is the one carrying spicy information.

America is where the week happened. US stocks ended roughly flat and debt got noticeably dearer. The S&P 500 closed at 7,722.72, down 0.3%.

The Nasdaq Composite managed +0.5%, smaller companies in the Russell 2000 slipped 0.2%. Meanwhile the ten-year US-treasury rose by 11 basis points to an astonishing 5.28% and thirty-year 13 to 5.63%. The two year, which tracks what people think the central bank will do soon, barely moved at all.

Europe is where the split was most visible to follow. Germany's ten-year borrowing cost fell 15 basis points to 3.46%. Italy pays 4.6% for the same ten-year loan, and that difference, what lenders charge for being less certain of getting their money back, grew by 24 basis points so almost a quarter of a percentage point this week, to 1.15 points. Investors sold the debt of the more indebted country and bought the debt of the less indebted one, same week, same reason. The Euro Stoxx 50 lost 1.0% to 6,241.75 and touched a three-and-a-half-month low before Friday's bounce.

Asia had a pretty solid week. The Nikkei 225 rose 2.9% to 68,309.46, its third winning week in a row, and Korea’s main index climbed back above 7,000. Japanese ten-year borrowing edged up 4 basis points to 3.11% after the Bank of Japan’s September minutes pointed to another rate rise before the year is out.

Gold is the biggest single move on the board and it goes the wrong way. Down 4.2% to $4,140.19, in a week when euro inflation hit a three-year high and oil sat above $100. The dollar explains some of it, up 0.9%. The rest is chapter three.

1. Europe’s borrowing costs split in two

Germany got cheaper to lend to. Italy got dearer.

A government borrows by selling bonds, and the rate it has to offer depends on how sure buyers are of getting their money back. Germany and Italy share a currency, a central bank and a rulebook. They do not share a debt pile, and this week that difference did all the talking.

German ten-year borrowing fell 15 basis points to 3.46%. Italy went the other way — the gap over Germany opened to about 115 basis points by Friday’s close, after touching 131 that morning. Three weeks ago it was near 91.

The mechanism runs through oil. Crude above $100 a barrel feeds into inflation, and euro area inflation duly printed 3.8% for September, with energy alone up 18.8% over the year. Higher inflation means the central bank keeps rates up for longer, and that raises the interest bill on every government that has to keep refinancing. A country with a small debt pile absorbs it. A country with a large one has to pay up to keep buyers interested. Italy’s debt is among the largest in Europe relative to the size of its economy, and France has the same problem.

Money leaving Italian bonds has to go somewhere. The obvious somewhere is German bonds. That is why one yield fell while the other rose.

Keep it in proportion, though. At 115 basis points Italy is paying a premium that looked unremarkable in 2023 and cheap in 2018. The move is fast rather than extreme, and that single Friday-morning print of 131 says more about how thin trading was than about Italy’s solvency.

2. Two counts of one month, and they disagreed again

ADP said the American labour market had a decent September. The government said it had a bad one, and then made the two months before it even worse.

ADP is a payroll company. It handles wage administration for a large share of American private employers, which means it can count who actually got paid. On Wednesday it reported 90,000 new private jobs against the 68,000 economists expected. A clean beat.

On Friday the Bureau of Labor Statistics, the government’s statistics office, reported the non farm payroll stats. These came in at 29,000 new jobs against 84,000 expected. It also revised July from a gain of 21,000 into a loss of 10,000, and August down from 162,000 to 133,000. Unemployment rose to 4.2%. Average hourly pay went up 5 cents in the month, which comes to 3.0% over the year.

Two organisations counted the same weeks and landed 61,000 apart. The one with the longer history pointed down.

Traders read it within minutes as the end of the rate-rise question. They had been pricing a 64% chance of the Federal Reserve raising rates at its meeting on 27 and 28 October. By Friday evening that was 16%.

A three-month average of 51,000 new jobs a month used to arrive with a recession attached. It does not look like that now, because the number of people available to work is also growing slowly, so fewer new jobs are needed to keep unemployment steady. That is the honest version. It also means there is very little room for one bad month.

3. Gold had every reason to rise but fell 4.2% instead

Inflation went up. Oil stayed above $100. And the metal people buy to protect themselves against both had its worst week in months.

Gold pays no interest and no dividend. Holding it costs you whatever you could have earned on a government bond instead, so the price tends to fall when that alternative gets more attractive. This week it got a lot more attractive. American ten-year borrowing paid 5.28% by Friday, thirty-year 5.63%.

The dollar did the rest. Gold is priced in dollars, so a stronger dollar makes the same ounce dearer for everyone outside America and demand falls. The dollar index rose 0.9% on the week while the euro fell 1.2%.

So the inflation story and the interest rate story pulled against each other, and the interest rate story won by a distance. Worth remembering the next time a gold rally gets explained purely by inflation fear. The same reasoning should have held this week. It did not.

Two things went genuinely well, and they deserve the same airtime as the rest. Japanese shares finished a third straight winning week, up 2.9%, in a country whose central bank is tightening rather than loosening. And the Group of Seven agreed to release up to 100 million barrels of emergency oil over four months — the first coordinated attempt to put a ceiling on the fuel price that is driving every inflation number in this letter.

Data Check

Five releases mattered, and the week splits neatly down the middle: what the private sector reported, and what the official statistics said two days later.

The euro area inflation row is the one worth a second look, because it was the quiet one. 3.8% against 3.6% expected is a two-tenths miss that moved nothing on the day, and it is also the highest reading in three years and the reason Italian borrowing costs had the week they had. The counter-example sits two rows above: ADP’s 90,000 was the only number all week pointing to a healthy labour market, and it came first.

The Credit Grind

The premium companies pay to borrow rose more this week than in any week of the past year, and it happened on both sides of the Atlantic at once.

Anyone lending to a company is paid for two things. One is the price of money itself, which governments set. The other is compensation for the chance this particular company stops paying. That second part is what this section tracks, and it has spent most of 2026 being remarkably cheap but ended this week.

Let’s start with the bottom row. Euro emerging market corporates now pay 156 basis points over governments, and the last column reads 100%. That column counts the share of the past year’s trading days that sat lower than today, so 100% means this is the dearest these companies have been to lend to in twelve months. There is no cheaper day left to compare it with.

The two high yield rows, the debt of companies rated least likely to repay, sit at the 97th percentile on both continents. Europe rose 40 basis points in the week to 317, America 44 to 324. Even investment grade, the debt of large and solid American companies, rose 7 to 86 and sits at the 92nd percentile.

Four indices moving together like that is a different animal from one index moving alone. It says the price of taking company risk got reset across the board, rather than one borrower getting into trouble.

The government rows underneath tell you what reset it. Two-year euro borrowing fell 12 basis points while ten-year rose 3, a market that expects the central bank to sit still and still wants more to lend for a decade.

If you hold a corporate bond fund, the compensation you have been missing all year arrived in a single week. The price of it arriving is that the bonds you already owned fell.

The Onchain Pour

Crypto went up while credit sold off, which is a combination worth sitting with for a second.

Bitcoin ended near $86,000, up about 2.4% against the $84,000 published here last Sunday. Ether finished around $2,700, up 5.5%, breaking out of the $2,600 range it had held since late September. Both did their best work on Friday, when falling bond yields and the collapse in rate-rise odds did the heavy lifting.

The flows point the same way. More than 40,000 bitcoin have left exchanges since 22 September, which usually means buyers moving coins into storage rather than keeping them handy to sell. CryptoQuant counts 75,000 bitcoin picked up by large holders over thirty days.

Glassnode adds the uncomfortable detail. The people selling into this rally are the ones who bought the 2025 run at around $97,000 and $89,000, and they are offloading more coins per day than any other group this year. The ones who bought the decline are holding. A rally that runs on earlier buyers giving up has a ceiling made of their average purchase price, and that ceiling sits above today’s level.

The Week Ahead

After a week like that one, the calendar looks thin, and that is the point. No central bank meets. No American inflation report lands. The week’s job is to tell you whether the credit selloff was a repricing that has now happened, or the opening move in something longer.

Wednesday evening carries the most weight. The Federal Reserve publishes the account of its September meeting, the one where it raised rates, and traders will comb it for how close the vote was and what would make the committee move again. It arrives into a market that has just cut the odds of an October rise from 64% to 16%, so any hint that September was more contested than it looked would land hard.

The other thing to watch is the auctions. America sells ten-year debt on Wednesday and thirty-year debt on Thursday, in a week when thirty-year borrowing just hit its highest since 2002. Who turns up tells you whether this week was investors repricing risk or investors stepping away.

US ISM services survey — Monday, 16:00.
Consensus 55.1 after 55.4. Services employ most Americans, and the survey’s own employment component has been below 50 for months while the headline sits comfortably above it.

German factory orders — Tuesday, 08:00.
Consensus −1.0% after +2.5%. German factories are the part of Europe most exposed to an energy price up 18.8% over the year.

Euro area retail sales — Tuesday, 11:00.
Consensus +0.3% after −0.6%. A first read on whether European households are absorbing higher fuel bills or cutting elsewhere.

German industrial production — Wednesday, 08:00.
Consensus +0.5% after −1.1%.

Federal Reserve minutes — Wednesday, 20:00.
The account of the 15–16 September meeting. A ten-year debt auction runs during the afternoon.

ECB meeting account — Thursday, 13:30.
The European equivalent, and the better guide to whether 3.8% inflation changes anything in Frankfurt. Jobless claims at 14:30, consensus 200,000. A thirty-year auction follows.

Canadian employment — Friday, 14:30.
Consensus +9,500, unemployment expected to tick up to 6.5%.

University of Michigan consumer sentiment — Friday, 16:00.
Consensus 48.1, unchanged and close to the weakest readings this survey has ever produced.

☕ Last Sip

Two weeks ago the question was whether the Federal Reserve would raise rates again. This week the market answered with a 16% probability — and spent the four days before that answer repricing the cost of lending to companies by more than it has in any week of 2026.

Both things can be true at once. A central bank that stops raising rates helps borrowers who already have cheap debt. It does nothing for the ones who have to refinance into a market that just decided risk was underpriced.

That is the tension to carry into the autumn, and the Fed’s own account on Wednesday is the next place it turns up.

Enjoy the first proper weekend of October. See you next Sunday.

Sources

Market levels from Yahoo Finance, the US Treasury, Trading Economics, ANSA and Sky TG24 for the Italian–German spread. Economic releases from ADP, the Institute for Supply Management, Eurostat and the US Bureau of Labor Statistics. Credit and euro curve data from ICE BofA indices via FRED and the ECB Data Portal, as of 1 October. Onchain from Glassnode and CryptoQuant via Cointelegraph. Weekly changes are measured against the levels published in edition #10. Analysis and personal opinion — never investment advice.

The weekend got away from me since there was plenty on, and here in Austria there was something worth celebrating, so the writing waited until the coffee was finished. Thank you for your patience. The markets, helpfully, did not move over the weekend.

The thirty-year American borrowing cost finished the week at 5.50 per cent, marking it the highest since 2004. The ten-year also touched its highest level since 2007. Both rose about sixteen basis points, a basis point being one hundredth of a percentage point, while the two-year rose only seven.

That is the exact shape this letter said would falsify its own reading: the test named was the thirty-year climbing back toward 5.5 per cent while the two-year held. Seven days later, both conditions are met. Equities rose anyway, with the S&P 500 gaining 1.2 per cent and closing within 0.7 per cent of its record, because oil did the heavy lifting on Friday and Brent dropped below 98 dollars a barrel.

The Scoreboard

Nineteen numbers, and the American bond block has inverted its own story from a week ago.

America rose across the board and did it unevenly. The Dow Jones ended a three-week losing streak with a 0.3 per cent gain. The Nasdaq added 2.1 per cent, helped by Meta Platforms rising nearly 13 per cent on the week after news about its artificial-intelligence agent Muse. Information technology gained 3.1 per cent, more than any other sector in the S&P 500. The Russell 2000 that represents smaller American companies, who borrow at floating rates, fell 0.8 per cent.

Europe followed America. The Euro Stoxx 50 gained 1.1 per cent, and German ten-year borrowing costs rose nine basis points to 3.61 per cent. What Italy pays above Germany narrowed by one, to 91 basis points, because German yields rose faster than Italian ones.

Asia was strong and partly closed. The Nikkei 225 rose 2.1 per cent to 66,364.20, its fifth consecutive daily gain, with 31 of the Tokyo exchange’s 33 industry groups finishing higher on Friday. Japan’s ten-year borrowing cost reached about 3.07 per cent, the highest since August 1996. The yen weakened a little further, to around 157.75 per dollar.

1. Short money paused and long money kept going

A week ago the gap between two-year and ten-year American borrowing costs was 25 basis points, the narrowest of the year. It now sits near 36.

Three things drove it, and the first is now measurable. American purchasing manager surveys landed on Wednesday at 57.0 for manufacturing against 53.6 expected, and 58.7 for services against 56.0, the fastest private-sector expansion in the series this year. Federal Reserve Governor Michael Barr spoke in hawkish terms the same afternoon. Energy prices stayed high through midweek because of the Iran conflict.

Then the Treasury tried to sell debt into it. The two-year auction on Tuesday cleared at 4.787 per cent against 4.204 the previous month. Wednesday’s five-year went at 5.033, above five per cent for the first time. Thursday’s seven-year cleared at 5.085. Each sale landed worse than the one before, which is the plainest evidence that buyers wanted paying more. Traders now price a roughly 64 per cent chance of another rate increase in October, up from about half a week earlier.

Thirty-year rates reflect how much debt is being sold and what inflation might average over decades, rather than what the central bank does next month. When they rise on their own, the cost of a home loan follows, and American thirty-year mortgage rates are now near seven per cent. So the correction to last week’s reading is this: the flattening was the long end lagging, and it has now caught up. I said that would change my mind, and it has.

2. Equities rose and almost nothing participated

The headline numbers were good. What sat underneath them was not. Friday was the ninth consecutive session in which more American companies hit a fifty-two-week low than a fifty-two-week high.

Across September, technology gained 5.4 per cent while utilities lost 5.9 per cent, and five of the eleven S&P 500 sectors fell 4.6 per cent or more. An index can rise while most of its members fall, as long as the largest few rise enough. That is what happened. It works until the largest few stop working as well.

3. Oil turned, and Friday turned with it

Brent crude fell below 98 dollars on Friday, down roughly six per cent on the week, and commodities as a group fell for the second week running. The trigger was diplomatic. Houthi forces in Yemen wrote to the European Union to say they would stop targeting European ships in the Red Sea, and reports circulated about a possible opening around the Strait of Hormuz. Iranian state media disputed that, which is a useful reminder of how thin the evidence is.

Cheaper oil lowers expected inflation, which lowers expected interest rates, which is why American companies rallied into the close and Treasury yields eased on Friday even after a brutal week. Gold went the other way, falling 2.1 per cent to 4,321 dollars. A rising dollar and rising real yields make an asset that pays no interest more expensive to hold, and this week that beat every geopolitical argument for owning it.

Data Check

Nine releases, and eight came in stronger than economists had forecast.

The American housing figures are the correction this letter owes its readers. A week ago this section read weak starts and permits as the first crack from higher borrowing costs. New home sales then rose 6.4 per cent in August to 684,000, well ahead of the 620,000 expected. Builders are discounting hard and buyers are taking it, which is a different story from the one I told.

What did break is how people feel. The University of Michigan’s final reading of consumer sentiment for September came in at 48.1, down from 51.7 in the preliminary cut three weeks earlier. Households now expect prices to rise 4.6 per cent over the coming year, against 4.0 per cent in the first estimate. Sentiment falling that far inside one month, while sales and hiring hold up, is the gap a central banker has to decide how to read.

The Credit Grind

European governments now pay more to borrow than on any day of the past year. And this week, for the first time in a month, what companies pay on top moved with them. Anyone lending to a company is paid for two things. One is the price of money itself, which governments set, because a state is the safest borrower in its own currency. The other is compensation for the chance this particular company stops paying. The board below separates the two.

Both euro government rows read 100 per cent, which means no day in twelve months saw a higher yield. Two-year money in the euro area costs 3.23 per cent, up 112 basis points since January. The company rows are the news. European high yield which ia the debt of firms rated least likely to repay, rose eight basis points to 277, moving from the 47th percentile of its year to the 66th in a single week. American high yield rose ten to 280, from the 12th percentile to the 43rd. Euro emerging-market corporates sit at 130 basis points, the 89th percentile.

Last week this section said the risky part was priced as though nothing could go wrong. That changed. And the place it changed first is worth naming: Goldman Sachs reported midweek that insurance against default by the large cloud computing companies had moved two to four basis points wider while their shares sat unchanged. Those firms have been issuing debt heavily as yields climb. The equity market is still paying for the artificial-intelligence buildout. The credit market has started asking what it ultimately costs.

The Onchain Pour

Bitcoin broke above a resistance level early in the week and held there, finishing near 84,000 dollars, up roughly 3.5 per cent against the level published here last Sunday. Ether went the other way, slipping about two per cent to around 2,560. What stands out is the contrast with shares: in a week when barely anything in the American stock market participated in the rally, 44 of 46 tokens in one widely followed ecosystem index traded above their fifty-day average. Breadth collapsed in one market and held in the other.

Against that, two pieces of housekeeping. The exchange Bitget raised the figure for its security incident to 387.5 million dollars and launched a recovery bounty while preparing to restart withdrawals. And the United States seized bank accounts belonging to a payments group linked to Tether, the largest stablecoin issuer. Stablecoins are the plumbing that the institutional case for this asset class rests on, and pressure there is a different kind of risk from price.

The Week Ahead

This is the week the American labour market and the European price level both report, two days apart.

Monday, 16:30 — US Dallas Fed manufacturing index.
Last at 11.6, expected to fall to 1. Regional surveys have been the first to wobble.

Tuesday, 06:30 — Reserve Bank of Australia decision.
From 4.35 per cent, with economists expecting a rise to 4.6. A fourth central bank joining the tightening wave.

Tuesday, 16:00 — US job openings and consumer confidence.
Openings last at 7.271 million. Confidence at 89.4, expected at 90.

Wednesday, 03:30 — Chinese purchasing manager surveys.
Official manufacturing last at 49.8, expected at 50 — the line between contraction and growth.

Wednesday, 14:00 — German flash inflation for September.
Last at 2.9 per cent, expected at 3.2. The euro area’s largest economy moving further from target.

Wednesday, 14:30 — US personal consumption expenditures price index.
The Federal Reserve’s preferred inflation measure. Core last at 0.2 per cent month on month, expected at 0.3. This lands three weeks before the October meeting.

Thursday, 16:00 — US manufacturing survey from the Institute for Supply Management.
Last at 54.6. The prices component, last at 71.1 and expected at 72.5, is the one to read.

Friday, 11:00 — euro area flash inflation.
Last at 3.2 per cent, expected at 3.6. That is a large jump for one month, and the European Central Bank meets in October.

Friday, 14:30 — American jobs report.
162,000 jobs were added in August. Economists expect 84,000 for September, roughly half. Unemployment is seen holding at 4.1 per cent.

Sunday, 4 October — OPEC and non-OPEC ministerial meeting.
Held days after Brent fell six per cent.

☕ Last Sip

Being wrong quickly is better than being wrong slowly, and this letter was wrong twice. The thirty-year went where I said would break last week’s argument about the yield curve, so that argument breaks. And the housing crack I described in the same edition did not widen — new home sales beat comfortably. What replaces both is simpler. American long-term borrowing costs are at twenty-year highs because the economy keeps running hot and the government keeps selling debt into it, not because anyone expects a recession. Mortgage rates near seven per cent are what that looks like from a kitchen table, and consumer sentiment at 48 is what it feels like.

Watch Wednesday afternoon and Friday morning. Enjoy the start of the week.

Sources

Market levels from CNBC, Yahoo Finance, Investrade, the Associated Press, Trading Economics, ANSA and STL News. Economic releases and consensus figures from Trading Economics, the US Census Bureau and the University of Michigan. Auction results from the US Treasury. Credit and euro curve data from ICE BofA indices via FRED and the ECB Data Portal, as of 24 September. Charts and market colour from posts by @MikeZaccardi, @zerohedge, @unusual_whales, @Andre_Dragosch, @jvisserlabs and @Cointelegraph. Weekly changes are measured against the levels published in edition #9. Analysis and personal opinion — never investment advice.

The Federal Reserve raised its interest rate on Wednesday for the first time in three years, to a range of 3.75 to 4.00 per cent. Kevin Warsh, who runs it, then spent the press conference making clear that more may follow soon. The Dow fell 800 points while he spoke.

Thursday belonged to London. The Bank of England left its rate at 3.75 per cent and announced it would stop selling the government bonds it had bought years earlier — a decision that removes a steady seller from the market and pushed British borrowing costs down sharply.

Friday was all about Tokyo. The Bank of Japan lifted its policy rate to 1.25 per cent, the highest since 1995, three months after its last increase. Two of the nine board members voted against. The yen fell anyway.

Underneath all three sits one number that moved in the opposite direction from everything else. What America pays to borrow for thirty years ended the week one basis point lower — a basis point being one hundredth of a percentage point — while two-year money rose eleven.

The Scoreboard

Nineteen numbers, and the ones that matter are all in the American bond block.

The thing to look for is the gap between the short and the long end. When two-year money rises hard and thirty-year money does not, the market is saying it expects tight policy soon and something weaker later.

America looks flat and wasn’t. The S&P 500 ended the week 0.1 per cent lower, which hides a Wednesday sell-off and a Thursday recovery. The Dow lost 1.7 per cent, its worst week since March. The Nasdaq gained 0.7 per cent, carried almost entirely by the largest technology companies, while the Russell 2000 — smaller American firms, who borrow at floating rates and feel a central bank first — fell 1.5 per cent.

Europe had the worst Friday. The Euro Stoxx 50, which tracks the fifty largest listed companies in the euro area, lost 1.4 per cent, and money moved out of every euro government bond except Germany’s. What Italy pays above Germany widened seven basis points to 92. France went further: its ten-year borrowing cost now sits 105 basis points above Germany’s, the widest since 2012, and after the close the rating agency Scope cut France one notch to A+.

Asia did the opposite of what a textbook would predict. Japan raised rates and the yen weakened, ending near 156.7 per dollar, about 2.1 per cent softer. The Nikkei 225 rose 2.5 per cent to 65,018.95. Korea’s KOSPI jumped 2.7 per cent on Friday alone and still finished the week slightly lower, at 6,894.23.

1. The Fed raised rates and thirty-year borrowing got cheaper

Wednesday’s increase was fully priced before it happened. What moved markets was the projection sheet published alongside it: slightly stronger growth, slightly higher inflation, and a higher path for the policy rate than officials had sketched in June.

Traders now put the chance of another increase in October at roughly one in two, with some measures nearer 58 per cent. That meeting falls one week before the midterm elections, which is its own kind of problem.

Look at what happened across the maturities. Two-year borrowing costs rose eleven basis points to 4.74 per cent. Ten-year rose four, to exactly 5.00 per cent. Thirty-year fell one, to 5.34 per cent. The gap between two-year and ten-year money is down to about 25 basis points, the narrowest this year.

Two-year rates mostly reflect what the central bank is expected to do over the next couple of years. Thirty-year rates reflect something slower: how much debt is being sold, who wants to hold it, and what inflation might average over a working lifetime. When the first rises and the second does not, the market is pricing tighter money now and a weaker economy later.

The honest objection is that this reads too much into one shape. Thirty-year yields have been pinned high all year by the sheer volume of debt America is issuing, and a curve can flatten because the long end is heavy rather than because growth is fading. Foreign holdings of Treasuries fell 50 billion dollars in July, to 9.25 trillion — the fourth monthly decline in five. That is a supply story, not a recession signal.

My own lean is that the flattening is real and worth taking seriously. What would change my mind is the thirty-year rising back toward 5.5 per cent while the two-year holds: that would tell you the long end was simply catching up, and the message I am reading into it was never there.

2. Britain stopped selling, and gilts noticed

The Bank of England did the expected thing with its interest rate and an unexpected thing with its balance sheet.

Between 2009 and 2021 the bank bought hundreds of billions of pounds of British government bonds. Since 2022 it has been selling them back into the market, a process that adds a large, predictable, price-insensitive seller on top of everything the Treasury already issues. On Thursday it stopped.

Gilt yields fell immediately. The thirty-year dropped ten basis points to 5.76 per cent, its biggest one-day fall since May. Removing a seller raises the price of what remains, and bond prices and yields move in opposite directions.

British inflation had come in at 3.1 per cent for August the previous morning, up from 2.6 per cent, which made the hold a harder call than it looked. The bank’s answer was to keep the rate steady and ease the pressure elsewhere. Whether that counts as tightening or loosening depends on which lever you think matters more.

3. Japan raised rates and the yen fell

For three decades money has been borrowed cheaply in yen and invested in things that pay more elsewhere. Higher Japanese rates narrow that gap and, in theory, pull some of that money home — which would lift the yen.

Friday did the reverse. The yen weakened past 157 during the session before settling near 156.7.

Two things explain it. The vote split seven to two, with dissenters arguing that core inflation, at 1.7 per cent in August, still sits below the two per cent target. And Governor Kazuo Ueda declined to signal anything about the next move. A rate increase delivered without a promise of more is read as the end of something rather than the start.

That reading matters beyond Tokyo. Japanese retail investors held roughly 24 billion dollars in bets against their own currency last week. If the Bank of Japan does turn more aggressive — the next meeting is 29 and 30 October, with fresh forecasts attached — those positions unwind fast, and the unwinding shows up as selling in New York and Frankfurt.

Data Check

Nine scheduled events, and the three that carried the most weight all landed exactly where forecasters said they would.

The rows worth a second look are the American housing pair. Starts came in at 1.275 million against 1.320 million expected, permits at 1.394 million against 1.408 million, and homebuilder confidence is near its Covid lows. Those are the first numbers to soften when borrowing costs rise, and they softened in the same week the Fed promised more of them.

Against that, American consumers spent 1.2 per cent more in August than in July, the strongest month in five, and new claims for unemployment benefit fell to 196,000 — the lowest in 57 years. An economy where housing cracks while spending and hiring hold up is the awkward middle that central bankers find hardest to read.

The Credit Grind

European governments have never paid more to borrow in the past year than they did this week. What companies pay on top of that barely moved.

Anyone lending to a company is paid for two things. One is the price of money itself, which governments set, because a state is the safest borrower in its own currency. The other is compensation for the chance this particular company stops paying. The board below separates the two.

Both euro government rows read 98 per cent. Two-year money in the euro area costs 3.15 per cent, up 104 basis points since January.

Now the company rows. European high yield — the debt of firms rated least likely to repay — rose one basis point to 269, the 47th percentile of its year. American high yield did not move at all, sitting at 270. The gap between the two continents is one basis point.

So a European bond fund earned more this week, and essentially all of it came from the government part. The thing you are taking risk for is still priced as though nothing could go wrong — in a week when three central banks tightened and France lost a rating notch.

The Onchain Pour

Bitcoin spent Tuesday and Wednesday falling through support near 76,000 dollars, touched roughly 74,800, failed to hold the breakdown, and finished the week near 81,000 — up 5.7 per cent against the level published here last Sunday. Ether gained 6.6 per cent to around 2,612 and closed its best third quarter on record.

The flows explain the shape. American bitcoin funds lost roughly 746 million dollars on Tuesday and Wednesday, then took in about 593 million across Thursday and Friday, including 433 million on Friday alone. Traders who had sold below 76,000 had to buy back, which amplified everything.

Two explanations are circulating. One points to the Securities and Exchange Commission, which issued an order on tokenised securities late in the week. The other points to the Gulf, where drone attacks on Saudi infrastructure and a possible American operation against Houthi forces have people thinking about wealth that can leave a country quickly. Fund flow data records the wrapper, not the buyer’s nationality, so neither can be proven from the numbers alone.

The Week Ahead

No central bank meets this week, which makes it the week the data either supports what three of them just did, or doesn’t.

Underneath the releases sits something quieter. The US Treasury sells two-year, five-year and seven-year notes on Tuesday, Wednesday and Thursday — precisely the maturities that moved hardest last week. Weak demand at any of the three would tell you the repricing has further to run.

Monday, 03:15 — China’s benchmark lending rate.
Expected to stay at 3.0 per cent. China is still exporting falling prices into a world that is importing rising ones.

Tuesday, 16:00 — Euro area consumer confidence.
Last at −15.5, expected to weaken to −16.5. The first read on how households took the energy bill.

Tuesday, 19:00 — US two-year note auction.
The previous sale went at 4.204 per cent. Two-year yields have since risen to 4.74. This is where you find out who wants them at the new price.

Wednesday, 09:15 to 15:45 — flash purchasing manager surveys.
France, Germany, the euro area, Britain and the United States in sequence. These ask companies directly whether business got better or worse this month, which makes them the earliest signal of a slowdown. German services sat at 49.7 last month, just below the line that separates growth from contraction.

Wednesday, 11:30 — German thirty-year Bund auction.
Last cleared at 3.90 per cent. Europe’s long end has been the calm one, and this is the test.

Thursday, 14:30 — US jobless claims.
Last week’s 196,000 was the lowest in 57 years. Economists expect 202,000. A second week that low would make the Fed’s hawkishness considerably easier to defend.

Thursday, 16:00 — US new home sales.
The previous month fell 10.5 per cent. Expected at 610,000. Housing is where higher borrowing costs land first, and last week’s starts and permits both disappointed.

Thursday, time to be confirmed — the Trump–Xi summit.
Scott Bessent and He Lifeng have been negotiating over trade, artificial intelligence and rare earths in the run-up.

Friday, 14:30 — US durable goods orders.
Last month rose 1.1 per cent, and economists expect a fall of 0.5 per cent. Orders for machines and aircraft say what companies are willing to commit to eighteen months out.

Friday, 16:00 — University of Michigan consumer sentiment, final.
The number to watch this week. The preliminary reading was 51.7 and the final is expected at 47.8 — a collapse of nearly four points within the same month. Inflation expectations for the year ahead are seen at 4.6 per cent against 4.0 per cent in the first cut.

☕ Last Sip

Last week this letter said three central banks were leaning the same way for the first time since 2022. All three have now moved, and none of them surprised anybody.

What is awkward is what the bond market did with it. Short-term borrowing costs rose, long-term costs did not, and the gap between them is the narrowest this year. That shape has a reputation, and it is not a flattering one.

Oil came off its peak, gold rose, and American shares ended roughly where they started. Enjoy what is left of the weekend. See you next Sunday.

Sources

Market levels from CNBC, Yahoo Finance, Investrade, Trading Economics, ANSA, Il Sole 24 Ore Radiocor and Rio Times. Economic releases from the US Census Bureau, the Department of Labor and the Office for National Statistics. Credit and euro curve data from ICE BofA indices via FRED and the ECB Data Portal, as of 17 September. Weekly changes are measured against the levels published in edition #8. Analysis and personal opinion — never investment advice.

Europe’s central bank raised interest rates on Thursday, for the second time this year. Christine Lagarde, who runs it, called the decision an easy one and named her reason plainly: the war in the Middle East keeps pushing energy prices up, and energy gets into the price of almost everything else.

Then came Friday, and American inflation figures that matched the forecast to the decimal. Prices were 3.4% higher than a year ago, as predicted. Leave out food and fuel and you get 2.4%, also as predicted.

Markets repriced anyway. Within an hour the chance of the Federal Reserve raising rates next Wednesday had gone from 72% to 87%. What America pays to borrow for two years finished the week 26 basis points higher — a basis point being one hundredth of a percentage point.

One number sits underneath all of it, and it is not an inflation rate. Brent crude, the oil price most of the world quotes, rose 8.5% this week to $104.61 a barrel.

The Scoreboard

Nineteen numbers, and the ones worth your attention are all in the same place: what governments pay to borrow.

The thing to look for is which maturity moved. When two-year money moves much further than thirty-year money, the market is saying something quite specific — that it expects the central bank to act soon.

America is where the week happened. What the government pays to borrow for two years rose 26 basis points to 4.63%, for ten years 18 to 4.96%, for thirty years only 11 to 5.35%. Shares fell for four days and took most of it back on Friday: the S&P 500 closed at 7,656.98, down 0.8% on the week. Smaller American companies did worse, with the Russell 2000 down 2.4% — they borrow at floating rates and feel a central bank sooner than large firms do.

Europe had the rate rise and still finished calmer. The Euro Stoxx 50, which tracks the fifty largest listed companies in the euro area, lost 1.2% to 6,322.25. Germany’s ten-year borrowing cost rose 16 basis points to 3.50%, its highest since August 2009. What Italy pays above Germany moved 4 basis points, to 85.

Asia split. The Nikkei 225 fell 2.4% to 63,442, most of it on Friday, as higher oil and rising domestic borrowing costs landed on Tokyo together. Korea went the other way and finished 3.3% higher at 6,909.91 — foreign and institutional money came back earlier in the week and that was enough to carry it. Japan’s ten-year rate rose 8 basis points to 2.99%, just under the 3% it touched last week for the first time since 1996.

Gold is the one worth a pause. It fell 3.5% to $4,350.36, a third losing week in a row, in the same days inflation turned out to be stickier than hoped. Higher interest rates make holding something that pays you nothing more expensive, and this week that beat the inflation story completely.

1. Europe raised rates against a price it cannot control

The ECB lifted its rate a quarter point on Thursday, and the thing pushing euro area inflation up is a barrel of oil in the Persian Gulf.

The European Central Bank sets the price of money for twenty countries. Its main tool is the deposit rate, which is what commercial banks earn for parking money there overnight. On Thursday it went from 2.25% to 2.50%. Everything else in the euro area — mortgages, business loans, savings accounts — sits downstream of that number. The bank’s own statement gave the reason: the Middle East conflict keeps driving prices up, and inflation should stay well above the 2% target for most of 2027.

Oil is not only fuel. It is the cost of moving a pallet, heating a factory, making plastic and fertiliser. When crude rises, those costs reach consumer prices a few months later. American wholesale prices, published Thursday, show it happening: up 0.4% in August, with more than three quarters of that coming from energy. Diesel alone rose 24.1% in the month.

Now the obvious objection, and it is a good one. A central bank cannot drill a well. Raising the cost of borrowing in Frankfurt does nothing about a tanker route near Iran.

What the bank is defending is the second round. Energy passes into wages, into the prices firms feel safe setting next year, and into what people simply expect inflation to be. Those are things a rate can reach. The cost of being wrong is that borrowers across twenty countries pay more for a barrel of oil nobody in Frankfurt can move.

2. The forecast was right and the market moved anyway

August consumer prices came in exactly as predicted on the year, and traders spent the morning repricing the Federal Reserve regardless.

Friday at 14:30 CET the Bureau of Labor Statistics, the American government’s statistics office, published August consumer prices. Up 3.4% over twelve months against 3.4% expected. The core measure, which leaves out food and energy because those swing for reasons a central bank cannot influence, came in at 2.4% against 2.4% expected.

A forecast that lands twice is usually a quiet morning. The reason this one was not is a number that gets much less coverage.

Month to month, core prices rose 0.3%. Economists had it at 0.2%, and July had been 0.2%. That tenth of a point is the fastest monthly pace since April, and monthly figures are what tell you about now. The annual number is twelve months averaged together, most of it already known.

Petrol explains the headline: up 3.9% in the month, more than a third of the entire price rise. Take fuel out and the core still sped up, which is the part a central banker will worry about. The energy is no longer staying in the energy column.

Traders repriced within the hour. The chance of a rate rise on Wednesday went from 72% to 87% on CME’s measure, briefly touching 91%. Prediction markets, where people bet with their own money, moved from 63.7% to 81.3%.

You can see the same thing in borrowing costs. Two-year money rose 26 basis points on the week, thirty-year money 11. The two-year rate mostly reflects what people think the central bank will do soon; the thirty-year reflects how much debt is being sold and who wants it. The gap between them narrowed sharply, and that shape means tighter soon, calmer later.

3. Japan decides last and may move furthest

The yen strengthened again without anyone buying it, and Japan’s central bank meets on Friday.

This letter has followed the yen since July, when Japan and the United States bought it together for the first time since 1998 to stop it falling. That is called intervention, and it worked for about two weeks.

What has been doing the job since is the expectation of higher Japanese rates. The yen ended the week around 153.5 per dollar, against 156.26 a week earlier, so it strengthened roughly 1.8% — in a week when the dollar was otherwise being pushed up. That takes some force.

Japan’s central bank meets Thursday and Friday and is widely expected to raise its policy rate to 1.25%, the highest in about thirty-one years. Its ten-year borrowing rate sits at 2.99%, a level that belonged to the 1990s until a few weeks ago.

Why this matters outside Japan: for three decades money has been borrowed cheaply in yen and invested in things that pay more elsewhere. Raising Japanese rates narrows that gap and gives some of that money a reason to come home. It gets sold in New York and Frankfurt, and shows up as weakness in places with no obvious link to Tokyo.

Two things went genuinely well this week. Oil fell 2.8% on Friday as the spike eased, and shares on both sides of the Atlantic took much of their weekly loss back in one session. And Italy’s borrowing costs stayed calm straight through a European rate rise — three years ago that would have been the story of the month.

Data Check

Six releases mattered this week, and the pattern across them says more than any single one. Four landed on forecast. The two that missed went in opposite directions, and the market cared about only one.

The row worth a second look is Friday’s monthly core figure: 0.3% against 0.2% expected. Everything else in that report was as predicted, and this one tenth moved rate expectations fifteen points. Monday was the opposite. German industrial production fell 1.1% in July against an expected rise of 0.3% — a far bigger miss, in the wrong direction for the ECB’s growth story, and it moved nothing. Two days later the ECB raised rates anyway.

The Credit Grind

The cost of borrowing for Europe’s safest governments has never been higher in the past year than it was on Thursday. What companies pay on top of that moved by three basis points.

Anyone lending to a company is paid for two things. One is the price of money itself, which governments set, because a state is the safest borrower in its own currency. The other is compensation for the chance this particular company stops paying. Central banks move the first directly. The second depends on how much money is looking for somewhere to go.

The board below has six rows and one job: to show what borrowing costs right now, and how unusual that is against the past year.

Both government rows read 100% in the final column. That column counts the share of the past year’s trading days that sat lower than today, so 100% means there was no day in twelve months when a top-rated European government paid more to borrow. Two-year money in the euro area now costs almost a full percentage point more than it did in January.

European high yield — the debt of companies rated least likely to repay — rose 3 basis points to 268, the 45th percentile of its year. Investors were paid more than this on roughly half the past year’s days. Against 21 basis points on the government row above it, that is close to nothing.

America is the surprise. Its high yield row rose 5 basis points to 270 in a week when the Federal Reserve did nothing at all, carrying it from the 2nd percentile of its year to the 11th. Two weeks ago that column read 0%. The gap between the two continents flips back to 2 basis points in Europe’s favour after touching parity last week.

So a European bond fund paid you more this week, and nearly all of that came from the government part. The thing you are taking risk for is still priced as though nothing could go wrong, with the Fed meeting on Wednesday.

The Onchain Pour

Bitcoin lost ground in a week built for it to lose ground, while money kept moving into ether.

Bitcoin ended near $76,800, down 4% against the $80,000 published here last Sunday. Ether barely moved by comparison, off 0.8% at around $2,450. Both face the same problem: higher rates for longer make an asset paying no interest more expensive to hold when the alternative pays 4.6%. Only bitcoin was sold on it.

The flows say the same thing. American bitcoin funds saw money leave on four consecutive trading days, $462.7 million over the week. Ether funds took in about $197 million across the same days, including $216.4 million on Friday alone, and finished a fourth straight week of inflows.

That split has been building all summer. Institutions that arrived for bitcoin as a scarce asset are increasingly buying ether as a bet on something else — the plumbing underneath stablecoins and tokenised assets, now around $346 billion across dozens of asset types. Whether that is the better bet is not obvious. It is certainly a different one.

The Week Ahead

This is the week the argument gets settled, or at least the American half of it. Three central banks meet within three days of each other, all of them looking at the same oil price, and none of them able to do anything about it.

The Federal Reserve goes on Wednesday with the market already 87% sure of the answer. That certainty is itself the risk: when a decision is this fully priced, what moves markets is not the decision but the sentence afterwards about what comes next. Britain follows on Thursday with a genuinely open question, and Japan on Friday, where a rise would take rates to their highest since the mid-1990s.

Underneath the three meetings, two data points are worth more than they look. American retail sales on Wednesday will say whether the consumer is still spending while prices climb. And British inflation the same morning lands hours before the Bank of England has to decide.

Canadian inflation — Monday, 14:30.
Last at 3.0% for the year. Canada is running its own experiment in what 50% American tariffs do to prices, and this is the first clean read on it.

Chinese industrial production and retail sales — Tuesday, 04:00.
Previously 5.9% and 6.4% over the year. China is still sending falling prices out into the world while everyone else imports rising ones.

British labour figures — Tuesday, 08:00.
Unemployment expected to stay at 4.9%. Wage growth here is what the Bank of England will be reading two days later.

British inflation — Wednesday, 08:00.
Last at 2.6%, closer to target than America’s, which is why Thursday is a harder call than Wednesday.

American retail sales — Wednesday, 14:30.
The previous month fell 0.6%. A consumer already pulling back would make a rate rise that afternoon considerably harder to defend.

The Federal Reserve decision — Wednesday, 20:00, press conference 20:30.
An 87% chance of a rise to 3.75–4.00%. The press conference matters as much as the decision, because the disagreement between Kevin Warsh and Christopher Waller has not been settled in public since.

The Bank of England decision — Thursday, 13:00.
From 3.75%, and genuinely unclear. Inflation nearer target, a weaker labour market, and the same oil price as everyone else.

The Bank of Japan decision — Friday, press conference 08:30.
Expected to raise to 1.25%, the highest in about thirty-one years.

☕ Last Sip

Last week this letter said three central banks were leaning the same way for the first time since 2022. One has now acted, and the other two answer this week.

What makes it awkward is the cause. Oil is up more than 16% in a month because of a conflict none of them can influence, and all three are raising the cost of borrowing for households and businesses who had nothing to do with it. That is the trade-off central banking actually is, most of the time. It usually just happens with a less obvious cause sitting in plain view.

Watch Wednesday evening, then Friday morning. If both the Fed and the Bank of Japan move, the cheapest money in the world gets meaningfully more expensive inside thirty-six hours.

Enjoy what is left of the weekend. See you next Sunday.

Sources

Market levels from the Associated Press, the US Treasury, Trading Economics, ANSA, News On Japan and Seoul Economic Daily. Economic releases from the US Bureau of Labor Statistics, Destatis and the European Central Bank. Credit and euro curve data from ICE BofA indices via FRED and the ECB Data Portal, as of 10 September. Weekly changes are measured against the levels published in edition #7. Analysis and personal opinion — never investment advice.

Kevin Warsh, who runs the US central bank, has spent two weeks warning that inflation is not coming down fast enough. On Thursday one of his own colleagues said the exact opposite in public.

Christopher Waller who sits on the Fed board and votes on interest rates said he would rather leave them where they are this month. Markets believed him. The dollar fell to a four-month low, and traders cut the odds of a September rate rise from about 60% to something close to 50-50.

Friday reversed these expectations. The official jobs report showed 162,000 new jobs in August against a forecast of 53,000, which was the best month since March, and the odds climbed back to 58%.

This report and the immediate change is the main story of this week. Not because of what it says, but because two other measurements of the same month say something quite different.

The Scoreboard

The board below is the week in nineteen numbers, and two of them stand out.

Oil moved further than anything else on the board, and it is the number to keep in mind while reading the rest of this letter. Fuel goes into making things, moving things and heating buildings. So when it climbs, prices climb behind it a few months later. That is why three central banks are now leaning towards higher interest rates at the same time.

The yen also went up. It has spent most of this year going the other way, through two interventions and a good deal of official worry, so a week of quiet strengthening is worth the notice.

The US equities finished the week where they began. Thursday’s rally and Friday’s fall cancelled out each other.

Europe had a worse week, and Tuesday was the worst day of it. Euro area inflation came in at 3.3%. German thirty-year borrowing costs reached their highest level since 2011. The Euro Stoxx 50, which tracks the fifty largest listed companies in the euro area, dropped to a one-month low and ended the week down 0.9%.

Asia fell even further. Japan’s ten-year government bond yield touched 3.00% on Tuesday, a level it last reached in 1996. The Nikkei lost 2.1% over the week and Korea’s main index 1.5%.

Those are three regions and a dozen numbers, all given equal space in a table. The chart below gives them the space they earned, and one of them takes up most of it.

The large green circle is Brent crude, the oil price most of the world quotes. It rose 7.6% to $96.40 a barrel after tensions in the Persian Gulf worsened again. US petrol should pass $4 a gallon this weekend.

Most of this letter comes back to that number. But why is that? Simply put: Energy costs feed into nearly every other price, which is why euro area inflation rose, why the European Central Bank is about to act, and why central bankers sounded more nervous this week than last month.

1. Three counts of one month

The government says American hiring got faster in August. Two other measurements disagree.

Let’s start with the official figure. The Bureau of Labor Statistics is the US government’s statistics office. On Friday morning it reported 162,000 new jobs for August, against a forecast of about 53,000 from a survey of bank economists.

It also revised July, which had shown a loss of 23,000 jobs, into a gain of 23,000. Unemployment stayed at 4.1%. Of those 162,000, private companies accounted for 127,000. The rest was government hiring. Now let’s look into the second count.

ADP is a payroll company. It handles wage administration for roughly a fifth of American private-sector workers, which means it can see how many people were actually on payrolls without asking anyone. Its August figure, published on Wednesday, was 38,000 new private jobs.

So what does that actually tell us? Two organisations counted the same weeks and landed 89,000 apart. Neither is inventing anything.

The third count is narrower but points the same way. Challenger, Gray & Christmas tracks layoffs that companies announce publicly. August brought 52,881 of them, against 33,429 in July, which signals a 58% rise in the month the official data says hiring sped up.

1.1 Why all three can be right at once.

The government’s headline number is seasonally adjusted. Some months always add jobs and some always lose them. Shops hire for Christmas. Schools stop paying staff over the summer. Seasonal work ends in August. So rather than publish the raw count, the statistics office compares each month against what that month normally does, and reports the difference. Without that step, the figures would swing wildly every year for reasons nobody needs to think about.

Now look at where August’s gains came from. Restaurants and bars added 59,000 jobs. Hotels and hospitality added 62,000. That happened in the month when summer work traditionally ends.

Taken at face value, that is a hiring boom in kitchens. Measured against the seasonal pattern, it more likely means employers cut fewer summer workers than expected — because they never took on the usual number in the first place.

The adjustment assumes companies hire and fire the way they always have. This year they have done far less of both. And every month that continues, the headline number rests on a shakier assumption.

None of this makes the report wrong. It makes it weaker evidence than 162,000 suggests, in a week when that one figure moved the entire bond market.

2. The Fed disagreed with itself in public

One governor wants to hold. The chairman has spent two weeks arguing the opposite. Both vote in eleven days.

Warsh gave his first speech as Fed chair last month at Jackson Hole, the annual conference where central bankers say the things they have avoided all year. His message: inflation is not improving, and money is still too cheap. Markets heard someone preparing to raise rates.

Waller said the opposite on Thursday. He would rather leave rates at 3.50–3.75% this month, provided inflation keeps easing.

Central bankers disagree often. It is less common for the second-most-watched person at the institution to reject the first one’s case in public, two weeks before a vote.

You can measure what that did, because people bet on Fed decisions. They do it through futures contracts, and the price of those contracts converts into a probability.

Before Waller spoke, the chance of a September rise was about 60%. By Thursday evening it was even. After Friday’s jobs report it was 58%. Bloomberg’s version of the same calculation reached 63%.

Twelve people vote on 15 and 16 September, Warsh and Waller among them. The evidence that settles it arrives first: US inflation figures on Friday 11 September.

The Fed is built to argue, so none of this is a crisis. It does mean nobody knows the answer yet, including the people who will decide. That is why a single data release moved expectations by ten points twice in two days.

3. The yen turned, and nobody had to buy it

Two interventions failed. The prospect of a rate rise did it in one week.

This letter has followed the yen since late July, when Japan and the US bought it together for the first time since 1998.

Buying your own currency to hold its price up is called intervention. A government sells foreign reserves and buys its own money, which lifts the price for as long as the buying lasts. Japan did it alone on 30 July and together with the US the day after. It held for about two weeks. Then the yen slid back through 160 to the dollar last week.

This week it rose 2.4%, to 156.26. On Friday it briefly touched its strongest level since 3 August. Nobody bought anything.

What changed is what people expect. Reports emerged that the Bank of Japan will raise its interest rate by a quarter point at its meeting on 17 and 18 September, and may go further after October. The market now puts the chance of that at 77%.

3.1 Why a Japanese rate rise moves the currency more than buying it does.

Japan’s official rate is 1.0%, but Japanese inflation runs higher than that. Subtract one from the other and the real cost of borrowing in Japan is around minus 0.9%. Money there is still effectively free, even after several increases. The central bank has room to keep raising without damaging its own economy.

Every rise also narrows a gap that has shaped markets for thirty years. Investors borrow yen at almost no cost and buy assets elsewhere that pay more. Raise Japanese rates and that stops being worth doing, so some of it gets unwound.

Japanese money is already coming home. The ten-year yield touched 3.00% on Tuesday, and a thirty-year auction this week met strong demand from Japanese pension funds, who bought their own government’s debt instead of sending the money abroad.

Both failed interventions taught the same lesson. Buying a currency moves the price for a few weeks. Only the interest rate changes anyone’s reason to sell it.

Data Check

Four economic releases landed this week that were worth watching. Two of them came in far above what economists expected, and both point the same way — towards a central bank that raises rates rather than holds.

The table below shows what was forecast and what actually arrived. The gap between those two columns is what moves markets, not the number itself.

The US services survey needs a short explanation. Every month the Institute for Supply Management asks purchasing managers at service companies one question: is business better or worse than last month? Fifty is the dividing line, so anything above it means growth.

August came in at 55.4, the second-highest reading since October 2024. The detail underneath is messier. New orders were the strongest since February 2023, which is good news. Employment came in at 47.8, meaning service companies cut staff for a second month running. Prices paid reached 72.6, the highest since July 2022.

Growing, hiring less, charging more. That combination is exactly what a central bank worried about inflation watches for.

The Credit Grind

Europe’s riskiest borrowers had their worst week in months. They now pay the same as America’s, for the first time this year.

A company that borrows pays two things added together.

The first is whatever the government pays. That sets the floor, because no company borrows more cheaply than the state it operates in.

The second is a premium on top, because companies go out of business and governments almost never do. The riskier the company, the larger the premium. That premium is what this section follows.

The table below has six rows and one job: to show what borrowing costs right now, and whether that is high or low compared with the past year.

The top two rows are governments. The four underneath are companies, sorted from riskiest to safest. And the last column is the one to read first — it tells you how unusual today’s number is.

Look at the first company row. European high yield means the debt of companies rated least likely to repay. It rose 9 basis points this week, to 265.

A basis point is a hundredth of a percentage point. So 265 means those companies now pay 2.65 percentage points more than governments do.

Nine basis points is the largest move on the board. It carried European high yield from the 6th percentile of its own year to the 34th. That final column counts how many days over the past twelve months were cheaper than today, so 34% means roughly a third of the year was lower. Two weeks ago the same column read 6%.

Now the row below. US high yield sits at 265 as well. That has not happened once this year. Europe has been the cheaper of the two for months — 19 basis points cheaper three weeks ago, 7 last week, and nothing at all now.

The two government rows above explain why. Both reached fresh twelve-month highs this week, up 10 and 9 basis points, because the ECB is expected to raise rates on Thursday. European companies were hit from both directions: the floor went up, and the premium on top went up with it.

US companies were not. Their high yield row moved 2 basis points and remains in the 2nd percentile — cheaper than 98% of the past year.

What this means if you own a corporate bond fund.
You are lending money to companies, so you should be paid for the risk that some of them fail. A fund holding European debt paid you more this week than last, which has not been true for a while. A fund holding US debt is still paying you close to the least it has all year, and the Fed may raise rates in eleven days.

The Onchain Pour

Bitcoin held $80,000 through a week that gave it every reason not to.

Bitcoin finished around $80,000, up under 1%. It crossed that level on Thursday, when Waller’s comments pushed the dollar to a four-month low. Tuesday had looked very different: $76,600, on the day Japan’s ten-year yield touched 3% and almost every risky asset in the world fell together.

That connection explains something that otherwise looks like coincidence. Cheap Japanese money has funded bets on higher-paying assets for thirty years, and crypto is one of them. When borrowing in Japan stops being nearly free, some of those bets get closed. The selling then shows up in places with no obvious link to Japan.

US bitcoin funds still took in $986.8 million across the week. The summer pattern holds: institutions keep buying while the price moves sideways. Ether recorded its third-best third quarter.

One oddity worth noting. Five wallets, each holding 50 bitcoin and untouched for sixteen and a half years, moved their coins this week. They date from 2010, when bitcoin traded for fractions of a cent and 50 coins was the reward for mining a single block. Each wallet is worth about $4 million today.

The Week Ahead

Thursday is the day that matters. The European Central Bank is expected to raise interest rates.

Monday 7 September.
German industrial production at 08:00, forecast 0.3%. Euro area growth for the second quarter is revised again at 11:00 and should confirm 0.4%.

Tuesday 8 September.
Japanese wage growth at 01:30, forecast 3.9%. The Bank of Japan watches this more closely than almost anything else before deciding. Chinese trade figures at 05:00 — exports rose 23.9% last month, and a slowdown there would matter for factories everywhere. German and French trade balances at 08:00.

Wednesday 9 September.
Chinese inflation at 03:30, forecast to slow to 0.7% from 0.9%. China is still exporting falling prices while everyone else imports rising ones. Christine Lagarde, who runs the ECB, speaks at 19:00 — the evening before her own decision.

Thursday 10 September — the one that counts. The ECB decides at 14:15.
The deposit rate is expected to rise to 2.50% from 2.25%. That rate is what banks earn for leaving money at the central bank overnight, and it sets the floor for what everything else in the euro area costs. Markets have almost fully priced the increase since inflation reached 3.3%. The press conference at 14:45 matters as much as the decision, because the real question is whether the ECB keeps going afterwards. US producer prices and jobless claims at 14:30.

Friday 11 September — US inflation at 14:30.
Forecast is 3.4% for the year and 2.4% for the core measure, which leaves out food and energy. This is the last major evidence before the Fed votes, and given how openly Waller and Warsh disagree, it may decide the outcome. UK growth figures at 08:00.

☕ Last Sip

Three central banks meet within eight days of each other, and the same thing is pushing all of them the same way: an oil price that will not stop rising.

Europe goes first on Thursday. America and Japan both decide the following week, almost at the same hour. Between now and then, the only new evidence is Friday’s US inflation report.

It has been a while since three of the world’s largest central banks were all leaning towards higher rates at once. You have to go back to 2022.

Enjoy the first proper weekend of September. See you next Sunday for the next episode of The Weekly Steep.

Sources

Market levels: CNBC and Investrade for the US closes; Trading Economics for European and Asian indices and yields; Bloomberg for the euro area government curve; Zaikei Shimbun and MyForex for the New York close in dollar-yen and euro-dollar; News On Japan for the Nikkei. Economic releases: US Bureau of Labor Statistics, ADP, Challenger Gray & Christmas, Institute for Supply Management, Eurostat, Trading Economics calendar. Credit and euro curve: ICE BofA indices via FRED and the ECB Data Portal, from my own pipeline. Onchain: Cointelegraph and CoinGecko. Weekly changes are measured against the levels published in edition #6. Analysis and personal opinion — never investment advice.

Wednesday’s inflation report missed by one tenth of a percentage point. The Federal Reserve’s preferred measure came in at 3.7% for the year against 3.6% expected — a rounding error by most standards.

It was enough. Gold, which had spent a month rising on worries about government debt, fell below $4,600 and closed the week at $4,506, down 3.4% on Friday alone. Bitcoin, which had reached $81,235 that morning, was under $78,000 by the American open.

Then Friday brought Kevin Warsh to Jackson Hole, in his first keynote as Fed chair. He said inflation indicators do not suggest the trend has meaningfully improved, and that describing financial conditions as restrictive is difficult. Traders moved the odds of a September rate rise from 35% to 57% in a single afternoon.

Underneath all of it, Nvidia reported the strongest quarter of its life and added $442 billion of market value in a day.

The Scoreboard

Look at the two ends of the American yield curve. They moved in opposite directions, and they swapped places with last week.

American shares finished higher for the week despite falling on Friday. The S&P 500 was up 0.5%, the Nasdaq 0.9%, and all that while smaller domestic companies dropped 1.4% during the week.

The two-year government borrowing rate rose 16 basis points while the thirty-year fell 6. That move is the exact opposite of what happened last week. Korea gave back the whole week on Friday alone, they are down 1.8% as foreign investors sold 1.76 trillion won of equities.

But the most interesting line on the board is the yen. On Friday it fell back through 160 per dollar which was the fifth falling day in a row. On 31 July, Japan and the United States bought yen together to stop it falling. They had not done that since 1998. The yen got stronger for about two weeks, but it has lost all of it already again. But why should a European reader even care about a Japanese currency that much you might ask?

Japan has kept interest rates near zero for thirty odd years. Money is cheap there and expensive almost everywhere else. So for decades, investors have borrowed yen cheaply and used it to buy things that pay more. Whether that was American government bonds, European bonds, shares, almost anything that yields more than owning a falling currency. Japanese savers and pension funds did the same with their own money. Japan became the world’s biggest lender.

That works while the yen keeps falling, because you repay your loan in a currency worth less than when you borrowed it. It’s like shorting a company that keeps on delivering bad earnings releases. A no brainer one might think. But what happens when the yen turns around, and then everyone tries to unwind the same trade at once?

They sell what they bought and buy yen to repay. That is why a currency most people never think about can move share prices in New York and bond prices in Frankfurt.

Two things pushed it down this week. Warsh made a US rate rise more likely, and higher American rates pull money out of yen and into US-dollars. Secondly Japan’s finance ministry said it may plan its 2027 budget around paying 3.8% interest on its debt, up from 3%. This would mark a government getting ready for its own borrowing to cost much more.

Above 160, people watch for another intervention. Treasury Secretary Bessent wrote to Senator Warren this week that big, messy moves in the yen would be bad for world markets. Nobody has said what level would make them act.

That is one line on the board. Here is the whole board at once, with each asset drawn as big as its move.

The two largest circles are both commodities, and both are red. Oil fell 4.5% as Iran and Oman kept talking about the Strait of Hormuz. Gold fell 3.5% for a completely different reason and that reason is the first chapter that we cover shortly.

Notice what is missing: any big green circle among the equity indices. The four stock markets on the board barely moved. Everything that did move was a currency, a commodity or a crypto asset. That is what a week looks like when the news is about central banks rather than companies.

1. The trade that needed soft inflation

Last week’s letter named the condition that would break the gold-and-bitcoin story. It broke three days later.

Two weeks ago this letter argued that gold and bitcoin were rising for the same reason: investors doubting that governments could handle their debts without inflating them away. It closed with a test to watch whether the two keep moving together, because if they separate, the story was doing less work than it looked.

On Wednesday the Bureau of Economic Analysis reported that the price index the Fed actually targets rose 3.7% over the year, against 3.6% expected. The core measure, which strips out food and energy, came in exactly on forecast at 3.3%.

One tenth of a percentage point, on the headline number only. Gold broke below $4,600 within the hour. Bitcoin gave up $3,000. By Friday, after Warsh spoke, gold had fallen another 3.4% to $4,506, its worst day in months.

But what does this tell us? If the trade were really about governments debasing their currencies over years, a one-tenth miss on a monthly inflation print should not touch it. What actually happened is that both assets were being bought on the expectation of falling interest rates. Higher rates make holding something that pays no interest more expensive, and both gold and bitcoin pay no interest and no dividends

So the trade was real, but its reasoning was shorter-term than its own story suggested. Worth remembering the next time a market move arrives with a grand explanation attached.

2. Warsh, and the curve that turned around

The Fed chair gave no guidance and moved the market anyway.

Last week’s letter described a market worried about government debt: long-term borrowing costs rising, short-term ones falling. This week it did the exact opposite and the reason was the Fed’s chairman.

Warsh spoke at Jackson Hole on Friday afternoon. He committed to nothing and declined to say what would make him move. But he said inflation data do not show meaningful improvement, that if inflation does not head to 2% quickly there is work to do, and that it is hard to call current financial conditions restrictive.

You can watch the market change its mind in real time, because people bet on this. Polymarket runs a market on the September Fed decision with $66 million of money in it — anyone can buy a share that pays out if the Fed raises rates, and the price of that share is the crowd's probability.

At two o’clock on Friday afternoon, before he spoke, the crowd gave a rate rise a 28% chance. By two in the morning it was 50%. As of Sunday morning it sits at 47%, against 51% for no change at all.

The futures market, where banks hedge rather than bet, moved even further. From about 35% to 57%.

The two-year yield, which mostly reflects what people expect the central bank to do, rose 16 basis points on the week to 4.34%. The thirty-year, which depends more on how much debt is being sold and who wants it, fell 6 basis points to 5.21%.

So what does that mean? Higher short-term rates slow the economy, and a slower economy means less inflation later. So the market is saying: tighter now, calmer later. It is the opposite of last week, when the message was that nobody wanted to lend long at any price.

But note where it landed. Not at “he will raise”, just at “it could go either way”. One speech turned a settled question into a coin flip, and Friday’s jobs report will most likely decide further steps.

3. Nvidia had the best quarter of its life

$96 billion of revenue, a forecast of 70% growth, and $442 billion added in a single day.

Nvidia reported on Wednesday after the close. Revenue of $96.22 billion, more than double a year earlier. Adjusted earnings of $2.22 a share. And guidance for roughly 70% revenue growth in the coming financial year. These numbers are just incredible to say the least.

Based on these news the stock rose nearly 9% on Thursday, adding $442 billion of market value. This was the largest single-day gain for any company since April 2025. That one move lifted the entire Nasdaq 100 by 1.4%.

Then on Friday it fell more than 3%, because Warsh spoke.

That sequence is the whole argument about this market in miniature. The company that carries the index delivered everything anyone could have asked for. But still, it was not enough to outweigh a central banker declining to rule out a rate rise. When borrowing costs go up, profits arriving years from now are worth less today. And Nvidia’s valuation rests almost entirely on profits arriving years from now not today.

4. France is where Europe is nervous now

Not Italy, not Greece. French banks fell 5% on Thursday over the country’s public finances.

Two weeks ago this letter noted, almost in passing, that France is not usually where European bond stress shows up first. It is now showing up there in equities as well.

On Thursday the CAC 40 fell about 1.7% to a one-month low while the broader European index dropped 0.7%. BNP Paribas, Société Générale and Crédit Agricole each lost roughly 4% to 5%. The trigger was a growing doubt about whether France can control its budget deficit while candidates for next year’s presidential election promise more spending.

Banks fall first in this kind of worry because they hold large amounts of their own government’s debt. If those bonds lose value, the banks holding them do too.

France’s ten-year borrowing cost touched 4.10% earlier this month, its highest since 2009. Germany’s sits at 3.29%, the highest since 2011. Meanwhile the gap between Italy and Germany finished the week at 82 basis points and barely moved. The country the market used to worry about is now the calm one.

Data Check

Every inflation report comes in two versions. The headline counts everything people buy. The core version leaves out food and energy, because those two jump around for reasons that have nothing to do with the wider economy. Let it be a cold winter, a war, a bad harvest.

Central banks watch the core number, because it shows the trend they can actually influence. So on paper, Wednesday was fine: core came in at 3.3%, exactly as forecast.

The headline came in at 3.7% against 3.6% expected. And that is the number the market traded.

The reason is simple enough. Households do not get to exclude petrol and groceries from their own budgets. Neither do the people setting wages, and neither do the traders trying to guess what a new Fed chair will do at his first meeting. When a central banker has spent the month sounding worried about inflation, a hot headline is what he gets asked about.

One tenth of a point, on the number that officially matters less, and it took 3.5% off gold and $3,000 off bitcoin.

A second number, quietly good.
Two weeks ago this letter flagged Friday’s annual revision to American job figures. This is the exercise where the statistics office goes back and corrects a year of its own counting. Last year it removed 911,000 jobs after the fact, which was one reason nobody trusted the labour data. This year it removed 79,000. A tenth as much.

The Credit Grind

Lending to America’s weakest companies pays less right now than on any day in the past year. In the same week the Fed chair said rates may go up.

When a company borrows money, the rate it pays consists of two things.

First, whatever the government pays. Governments are the safest borrowers there are, so their rate is the floor. Nobody borrows cheaper than the state they sit in.

Then you add a premium on top of that base rate. A company can go out of business and stop paying. A government almost never does so. That’s why lenders want more to take on that kind of risk. A large, solid company might pay a little extra. A struggling one pays a lot. That extra payment is what this section follows every week.

The size of it tells you two things at once. It says how likely lenders think it is that they will not get their money back. And it says how badly they want to lend at all. When there is more money looking for a home than there are potential borrowers, lenders compete, and competing means accepting less compensation.

This week the second reason (the competition among lenders)) is the one to watch. The premium did not fall because these companies got safer. Nothing about them changed. It fell because there is more money looking to lend than there are companies looking to borrow, so lenders simply undercut each other.

Now let’s look at the second row: America’s weakest companies. They now pay 263 basis points extra, which is 2.63 percentage points on top of the government rate.

The last column says 0%. That column tells you how many days in the past year were lower than today. Zero. None of them. In a whole year of trading, this is the day lenders asked these companies for the smallest extra payment.

Now look at the two rows above. Both are at 98%, almost the highest of their year. Those are European governments. So in one week: Europe’s safest borrowers are paying near their yearly high, and America’s riskiest are paying their yearly low.

One thing I got wrong. Last week the American investment-grade row was at 78%, and I wrote that borrowing trouble was spreading to companies. It is back at 45% now. That reading lasted seven days.

What this means if you own a corporate bond fund.
You are lending to companies, so you should be paid for the risk that some of them fail. Right now almost nothing you earn comes from that. It comes from the base rate. So essentially the part you would get from a government bond anyway. Simply put: you are taking on company risk and being paid government prices.

The Onchain Pour

Bitcoin ended the week higher, having been $1,700 higher still on Wednesday morning.

Bitcoin closed around $79,500, up 2.9% on the week. That number hides a round trip: it reached $81,235 on Wednesday before the inflation report, fell below $78,000 within hours, then recovered.

The month as a whole was remarkable. Bitcoin was near $62,800 at the start of August and peaked about 28% higher. Ether finished around $2,505, up 2.8%.

Ether funds took money in on ten consecutive trading days. Bitcoin funds broke their own streak on Friday with $201 million of net outflows, the first day of selling after the run.

Two structural items. Solana’s validators voted to double the rate at which new coin issuance falls, cutting an estimated 18.9 million coins from supply over six years. And American M2 money supply, which stands for the broad measure of cash and deposits, rose $102.8 billion in July to a record $23.22 trillion, its 27th consecutive monthly increase.

That second number is the entire argument for holding scarce assets, still running quietly in the background while the price of holding them went up this week.

The Week Ahead

This week has one question running through it, and every release either supports Warsh or undermines him. He argued on Friday that inflation is not falling fast enough and that money is still too easy. The week’s job is to check both halves of that.

Monday 31 August.
China’s factory survey at 03:30, expected to stay below 50 — the line between growing and shrinking. China’s factories set prices for a lot of what the rest of the world buys, so a weak reading is mild good news for inflation everywhere else. German inflation for August at 14:00, forecast to rise to 2.9% from 2.8%.

Tuesday 1 September — the European one.
Euro area inflation at 11:00, forecast at 3.2% against 2.9% last month. That would be the highest reading of the year, and it arrives two weeks before the ECB meets. If it comes in at or above forecast, the argument in Frankfurt shifts from holding rates to raising them, and every European mortgage and business loan sits downstream of that. US job openings at 16:00 (consensus 7.39 million) is the first of four labour readings this week.

Wednesday 2 September.
Private payrolls at 14:15 (consensus 59,000) — a private company’s count of American hiring, published two days before the official one and often the first hint of it. The Bank of Canada decides at 15:45, expected to hold at 2.25%, in the week its largest trading partner put 50% tariffs on some of its goods. The Fed’s survey of regional business conditions at 20:00 is worth more than usual: it is written by the same regional presidents who will vote in three weeks.

Thursday 3 September.
The US services survey at 16:00 — the sector that produced last week’s upside surprise, and the one that employs most Americans. Jobless claims at 14:30 (consensus 203,000). Then two Fed officials speak, Waller at 14:30 and Hammack at 21:00, both with a chance to tell us whether Warsh was speaking for himself or for the committee.

Friday 4 September — the one that counts. US payrolls at 14:30.
Consensus is 45,000 new jobs after last month’s loss of 23,000, with unemployment ticking up to 4.2% from 4.1%.

This is where the week resolves. A central bank raising rates into a weakening labour market is a very hard thing to justify in public. Weak numbers and Friday’s whole move reverses. Strong numbers and Warsh has what he needs for 16 September. Canada reports its own employment figures at the same hour.

☕ Last Sip

The week’s lesson is about how quickly a story can change owners. For most of August the market worried that governments were borrowing too much. By Friday it worried that a central bank would raise rates instead.

Both worries are real, and they pull in opposite directions for anyone holding bonds. Too much government debt makes long-term lending expensive. A central bank raising rates makes short-term lending expensive. This week the second one won the argument, and the shape of the yield curve simply flipped over to match.

There is a reason it can turn that fast. Neither worry is a fact yet. Nobody knows whether Warsh will actually raise rates in three weeks, and nobody knows how much debt the world will absorb before it demands a higher price. Markets have to put a number on both anyway, every day. So the number moves whenever someone credible says something new. Last week the most credible person in the room spoke for the first time in that job.

Friday brings the American jobs report, and it will settle rather a lot. Weak numbers make a September rate rise hard to justify and the whole week’s move goes into reverse. Strong numbers and Warsh has what he needs.

Enjoy the last days of summer. See you next Sunday for the next episode of The Weekly Steep.

Sources

Market levels: CNBC and Yahoo Finance for the US closes; Trading Economics for European and Asian indices and yields; Zai Diamond and Zaikei Shimbun for the New York close in dollar-yen, euro-dollar and gold; Korea Times and Asia Business Daily for the KOSPI. Economic releases: US Bureau of Economic Analysis, US Bureau of Labor Statistics, Eurostat, Trading Economics calendar. Credit and euro curve: ICE BofA indices via FRED and the ECB Data Portal, from my own pipeline. Onchain: Cointelegraph. Weekly changes are measured against the levels published in edition #5. Analysis and personal opinion — never investment advice.

On Wednesday morning the US Treasury announced it would at least double the size of its bond buybacks. This are the operations where the government buys back its own debt from its investors. As a result of these actions the long-term borrowing costs fell immediately. The thirty-year rate dropped about ten basis points.

But by Thursday the entire move had reversed. By Friday the thirty-year closed at 5.27%. This marks its highest weekly close since June 2007.

Two things went up instead. Gold rose more than 5%, with its fifth straight weekly gain. Bitcoin rose 22%, its best week in two years. Equities fell. When investors sell government bonds and buy things outside the fiat-backed system, they are usually saying something about their confidence in the government doing the borrowing.

Then, right after the week had ended: trade talks between the United States and Canada collapsed shortly before a midnight deadline on Friday, and 50% tariffs took effect on about $20 billion of Canadian goods. Canada has promised to match them dollar for dollar.

The Scoreboard

This week everything a government issues had a bad week. Everything a government cannot create had a very good one. That’s the major insight.

Reading the board. American equities fell for the first week since late July. The S&P 500 down 1.4%, the Nasdaq 2.1%, despite rising on Friday. Long-term government borrowing costs rose while short-term ones slipped, which is the shape of a market worried about debt.

Europe followed the same pattern. Germany's ten-year yield reached its highest since 2011 during the week, and France's touched 4.10% on Tuesday, a level last seen in 2009.

Japan’s Nikkei index fell 4.2%, the largest equity move on the board, after its own ten-year yield hit a thirty-year high mid-week.

But the interesting part is what rose instead: Brent crude up 6.2%, gold up 5.4%, bitcoin up 22.5%, ether up 30%. Oil had its own reason in the Iran conflict. The other three share one: nobody can create more of them by decision.

But now let’s dive deeper.

1. The intervention that lasted a day

The Treasury stepped into its own bond market on Wednesday but the market handed the money back by Thursday.

Let’s start with what a buyback is. The government has already sold bonds that will not mature for decades. A buyback means it goes into the market and buys some of them back early. But why would they do that, you might ask? It’s quite simple: Fewer bonds available means a higher price for the ones that remain, and a higher price means a lower yield since the two always move in opposite directions.

On Wednesday the Treasury said it would at least double these operations for bonds maturing in ten to thirty years, running from 9 September to 4 November. As a consequence bond yields with a long maturity fell straight away.

That sounds good generally speaking. But the problem is more complicated than one might grasp straight away, because who wants to finance this relocation? There really is only one way. The Treasury has to raise that money elsewhere. It does so by issuing more short-term bills. So the total debt does not shrink at all - it simply gets shorter. And short-term debt has to be refinanced sooner, at whatever rate happens to prevail then.

Investors worked this out within a day. The market’s own measure of expected inflation rose to its highest in more than two months, meaning traders read the operation as more likely to create inflation rather than to reduce borrowing costs. The thirty-year yield climbed straight back to where it had been before the announcement.

Secretary Bessent said afterwards that he has “a big tool kit”. This week the market tested one of the tools and gave it about twenty-four hours.

2. This is not an American problem

Germany, France and Japan all hit multi-decade highs in the same week.

It would be very easy to read this past week as a story about American deficits. Unfortunately the rest of the world faces the same problem.

Germany’s ten-year yield climbed above 3.25%, its highest since March 2011. France’s touched 4.10% on Tuesday, the highest since June 2009 — and France is not usually where European bond stress shows up first. Japan’s ten-year yield reached just under 3%, a thirty-year high, in a country that spent most of those thirty years with interest rates near zero.

The common factor is not any one government’s budget. Rich countries as a group are borrowing more, at the same time, into a market with a finite appetite for long-dated debt — while the central banks that spent a decade buying that debt have stopped. And governments are no longer the only ones queueing. American companies have issued almost $1.7 trillion of bonds this year, 27% more than last year, much of it to build AI capacity. That money competes for the same buyers.

If you are a European reader this matters more than you might think. The German ten-year yield is the reference rate for the whole euro area. When it rises, so does what Italy, France and Spain pay, and so eventually does what banks charge for mortgages and business loans.

3. Gold and bitcoin did the same thing this week

Two very different assets rose hard in the same days that government bonds fell and this is no coincidence.

Gold finished around $4,671 an ounce, up more than 5% and its fifth consecutive weekly gain. Bitcoin rose about 22% to roughly $77,240, touching $79,000 on Friday — its best week in two years. Both moved while equities fell and while government borrowing costs hit multi-decade highs.

Generally speaking, a government with a large debt has two ways out that do not involve raising taxes or cutting spending: borrow more, or let inflation quietly reduce what the debt is worth. Investors who think the second is likely want to hold things whose supply nobody controls. Gold has been that for thousands of years. Bitcoin was designed to be it.

On Friday one of the largest investors in the world said exactly this out loud. Ray Dalio, who built Bridgewater into the biggest hedge fund on earth, wrote that America’s finances are “at an inflection point” and told investors to hold fewer bonds and put 10% to 15% of their money into gold, plus “a bit” of bitcoin.

His point being: about $5.5 trillion of government revenue this year against $7.5 trillion of spending, roughly $1 trillion of that going on interest alone, and some $10 trillion of existing debt that has to be refinanced. His guess at when this breaks is three years, give or take two. What makes the timing striking is what he named as evidence: Japan selling American bonds, long-term yields at multi-year highs, and Bessent’s buyback announcement. Those are the first two chapters of this letter.

Take the recommendation with the caveat it deserves. Dalio has spent years arguing bitcoin cannot replace gold and warning about its privacy and security risks. “A bit” is doing real work in that sentence.

Bitcoin also had reasons of its own this week. Exchange-traded funds took in $1.67 billion over five consecutive days, BlackRock alone bought more than a billion dollars of bitcoin and ether in two days, and Washington made a renewed push to pass crypto market rules. Those would have lifted the price whatever bonds did.

Watch whether gold and bitcoin keep moving together. If they separate, the debasement story was doing less work than it looks.

4. The economy refuses to give one answer

Last week the consumer looked broken. This week business activity hit a four-year high.

Last week’s letter reported that American retail sales fell 0.6% in July and consumer confidence came in at 51 against a forecast of 55. The picture was of a household sector running out of road.

This week’s data says something else. The flash survey of American business activity for August came in at 56.0 against a forecast of 54.0, the strongest reading since 2022, with services doing the work at 56.8 against 54.0 expected. Manufacturing was the exception, slipping to 53.2 against 53.9 expected, though anything above 50 still means growth.

And new claims for unemployment support fell to 206,000, a level the American labour market has not been far from all summer.

That last number needs a word of explanation, because it looks impossible next to a month in which the economy lost jobs. Initial claims count people who have just been laid off. They measure firing, and firing alone. The payroll figure is a net number: everybody hired, minus everybody who left for any reason.

Both can be low at the same time. If companies stop firing but also stop hiring, jobs disappear quietly. Let it be through retirements, through people leaving voluntarily, through graduates who never find a first position. None of those people can file a claim. Someone who quits generally has no entitlement, someone who retires has none, and someone who has never worked has nothing to claim against.

So the picture is not a strong labour market and a weak one at once. It is a market where almost nothing is happening in either direction. Few people losing jobs, few people finding them. That is a fragile kind of stability, and it is why one bad quarter for company earnings would show up faster than usual.

Data Check

These are survey numbers, so it is worth knowing what they actually measure. Every month S&P Global asks purchasing managers at thousands of companies one simple question: is activity higher or lower than last month? The answers get turned into a single figure where 50 is the dividing line. The August readings below are "flash" estimates, published from roughly the first four fifths of responses, which is why they arrive two weeks before the final version.

The services figure was the surprise. At 56.8 against 54.0 expected, it is the strongest American business survey since 2022 — and services is where the large majority of Americans work, which makes it the more consequential of the two. Manufacturing went the other way, at 53.2 against 53.9, though it is still comfortably in growth territory.

Europe grew too, more modestly, at 52.1. The composition there is the opposite of America’s: factories did the work, largely in Germany, while services barely improved.

Keep one limitation in mind. A survey records how business feels to the people running it. It does not count anything. Last week’s retail sales figure counted actual purchases and fell 0.6%. Both numbers can be accurate at the same time, which is exactly the tension the chapter above describes.

The Credit Grind

The safest government bonds in the euro area now yield more than at almost any point in the past year. The riskiest companies in Europe are paying almost the least.

A borrowing company pays two prices added together: the base cost of money, set by what governments pay, and a surcharge for the chance it fails to repay. This week those two moved in opposite directions, which is the whole point.

Look at the top two rows. Both euro government maturities sit at the 99th percentile of their own year, meaning there was only one day in the past twelve months when lending to a top-rated European government paid better than it does now. Both are within a basis point of their twelve-month high, and both got there in a single week — up 10 and 12 basis points.

Now the bottom four. Euro high yield tightened again to 256 basis points, the 5th percentile of its year: investors were paid more than this on 95% of the past year’s trading days. Compare that with the American rows. US investment grade — the debt of large, solid companies — jumped from the 48th percentile last week to the 78th. US high yield went from the 9th to the 23rd.

So the bond stress is reaching American corporate borrowers and has not yet reached European ones. Europe now pays 19 basis points less than America for junk-rated debt, the widest gap in weeks.

If you hold a European corporate bond fund, nearly all your yield now comes from the government part, and that part just got more valuable. The compensation for company risk is close to the thinnest of the year.

The Onchain Pour

Bitcoin’s best week in two years arrived exactly where last week’s warning signal said a bottom might form. And the way it happened matters more than the size of it.

Last week’s letter noted that only 51.4% of all bitcoin was worth more than its owner had paid, the weakest reading since 2023. That level has historically appeared near market lows, and also shortly before further falls. This time it was the first.

Bitcoin ended around $77,240, up roughly 22% on the week. Ether rose about 30% to near $2,438. Solana passed $100 for the first time since 3 February.

Here is the detail worth having. Open interest — the total value of outstanding bets on bitcoin’s future price — fell about 12% over the week and now sits at a five-month low, while the price rose 20%. Those two normally move together, because rallies usually come with more borrowed money chasing them. This one came with less.

That reframes the $5 billion of leveraged positions liquidated over three days. Most of those were bets that the price would fall, and clearing them out left a market holding more coins and less debt. A rally built on people actually buying is a sturdier thing than one built on borrowed money, even if it is no guarantee of anything.

The caveat stands anyway: bitcoin now shows its most overbought reading since November 2024, which describes how far and how fast it has come, not where it goes next.

The Week Ahead

Monday 24 August.
Quiet on the data. The White House is expected to publish details of what President Trump has called an “economic D-day” against Iran — oil rose more than 5% last week on the anticipation alone.

Tuesday 25 August.
German business confidence at 10:00, forecast to improve to 87.2 from 86.6. US consumer confidence at 16:00 (consensus 91.2) is the first read on whether the collapse in sentiment continued into August. Then a two-year debt auction at 19:00 — the first of three this week, and the real test of whether last week’s buyback changed anything.

Wednesday 26 August — the one that counts. The July personal consumption expenditures price index at 14:30.
This is the inflation measure the central bank actually targets, rather than the consumer price index that gets the headlines. Consensus is 0.2% on the core measure for the month and 3.3% for the year. Second-quarter growth is revised at the same time, forecast down to 1.5% from 2.1%. A five-year auction follows at 19:00. And after the close, Nvidia reports earnings — the largest company in the index and the clearest read on whether the AI investment boom is still accelerating.

Thursday 27 August.
Jackson Hole opens, the annual gathering where central bankers say the things they have avoided all year. The ECB publishes the account of its July meeting at 13:30. Jobless claims at 14:30 (consensus 209,000), then a seven-year auction at 19:00.

Friday 28 August — the heavy one.
At 16:00, three things land together: Chair Warsh speaks at Jackson Hole, the final August consumer sentiment reading confirms or revises the 51 flash, and the Bureau of Labor Statistics publishes its preliminary annual revision to payrolls. Last year’s revision removed 911,000 jobs from the record after the fact. Euro area inflation flashes for France and Spain arrive during the morning.

Through the week.
Canada’s retaliation. Carney has promised to match the tariffs dollar for dollar but has not said on what. The direct sums are modest — about $20 billion of goods against nearly $900 billion of annual trade between the two countries, and the list runs to plywood, liquor and hockey equipment. What matters is whether it stops there.

☕ Last Sip

The lesson of the week is that a government can influence the price of its own debt for about a day. After that, the buyers decide.

Nobody failed to sell a bond this week, and nobody failed to pay one back. It is something slower and more ordinary: the group of people willing to lend to rich countries for thirty years has stopped growing as fast as the amount those countries want to borrow. Prices adjust until the two meet. What we watched this week was this adjusting.

The reason to pay attention, even if you own no bonds at all, is that this number sits underneath almost everything else.

Mortgages are priced off it. Company borrowing is priced off it. And every euro a government spends on interest is a euro it does not spend on anything else.

Next week gives us three auctions and a speech from Jackson Hole. That should tell us whether last week was a wobble or the start of something that keeps going.

Enjoy the last stretch of August. See you next Sunday for the next episode of “The Weekly Steep”.

Sources

Market levels: Trading Economics for indices, yields, currencies and commodities; CNBC and Yahoo Finance for the US closes; Il Sole 24 Ore and SoldiOnline for the Italian–German spread; Zai Diamond and Zaikei Shimbun for the New York close in dollar-yen and euro-dollar. Economic releases: S&P Global, US Department of Labor, Eurostat, Trading Economics calendar. Credit and euro curve: ICE BofA indices via FRED and the ECB Data Portal, from my own pipeline. Onchain: Cointelegraph and Yahoo Finance. Weekly changes are measured against the levels published in edition #4. Analysis and personal opinion — never investment advice.

Wednesday’s inflation report landed perfectly where economists expected it to. Consumer prices went up 3.4% over the year, down from 3.5%, with the core measure that strips out food and energy at 2.5%. Traders read it as the end of the rate-rise argument that has run all year. The probability of a US rate hike in September fell from about 57% a week ago to about roughly 30% by Friday. The equity market responded fast and share prices climbed to records.

The same Wednesday, the Treasury reported a $432 billion deficit for July. This marks the largest July on record. On Thursday it sold thirty-year debt at 5.216%, which is the highest rate since 2001.

But that is the odd thing that wonders investors. If the central bank is less likely to raise rates, borrowing should be getting cheaper. It got more expensive instead, because the government now has more bonds to sell than it can find buyers for at the old price.

Then Friday brought something nobody expected: American shoppers stopped spending.

The Scoreboard

Lets have a look at our weekly scoreboard to start our session. There are two major things that spark interest. The first is how little moved from one Friday to the next. Most of these numbers ended the week close to where they started, and a quick glance would tell you nothing much happened during this week.

But this is wrong. A lot has happened. American shares hit a record on Thursday but gave it back on Friday. The ten-year government borrowing rate touched a nineteen-month high on Tuesday before settling almost exactly where it began. Weekly numbers are a snapshot of two moments, seven days apart, but sometimes they hide everything in between. This was such a week where the space in between was the story.

There was not much happening in the US. The markets barely moved. The S&P 500 up 0.4%, the Nasdaq 0.1%, but both touched records on Thursday before Friday’s retail sales report knocked them back down.

Government borrowing costs look almost unchanged too: the two-year flat, the ten-year up 2 basis points. The two-year and the ten-year answer different questions, which is why they can move apart.

The two-year is basically a bet on the central bank. It barely moved because traders spent the week deciding a September rate rise was most likely off the table. This is because the week's data made a September rate rise look unlikely. Wednesday's inflation report came in on forecast rather than hot, and Friday's collapse in retail sales and consumer confidence made a central bank raising rates into a slowing economy hard to imagine. By Friday the market put the odds of an increase at roughly 30%, down from about 57% a week earlier.

The ten-year is a different kind of story. That is where the government does much of its real borrowing, and its price depends on something simpler: how many bonds are being sold, and how many people want them. On Tuesday it touched 4.75%, the highest in nineteen months, before easing to 4.68% by Friday.

This is the number to care about, because it sets prices well outside the bond market. It is what American mortgage rates are built on. Q thirty-year home loan costs about 6.7% today for this reason. Large companies borrow off it too.

So in essence the week’s message was this: Traders stopped expecting the central bank to raise rates in September. You can read that expectation off the futures market, where people bet on the decision months ahead. The prices of those bets translate into odds. A week ago they pointed to roughly a 57% chance of an increase. By Friday it was near 30%. Normally that alone would make borrowing cheaper across the board. It did not. The government raises money by selling bonds at auction. Each bond pays a fixed amount every year. So if buyers are scarce and the government has to accept a lower price, that fixed payment now represents a bigger return for whoever buys and a bigger return for the lender is a higher cost for the borrower. Price down, yield up. They are two ways of saying the same thing.

This week the government had a great deal to sell, and buyers were not enthusiastic. Thursday’s thirty-year auction went through at 5.216%, the highest since 2001. So the ten-year rate climbed even while the central bank looked less threatening. When those two move apart like that, the cause is usually not interest-rate policy. It is the sheer volume of debt coming to market, which is the subject of the first chapter below.

Europe was quiet, though Germany’s ten-year yield reached a fifteen-year high mid-week before settling at 3.20%.

Asia carried the real movement. Korea’s KOSPI rose 11.5%, the same index that fell 5.1% the week before. The yen weakened 1.6% to around 159 per dollar, its worst week in three months.

1. The government is borrowing faster than the market wants to lend

A record July deficit, and a thirty-year debt sale at the highest rate in a quarter century.

Governments spend more than they collect and borrow the difference by selling bonds. Buyers set the price. When there are more bonds than eager buyers, the government pays more.

July’s gap was $432 billion, a record for the month. For the fiscal year so far the total is $1.8 trillion which is already more than the government borrowed in all of last year, with two months still to run.

One honest caveat. August began on a weekend, so roughly $99 billion of August benefit payments went out in July. Adjusted, the shortfall was $333 billion, still up 18% on a year earlier, but not the big headline number.

The market answered on Thursday. The Treasury sold $25 billion of thirty-year bonds at 5.216%, the highest since 2001. Buyers put in $2.39 of bids for every dollar on offer, down from $2.44 at the previous sale. The government got its money, but it had to reach a little further to find it. Wednesday's ten-year sale drew the highest financing cost at that maturity since 2007.

Interest on money already borrowed has cost $1.17 trillion in the first ten months of this budget year which is 15% more than the same stretch a year ago. That interest is not something separate from the deficit. It is one of the largest items inside of it.

So the chain runs exactly like this: more borrowing pushes the rate up, a higher rate makes next year’s interest bill bigger, a bigger bill widens the deficit, and a wider deficit means still more borrowing.

Anyone who has carried a balance on a credit card will recognise this loop.

2. Prices cooled. Shoppers stopped anyway.

July inflation came in exactly on forecast, and Friday’s data showed the American consumer is pulling back hard.

Inflation at 3.4% is the second consecutive monthly slowdown, and core at 2.5% is back where it sat before the Iran conflict began. Petrol prices fell 2.9% during the month. This is what a fading energy shock looks like.

Then Friday came. Retail sales fell 0.6% in July against expectations of a small rise. This is the sharpest drop in over a year. Consumer sentiment came in at 51 against forecasts of 55. Weaker consumer sentiment, weaker economy.

But what caused this? One number explains the gap between the cheerful inflation report and the gloomy shopper. Prices rose 3.4% over the year. Average hourly pay only rose about 3.2%. Inflation slowing is not the same as prices falling, and the average American’s pay is losing the power.

Energy still does the damage even while it fades: petrol costs 24.6% more than a year ago, heating oil 39.1% more. The annual comparison improves every month. The bill each month does not.

3. Korea again — and I got half of it wrong

The index that fell 5.1% last week rose 11.5% this week, and the money came from abroad.

Last week I wrote that Korea’s market was falling because two memory-chip makers dominate its index, and that Korean savers were moving money into American shares. The first half held. The second needs correcting from my end.

The KOSPI closed Friday at 6,977.94, up 11.5% on the week and briefly back above 7,000, with five consecutive winning sessions. Memory chips led it up exactly as they led it down. Concentration cuts both ways, and one week is not enough to tell a structural problem from a bad fortnight.

What I did not anticipate: foreign investors bought a net 3.03 trillion won of Korean shares in a single session. So while Korean retail savers were buying American stocks in July, foreign institutions were buying the market those savers were leaving.

That is worth remembering about a whole class of story. “Locals are fleeing their own market” is compelling, and it was true. But it was also one side of the ledger.

4. The yen gave back half the rescue

Three weeks after the first joint US–Japan intervention since 1998, the currency is most of the way back down.

Last week’s letter said intervention buys time rather than fixing anything, because the interest-rate gap that pushed the yen down was still there. That resolved faster than expected.

The yen fell about 1.6% to roughly 159 per dollar, its largest weekly loss in three months, surrendering close to half the ground gained when Tokyo and Washington bought it together in late July. The 160 level that triggered the first intervention is back in no time.

There is a second-order effect that connects to the first chapter. Japan is among the largest foreign owners of American government debt. If Tokyo intervenes again it may need to sell US Treasuries to fund the purchase and traders priced that possibility into US long-term yields this week. A currency operation in Tokyo is now part of why Washington pays more to borrow and why the US has a specific interest in the yen.

The Bank of Japan meets on 17 and 18 September, and analysts expect it to raise its rate to 1.25% from 1%. To see why that matters more than another round of buying, look at the gap. Japan pays about 1% on its money. America pays between 3.50% and 3.75%. Money moves toward the better return, so an investor sitting on yen has a standing reason to sell them and buy dollars instead. That selling is the thing pushing the yen down, and it happens every day, in small amounts, from thousands of separate decisions.

Intervention does not touch any of that. It is one very large purchase on one afternoon. It moves the price for a few days but it changes nobody’s reason behind the moves.

Data Check

Two kinds of number came out this week, and they split cleanly.

One kind measures prices. Consumer inflation on Wednesday, producer prices on Thursday. Both landed almost exactly where economists had predicted.

The other kind measures what households actually do. How many filed for unemployment support, how much they spent in shops, how they feel about the months ahead. All of those came in worse than expected, and retail sales missed by a wide margin.

So the week was not really an argument about inflation. That question got answered on Wednesday and largely settled. What it opened instead is a different one: whether American households are still able to carry the economy the way they have for the past two years.

The Credit Grind

European junk-rated companies are paying the smallest premium over governments in almost the entire past year. In the same week those governments paid the most in decades.

A borrowing company pays two prices added together: the base cost of money, set by what governments pay, and a surcharge for the chance it fails to repay. This week those two moved in opposite directions, which is the point.

Euro high yield fell another 6 basis points this week, to 257. The direction is the confusing part: a smaller number means investors are asking for less extra payment to lend to these companies, not more.

How little? 257 sits in the 6th percentile of the past year. On 94% of the trading days behind us, lenders were paid more than they are being paid today. The American equivalent sits at the 9th percentile, almost as low.

Now hold that next to what governments pay. The two-year euro government yield sits in the 93rd percentile of its own year, the ten-year in the 92nd.

Normally these two move together. When borrowing gets more expensive in general, it gets more expensive for risky companies too. This week that broke apart. The same investors, in the same few days, wanted close to the most they have wanted all year to lend to governments and close to the least to lend to Europe’s weakest companies.

If you hold a corporate bond fund, that is worth knowing. Almost all of the yield you collect right now comes from the safe half. Very little of it is payment for the risk you are carrying.

The Onchain Pour

Bitcoin fell while the stock market held their records, and Washington moved closer to crypto than it has ever been.

Bitcoin ended Friday around $63,050, down 2.9% on the week. Ether finished near $1,870, down 2.1%. CryptoQuant reported that the share of bitcoin held at a profit has dropped into territory it has historically reached only near market bottoms. This is either a floor forming or a floor about to break. We will keep monitoring the current situation.

But the policy news was larger than the price news. A US regulator conditionally approved a bank charter for World Liberty Financial, the Trump family’s crypto venture: a sitting president’s family business receiving permission to operate as a bank in an industry his administration regulates.

The White House hosts executives from Coinbase, a16z, Ripple, Chainlink and the prediction market Kalshi on Wednesday.

The Week Ahead

Monday 17 August.
Japan’s second-quarter growth at 01:50 (consensus 0.5%). China’s July industrial production and retail sales at 04:00 — retail sales are forecast to pick up to 1.5% from 1%. Then the one worth marking: US foreign investment flows for June at 22:00, which shows how much American government debt overseas buyers actually bought. Given this week, that number matters more than usual.

Tuesday 18 August.
UK unemployment and wages at 08:00 (pay growth forecast to slow to 4% from 4.3%). German investor sentiment at 11:00, expected to improve to 30 from 26.3. US housing starts and building permits at 14:30, then pending home sales at 16:00 after last month’s 5.4% collapse.

Wednesday 19 August — the busy one.
UK inflation at 08:00 is forecast to jump to 2.9% from 2.6%. Lagarde speaks at 09:10. Euro area final July inflation at 11:00 (consensus 2.9%). A US twenty-year bond auction at 19:00 is the next test of the demand that wobbled on Thursday. Then the record of the July Fed meeting at 20:00 — the one where three officials voted to raise rates. What they said to each other is the clearest guide available to September.

Thursday 20 August.
The ECB publishes the account of its own July meeting at 13:30. US jobless claims at 14:30 (consensus 210,000) and the Philadelphia Fed factory survey, forecast to fall hard to 25.3 from 41.4.

Friday 21 August.
Japanese inflation at 01:30, with the core measure expected to rise to 1.8% from 1.6% — relevant to whether the Bank of Japan moves in September. Flash business surveys for France, Germany, the euro area and the UK through the morning, and the US equivalent at 15:45.

Through the week.
Home Depot, Lowe’s, Target and Walmart report. This the clearest read on whether Friday’s retail sales drop was a one month event or a turn.

Unscheduled and larger than all of it: the US–Iran ceasefire was reported at a standstill on Friday evening, and President Trump said he would declare the Strait of Hormuz United States territory.

That happened after markets closed and is in none of the numbers above.

☕ Last Sip

Two true things pointed in the exact opposite directions this week.

Inflation is fading and the central bank is standing down. The government’s borrowing costs hit a twenty-five-year high anyway. Both will still be true on Monday.

Enjoy what is left of August. See you next week for the next episode of The Weekly Steep.

Sources

Market levels: Trading Economics for indices, yields, currencies and commodities; Kiplinger and Yahoo Finance for the US closes; Il Sole 24 Ore for the Italian–German spread. Economic releases: US Bureau of Labor Statistics, US Treasury, Eurostat, Trading Economics calendar. Credit: ICE BofA indices via FRED, from my own pipeline. Onchain: Cointelegraph and CryptoQuant. Weekly changes are measured against the levels published in edition #3. Analysis and personal opinion — never investment advice.

The Week in One Sip

At 14:30 on Friday the US Bureau of Labor Statistics reported that the American economy shed 23,000 jobs in July. Economists had expected a gain of about 80,000. Within minutes the S&P 500, the index of America’s 500 largest listed companies, was heading for a record close.

That looks strange until you know what the argument in markets has been about all year. The US central bank has spent 2026 debating whether to raise borrowing costs, not lower them, and three of its officials voted for a rise in July. A labour market shedding jobs makes that case much harder, and cheaper borrowing lifts share prices. A bad number for workers arrived as a good number for shareholders.

Underneath ran a second story that never resolved. Oil fell hard early in the week as Iran and Oman negotiated a corridor through the Strait of Hormuz, the waterway that carried roughly a fifth of the world’s oil before the war.

Then Iran published its terms, they were stricter than anyone expected, and oil turned back up on Friday.

The Scoreboard

Look for a clean split. Everything that benefits from cheaper money rose, everything priced off the dollar or off oil went the other way, and one market broke ranks entirely.

Reading the board.
America ran the week: the S&P 500 up 3.6% to a record 7,757.64, the Nasdaq 5.2% as chip stocks bounced, and the Russell 2000 which consists of smaller, domestic companies rather than global technology, keeping pace at 3.5%. Government borrowing costs fell across every maturity.

Europe followed rather than led. The gap between what Italy and Germany pay to borrow for ten years narrowed to 77 basis points, which is the market saying it sees no particular stress in the eurozone.

Asia is where the calm number hides everything. Japan’s Nikkei rose 1.9% while the yen strengthened 2.1%. Korea’s KOSPI fell 5.1% in the same week the S&P hit a record. Gold did the most work of anything on the board, up 7.8% to around $4,343 an ounce.

1. The revisions mattered more than the headline

July’s jobs report was bad. What it revealed about May and June was worse.

A monthly jobs report carries two things: what happened last month, and a correction to what you were told about the months before. Almost all the coverage goes to the first. The money is usually in the second.

July came in at minus 23,000. But May and June were marked down too, heavily enough that average monthly job creation over the past year has collapsed to roughly 34,000. That reframes everything. This wasn’t one weak month interrupting a strong run — the run was already far weaker than the official numbers showed.

One detail explains a headline that otherwise makes no sense. The unemployment rate fell, to 4.1%, in a month when the economy lost jobs. The rate only counts people actively looking for work. The share of adults working or looking dropped to 61.4%, down 0.7 points this year, as close to 1.4 million people stopped looking. Leaving the labour force takes you out of the count.

So the headline flattered a genuinely poor month. Wages agreed: average hourly pay rose 3.2% over the year against the 3.5% expected. If you’ve been reading that the American labour market is fine, that claim got harder to make this week.

2. Record profits, and where they came from

Company earnings are running at levels not seen since the pandemic rebound. A meaningful slice is companies revaluing what they own.

With 88% of the S&P 500 reported, earnings are up more than 50% on the same quarter last year against an expectation of 23%. Margins are at a record. On the face of it, an extraordinary boom.

Two numbers deserve a second look. Alphabet booked roughly $98 billion in gains on investments it holds but hasn’t sold, and Amazon around $53 billion from its stake in Anthropic. Both are legitimate under accounting rules. Neither is money from selling anything to a customer. Strip them out and growth is still above 30%, so the boom survives — but a profit from marking up a shareholding behaves very differently in a downturn than a profit from sales.

Look one layer below the index heavyweights and it gets uneven. United Wholesale Mortgage, the largest mortgage lender in the United States, fell 34.8% in a single session on 6 August to an all-time low. Lumber has traded down eleven days running, its longest losing streak since April 2024. Housing and construction are not having the year the index is having.

Both things are true at once, and the distance between them is what to watch.

3. Korea fell five percent while everyone else rose

A memory-chip selloff took Korea’s market down hard in a global rally — and Korean savers are voting with their money.

Korea’s KOSPI closed Friday at 6,258.77, down 5.1% on the week. The S&P 500 rose 3.6% over the same days. That’s an unusually wide gap between two markets that normally move together.

The cause is concentration. Korea’s index leans heavily on two memory-chip makers, and memory had a bad week — SK Hynix dropped 4.8% on Friday alone, extending a selloff that has run for weeks. When two companies carry that much of an index, their bad month is the country’s bad month.

What makes it more than a market story is where Korean money went instead. Retail investors in Korea bought $4.6 billion of US shares in July, roughly seven times June’s figure. Ordinary savers are moving out of their home market into someone else’s. That’s slow-moving rather than a crisis, but a country whose savers prefer foreign assets has a harder time funding its own companies.

4. The currency two governments moved

The US and Japan bought yen together for the first time since 1998, and the American reason is about American borrowing costs.

When a currency falls too far, a government can step into the market and buy it with its own reserves. It’s a blunt tool and usually a national one. In late July it stopped being national.

Japan intervened alone on 30 July, with the yen close to 164 per dollar — roughly a forty-year low. The next day the US Treasury joined, selling euros and buying yen through the New York Fed. Coordinated intervention between the two hadn’t happened since 2011, and coordination in the direction of buying yen not since June 1998. It has held so far: the yen closed Friday at 156.68, a little over two percent stronger on the week.

The interesting part is the American motive. A weak yen and rising Japanese interest rates travel together, and Japanese investors are among the largest owners of US government debt. If yields at home rise far enough, that money goes back to Tokyo and America pays more to borrow. Washington had a domestic reason to defend the yen.

Intervention buys time. The gap in interest rates that pushed the yen down is still there.

Data Check

Everything pointed the same way before Friday. Job openings, private payrolls and the services survey all came in below expectations, and then the employment report confirmed the direction.

Consensus is the median forecast from Reuters and Bloomberg surveys of economists, published free by Trading Economics. Markets move on the gap between forecast and outcome.

The payrolls miss was the largest surprise of the week by a distance — 103,000 jobs between forecast and outcome. Worth remembering when the unemployment rate gets quoted on its own over the coming months.

The Credit Grind

One number: what investors charge America’s weakest-rated companies for the risk that they fail to repay.

It fell to 271 basis points this week — 2.71 percentage points of extra yield above what the US government pays to borrow. That is the 8th percentile of the past year, meaning investors were paid more than this in 92% of the last twelve months. They are accepting close to the least compensation for corporate default risk on offer in a year, and they accepted less of it this week than last.

Extra yield over government bonds for high yield borrowers, past twelve months. Pipeline run of 6 August 2026, ICE BofA indices via FRED.

One corner went the other way. Euro-denominated debt from emerging-market companies widened 9 basis points — the only part of this market where investors demanded more. Those borrowers sit closest to the energy story, and Hormuz moved against them.

High yield covers companies rated below investment grade. Levels are 6 August, one day behind everything else on the Scoreboard.

If you hold a corporate bond fund, this is what you are being paid to take risk — and right now it is thin.

The Onchain Pour

Bitcoin went nowhere in particular, which given everything else happening is itself informative.

Bitcoin ended Friday around $64,940, up 1.6% on the week after touching $65,300 on the jobs report. Ether finished near $1,910. Funds that hold bitcoin directly took money in every single day of the week.

The structural news mattered more. The US Treasury sanctioned two Iranian exchanges for allegedly moving money for the Revolutionary Guard — the same conflict driving oil, arriving in crypto. Legislation to give the industry clear US rules hit another setback, with Senator Lummis promising to keep pushing. And developers are arguing over a proposed change to Bitcoin’s rules, BIP-110, openly enough that a network split is being discussed.

One thing for anyone holding coins themselves: losses from the Coldcard hardware wallet exploit have passed $111 million.

The Week Ahead

Monday 10 August.
Quiet. Japan’s central bank publishes the summary of opinions from its July meeting at 01:50, worth a glance for how close another rate rise is.

Tuesday 11 August.
Australia’s central bank decides at 06:30, expected to hold at 4.35%. US small-business optimism at 12:00 (consensus 97.1). Existing home sales at 16:00 (consensus 4.07 million) matter more than usual after this week’s mortgage-sector trouble. A three-year debt auction at 19:00 gives the first read on demand since the jobs shock.

Wednesday 12 August — the one that counts. US consumer price inflation for July, 14:30.
Consensus is 3.4% for the year, down from 3.5%, with the core measure that strips out food and energy at 2.5%. This decides whether Friday’s jobs report ended the rate-rise debate or merely paused it; a hot print puts September back in play immediately. Germany’s final July inflation at 08:00. The IEA oil market report at 11:00 and OPEC’s monthly report later both carry extra weight with Hormuz unresolved.

Thursday 13 August.
UK second-quarter growth at 08:00 (consensus 0.4%, down from 0.6%). US producer prices at 14:30 — the inflation companies face before it reaches shop shelves, consensus 0.2% on the core measure — alongside weekly jobless claims at 198,000. Then Hammack speaks at 14:15. She voted to raise rates in July, so how she sounds after a negative jobs print tells you how firm that camp still is. Barkin follows at 14:40.

Friday 14 August.
Euro area second-quarter growth, second estimate, at 11:00, consensus confirming the 0.4% flash. US retail sales at 14:30, consensus 0.2%, the cleanest read on whether households keep spending while the job market softens. Michigan consumer sentiment at 16:00 (consensus 54, down from 55.2).

Unscheduled and larger than anything on that list: Iran’s parliament is reviewing the draft terms for reopening the Strait of Hormuz.

☕ Last Sip

Two weeks ago the question was whether the US central bank would raise rates in September. After Friday that mostly answers itself, which loads the whole week onto Wednesday’s inflation number.

Enjoy the quiet part of August while it lasts. Bis nächsten Sonntag zur nächsten Folge von The Weekly Steep.

Sources

Market levels: Trading Economics for indices, yields, currencies and commodities; Reuters and CNBC for the US closes; Il Sole 24 Ore for the Italian–German spread. Economic releases: US Bureau of Labor Statistics, Eurostat, Trading Economics calendar. Credit: ICE BofA indices via FRED, from my own pipeline. Onchain: Cointelegraph and Coinbase.

Weekly changes are measured against the levels published in edition #2. Analysis and personal opinion — never investment advice.

One piece of groundwork first, because everything below runs on it.

When a government spends more than it collects in taxes, it borrows the difference by selling bonds. A bond is an IOU: give me €100 today, I’ll pay you interest every year and return the €100 on a fixed date. The interest rate on that IOU is the yield, and a US government bond is a Treasury. The two-year is repaid in two years and mostly reflects what traders expect the central bank to do soon. The thirty-year is repaid in thirty and reflects something slower.

Now the week. Oil dropped about 8% on hopes of peace with Iran. The US economy grew 1.5% last quarter against a 2.1% forecast, and slower growth normally means less pressure on prices. Sure enough, the inflation measure America’s central bank watches most closely cooled as well. That combination almost always makes long-term borrowing cheaper.

On Wednesday the Federal Reserve held its main interest rate steady for the fifth meeting running, with three of its own officials voting for a rise. Chair Kevin Warsh declined to say what happens next.

The two-year yield fell, exactly as you’d expect. The thirty-year rose to 5.25%, the highest since 2007.

Short-term borrowing got cheaper. Thirty-year borrowing got dearer. In a week when every piece of news pointed the other way. The chapters below explain why, and what it costs you.

One more thing happened while I was writing: late Saturday night, President Trump called off a planned strike on Iran and said a framework for a deal had been agreed. That will move Monday’s open, and I come back to it at the end.

This week’s chapters

1. The bond market stopped listening — why long-term rates rose against the news
2. Japan came home — America’s most reliable lender is buying at home instead
3. What goes up too fast comes down — Korea, chips, and borrowed money
4. Dancing until the music stops — AI spending meets the cash register
5. Something is working — jobs held up, and a record in new businesses

The Scoreboard

Here is where markets closed on Friday, and how each moved over the week.

Before your eye lands on the numbers, look at the three American bond lines. The two-year fell while the ten-year and thirty-year rose. Everything else on the board moved more or less together. That disagreement is the story.

A basis point (bp) is one hundredth of a percentage point. A rising yield means dearer new borrowing and a lower price for bonds already issued. Closes: Asia 08:00, Europe 17:30, US 22:00 CET. Levels marked ~ are approximate.

Reading the board. America looks calm and wasn’t: the Nasdaq’s 1.6% covers six falling days into Wednesday, then a surge when Microsoft alone gained $450 billion in a single session.

Europe quietly had the best of it, with German shares leading and the Euro Stoxx 50 up 1.9%. The line marked BTP–Bund is the extra interest Italy pays over Germany to borrow for ten years, the standard gauge of nerves about the eurozone. At 81 basis points it sits near a fifteen-year low.

Asia was where the violence was. Korea’s KOSPI rose 17.9% on Friday, its biggest single day ever, and still ended the week down 1.4%. That tells you how bad Tuesday and Wednesday were.

1. The bond market stopped listening

Long-term borrowing costs rose against every piece of news this week, because lenders are charging extra for uncertainty rather than for inflation.

The Federal Reserve does not set long-term interest rates. It sets one rate — what banks charge each other overnight. Everything longer is set by investors buying and selling bonds, and when an investor decides what to demand for a thirty-year loan, they are pricing two separate things. Where will that overnight rate average over thirty years? And how much extra do I want for the risk that I’m wrong? Call the second part a nervousness fee.

Usually both move together. This week you could watch them come apart. Traders cut the odds of a September rate rise from around 80% to 63%, and the two-year yield fell seven basis points with them. The thirty-year went the other way, and after stripping out expected inflation, what lenders actually earn on it is now the highest since 2008.

Two things are feeding that fee. Warsh spoke forcefully about inflation on Wednesday and then held rates for the fifth meeting running, with no guidance on what comes next, and reports on Friday suggest he wants fewer scheduled meetings each year. Meanwhile American government debt passed $40 trillion — debt the government refinances roughly a third of every year, which makes it closer to a mortgage that resets than one fixed for the term. Market rates reach its interest bill in months.

The Fed has been cutting since 2024. The thirty-year yield is clearly higher than when those cuts began. A central bank lowering short rates while long rates climb is being overruled by the people who actually hand over the money.

What this means for you: the rate on a thirty-year mortgage is set by investors’ willingness to lend for thirty years, not by Fed announcements. That’s why headlines can say “the Fed cut rates” while your quote doesn’t move.

2. Japan came home

Japanese institutions have bought American government debt for decades. Rising interest rates at home are ending that.

For thirty years Japanese government bonds paid almost nothing, because the Bank of Japan held the ten-year rate near zero on purpose. So Japanese pension funds and insurers, sitting on vast savings and earning nothing at home, bought foreign bonds instead — above all American ones. Japan became the largest foreign owner of US government debt.

That is unwinding, faster than most people notice. The Bank of Japan dropped its cap in 2024 and raised its main rate to 1% in June. Japan’s ten-year yield now sits at 2.80%, a level last seen in the 1990s. This week the central bank held while warning that inflation might overshoot.

The consequence is simple. A Japanese insurer can finally earn a decent return at home, in its own currency, with no exchange-rate risk. So it stops buying American paper and brings the money back — $96 billion of it over the past three months, leaving holdings at $1.14 trillion, the lowest since April 2025.

Then there’s the yen, which had fallen to a forty-year low against the dollar. That makes everything Japan imports dearer, and on Thursday Tokyo stepped in, selling dollar reserves to buy back its own currency. It spent an estimated $53 billion in one day, the largest such intervention on record anywhere. Reuters reports America is now helping.

For all that, Japan’s ten-year yield finished the week one basis point lower than it started. A central bank meeting, a record intervention and a currency at forty-year lows moved it almost not at all, which tells you the pressure is structural rather than event-driven.

What this means for you: one of the most dependable buyers of American government debt is stepping back, exactly when America needs to sell more of it.

3. What goes up too fast comes down

Chip shares ran too far, Korean small investors bought the top with borrowed money, and the unwind ran automatically.

You’ve seen this shape before, even if you’ve never owned a chip stock. A sector runs hard for over a year. Money floods in near the peak, much of it through leveraged products, which multiply gains and losses two or three times over. Small investors, who by definition arrive after the good performance has already happened, take their biggest position at the highest price.

Then it turns. Chip shares worldwide fell 23.2% in July, their worst month since 2002. In Korea, where appetite for these products ran especially hot, ordinary investors had bought more than $60 billion of shares since April while foreign institutions sold over $70 billion, and borrowed money hit a record 38.6 trillion won in late June.

Then the market fell so fast on three consecutive days that trading halted automatically each time, something never seen before. Citi estimates more than 360,000 people were forced to sell everything, 62% of them under 35, with losses near $38.7 billion. Almost all of it happened without anyone deciding anything, triggered by clauses written into the contracts.

On Friday the finance ministry capped these products, foreign institutions bought a record amount as small investors sold a record amount, and the index jumped 17.9%. It still ended the week down.

The mechanism is worth understanding. Borrow to buy, and a falling price forces you to sell. Your selling pushes the price lower, which forces the next person out. Being right about the long term doesn’t help.

What this means for you: you can’t control whether a hot sector cools. You can control whether you’re forced to sell when it does. Borrowed money takes that choice away at the worst possible moment.

4. Dancing until the music stops

The four biggest cloud companies will spend three quarters of a trillion dollars on AI this year, and investors have started asking whether it comes back as cash.

In 2007 the boss of Citigroup said that as long as the music was playing, you had to get up and dance. The music stopped shortly after, and it did not end well.

The four largest cloud companies are still dancing. All four spent enormously this quarter, all four guided higher, and together they plan $720–745 billion of building this year. Nobody wants to be the one who sat down early and looked foolish.

What changed this week is where the strain showed. Microsoft justified it: its cloud business grew 43% and passed $100 billion in annual revenue for the first time, and the shares had the best single day any company has ever recorded. Amazon rose 15.6% on Friday telling the same story.

Meta went the other way. Record revenue of $60.8 billion, costs up 55%, and free cash flow down 91% from a year ago to $784 million. Free cash flow is what’s left after a company both runs itself and pays for what it’s building — the money that funds dividends and services debt. Two years ago these were the most cash-generating businesses on earth, and several are now scraping zero.

Companies that can’t fund a building spree from cash borrow instead. Around $400 billion of new borrowing is expected, and the cost of insuring these companies against default hit a record this week.

What this means for you: the AI trade started as a story about asset-light companies with enormous margins. It’s becoming a story about heavily indebted ones building warehouses of chips, and those two behave very differently when money gets tight.

5. Something is working

Jobs held up and record numbers of people are starting companies, and both deserve as much attention as the problems above.

I don’t want to leave you on the debt, and I don’t need to.

Each week America counts first-time filings for unemployment benefits, the fastest signal available on whether people are losing work. That number came in at 197,000, near its lowest in decades. The fear at the end of last year was an unemployment rate rolling over, and it hasn’t happened.

The second one I like more. Record numbers of new technology companies are being registered, and many are one or two people working from a bedroom. The barrier to starting a software business keeps collapsing.

There’s a link back to chapter four worth spotting. Enormously expensive building booms have rarely been solved by building more, but by someone finding a way to need less — cheaper extraction after an oil shock, better chips after a shortage. That someone usually works at a tiny new company, and there have never been more of those.

What this means for you: the boom and the correction can both be true. Expensive capital cycles routinely disappoint the investors funding them and benefit everyone who ends up using what got built.

The Week Ahead

Two events dominate, and around them sits the busiest week for government bond sales in months.

Monday 3 August

• US factory survey, 16:00 CET. A poll of purchasing managers where anything above 50 means expansion. Expected at 54.0 after 53.3. Read the prices component first: it was 73.0 last month and is forecast at 70, and that’s where July’s oil move shows up before it reaches anything else.

• US bank lending survey, 20:00 CET. Banks report whether they’re tightening standards on loans. It rarely makes headlines, but it’s the earliest warning that credit is getting harder to obtain, which matters a great deal given chapter four.

• US Treasury borrowing estimates, 21:00 CET. The warm-up for Wednesday.

• Eurozone factory survey, 10:00 CET, expected at 52.0.

  ---

Tuesday 4 August

• SpaceX first earnings as a public company, after the US close. It listed in June in the largest IPO in history and has since fallen roughly 45% from its peak. Two trading days later, up to 911.5 million shares held by employees and early investors become sellable — about $110 billion of stock arriving in a market where only 5% of the company trades freely. Musk’s own stake stays locked until June 2027.

• AMD earnings, after the close. The first large chip company to report since the sector fell 23% in July. Analysts expect revenue up 47% from a year ago, so the question is whether the business is still growing while the shares fall.

• Japan sells ten-year debt, 05:35 CET. The last auction cleared at 2.729%. Weak demand pushes Japanese yields higher and feeds straight back into chapter two.

• US job openings, 16:00 CET, expected at 7.25 million after 7.59 million.

  ---

Wednesday 5 August

• US Treasury refunding announcement, 14:30 CET. The most important scheduled event of the week, and almost nobody covers it. Treasury says how much it plans to borrow and at which maturities. With thirty-year borrowing at 5.25%, the split between cheap short-term bills and expensive long bonds tells you what the government itself expects to happen.

• US private payrolls estimate, 14:15 CET, expected at 75,000 after 98,000. A rough steer into Friday.

• US services survey, 16:00 CET, expected at 54.2. Services are roughly four fifths of the American economy, so this outranks Monday’s factory number by some distance.

• Germany sells ten-year and thirty-year debt, 11:30 CET.

• Japanese wage data, 01:30 CET, expected at 3.4% growth. Wages are what the Bank of Japan watches before raising rates again.

  ---

Thursday 6 August

• SpaceX share unlock. The shares described above become sellable.

• Japan sells thirty-year debt, 05:35 CET. The last one cleared at 3.993%, and Japan’s longest-dated bonds have been the most fragile part of that market all year.

• US weekly jobless claims, 14:30 CET, expected at 200,000.

• US unit labour costs, 14:30 CET, expected to jump to 2.7% from 1.8%. This measures what it costs an employer to produce a unit of output, and rising costs eventually reach prices.

• Eurozone retail sales, 11:00 CET.

  ---

Friday 7 August

• US jobs report, 14:30 CET. How many jobs America added in July, and the last major labour reading before the Fed meets on 16 September. Expected at 91,000 after a weak 57,000, with unemployment forecast to tick up to 4.3% from 4.2% and wages up 3.5% from a year ago.

• China trade data, 05:00 CET. Exports grew 27% last month, a figure few economists take at face value.

• US consumer inflation expectations, 17:00 CET, last at 3.7%. What ordinary Americans think inflation will be, and the number that decides whether the Fed’s credibility is holding.

Data Check

Every scheduled release comes with a forecast attached before it lands. Read the gaps rather than the levels — a number arriving exactly where everyone predicted moves nothing, because the price already reflects it.

Consensus is the median forecast from surveys of bank and research economists run by Reuters and Bloomberg, published free by Trading Economics.

Two deserve a second look. The eurozone grew faster than expected while inflation rose, which is why investors now fully expect the European Central Bank to raise rates twice more by early 2027.

And look at the oil storage number. American stockpiles fell by 7.17 million barrels when forecasters expected a rise of 1 million, which describes a market genuinely short of oil. The price fell 8% anyway. That tells you how much of July’s price was fear about the war rather than actual shortage.

The Onchain Pour

Bitcoin ignored a strong week everywhere else, which is the most interesting thing it’s done in months.

It held near $64,300 through a session in which Korea surged 17.9%, chip shares soared and the Nasdaq rallied, finishing the week down about 2% at $63,870, with ether at $1,890.

All year Bitcoin has moved like an excitable technology share, so sitting out a rally like that is a real change. It did move on the Fed decision, which suggests it’s tracking interest rates now rather than the stock market. Whether that holds is what I’ll watch in August.

Two developments matter more than the price. Circle won a limited banking licence from New York’s financial regulator. And Open USD launched — a stablecoin, meaning a token designed to always be worth one dollar, this one backed by more than 140 businesses including Visa, Mastercard and BlackRock. It launched on Ethereum, a public network anyone can build on, rather than a private chain of its own. The plumbing of the financial system is being rebuilt by the incumbents, in public, on open networks.

Last Drop

Something happened while I was writing this. Late Saturday night, Trump announced he had cancelled a planned attack on Iran, saying the outline of a deal had been agreed — one that would fully reopen the Strait of Hormuz and end Iran’s nuclear programme. Israel has joined the commitment, and mediators from Qatar and Oman met Iran’s foreign minister on Saturday. Nothing is signed, and as I write Iran hasn’t confirmed.

About a fifth of the world’s seaborne oil passes through that strait, and part of today’s oil price is simply payment for the risk it stays shut. If it genuinely reopens, that part disappears fast, and oil at $88 has a long way to fall.

Then watch what America pays to borrow for thirty years, because it faces a real test. Cheaper oil means less inflation, which normally means lower rates. This week that logic was ignored completely. If the thirty-year yield holds above 5.20% even as a peace deal arrives, you have your answer about what’s really driving it — and it isn’t the war.

See you next Sunday ladies & gents.

Sources

_{Scoreboard levels come from the Advisor Perspectives Treasury Yields Snapshot and CNBC for US yields, Yahoo Finance and CNBC for equity closes, Il Sole 24 Ore and ANSA for the BTP–Bund spread, Rigzone for Brent, Seoul Economic Daily for the KOSPI, Trading Economics Japan for the JGB, and CoinDesk for crypto. The 10-year Bund level is calculated from the Italian yield minus the published spread rather than reported directly. Market data as of Friday 31 July 2026. News and events to Sunday 2 August 2026, 09:15 CET.}

This paper started with a single sentence. Mark Connors, who has come to serve as a mentor to me in the macro space, pointed out that roughly a third of US Treasuries now sit in short-term paper. I went into the Treasury’s records to verify it, and found something even better: the precise number is more interesting than the round one, and the story behind it is bigger than the statistic. What follows is that story, condensed to one page first.

The situation.
A third of all US marketable debt, 10.1 trillion dollars, matures within the next twelve months. The government repays none of it out of revenue. It sells new paper to redeem the old, week after week, at whatever rate the market demands that day. I computed these figures from the Treasury’s security-level records myself and cross-checked them against the official totals to within four billion dollars.

The choice.
This structure was built, deliberately. Treasury bills, paper shorter than one year, make up 21.5 percent of the stock, above the 15 to 20 percent band the Treasury’s own advisors recommend, and 85 percent of everything currently being issued. The United States has financed itself like a floating-rate borrower: it profits within weeks from every rate cut and pays just as fast for every hike.

The cost.
The average interest rate on the debt doubled in five years, from 1.54 to 3.35 percent, because a third of the debt reprices every year. Net interest now runs 3.5 billion dollars per day, more than the entire defense budget.

The buyers.
Demand remains strong, with nearly three dollars of bids for every dollar of short paper sold. Who provides it has changed: foreign central banks stepped back, money market funds and stablecoin issuers stepped in, the latter required by US law to hold exactly this paper.

The argument.
By financing short, the United States has wired its budget directly to the Fed’s next decision. That wire is a transmission channel rather than a crisis, and it is the most useful lens available for reading every rate meeting, every refunding announcement, and every auction from here on.

Chapter 1: The number worth understanding

The United States owes 39 trillion dollars. That number gets quoted constantly and explains almost nothing. It is too big to feel, it grows every day by design, and on its own it carries no information about risk, cost, or timing. A household owing 500,000 dollars on a thirty-year fixed mortgage and a household owing 500,000 dollars on a credit card are not in the same situation, even though the headline number is the exact same. What really matters and makes the difference is the structure of the debt. When it comes due, what it costs, and how fast that cost can change.

And here comes a number that carries some very spicy information. 10.1 trillion dollars of US government debt comes due within the next twelve months. Due means what it means for any loan. The money has to be repaid on a fixed date. That is 32.6 percent of all the debt the US government has issued to its investors. So essentially a third of the American debt has a clock attached to it with less than a year maturity. I computed this number from the Treasury’s own records, which list every outstanding security with its amount and its due date, and cross-checked my total against the official summary statistics. The two match to within four billion dollars on a base of 31 trillion, which in this world counts as exact. Apollo, one of the largest asset managers, puts the same figure at 33 percent. The government’s own auditors report that 9.1 trillion dollars of maturing debt had to be refinanced in fiscal year 2025 alone. This means that a third of the debt rolls over inside a year.

_{Marketable Treasury debt by remaining maturity, June 2026 · Source: US Treasury MSPD}

The chart shows the full picture, and it is worth a moment. The two left bars carry the most weight. As one can see a third of the debt matures inside one year, another third between one and five years. Together, two thirds of everything the United States owes gets repriced within five years. Most people picture government debt as the opposite: long, patient bonds that were sold decades ago and will not bother anyone for decades more. The chart shows how little of the debt actually works that way. The paper beyond ten years, the image everyone carries in their head, is the smallest of the four bars. The everyday reality of American debt is the tall bar on the left, and it comes due within the year.

The same debt, cut by calendar year instead of bucket, makes the shape even clearer.
The wall stands at the front, and the thin tail running to 2056 is all that carries the famous long average.

_{Marketable Treasury debt by calendar year of maturity, 2026–2056 · Source: US Treasury MSPD}

Why that matters, and why it is neither an accident nor an automatic catastrophe, starts with what this debt actually is.

Chapter 2: What the government actually sells

When the US government spends more than it collects in taxes, it borrows the difference by selling securities. This happens via standardized IOUs. IOU is short for “I owe you” and describes a written note that acknowledges a debt between two parties. You give the government money today, it repays you on a fixed date, and you earn interest for lending out your money and waiting. These IOUs come in three main forms, and the only real difference between them is how long you wait.

Treasury bills are the short ones, running from four weeks to one year. Bills pay no interest along the way. You buy them at a discount, say 98 dollars for a bill that repays 100, and the 2 dollars of difference is your return. Because the wait is short, bills are the closest thing on earth to cash that pays.

Treasury notes run from two to ten years and pay interest every six months. The ten-year note is the one quoted in every news segment, the benchmark for mortgages and corporate borrowing worldwide.

Treasury bonds run twenty to thirty years, on the same schedule, and are bought mostly by institutions with obligations decades out: pension funds, insurers, sovereign reserves.

The government sells this paper at auctions, and an auction means what it always means: investors bid, and the interest rate the government ends up paying is whatever buyers demand that day. Nobody decrees the rate. The market sets it, fresh, at every auction. The Treasury runs these auctions on a published calendar, week in and week out, in sizes that would count as a major corporate bond deal anywhere else and here are simply Tuesday.

One more piece. The Federal Reserve sets a short-term policy rate, and that rate anchors what investors demand at bill auctions, because a bill competes directly with parking money at rates the Fed controls. Raise the policy rate and bill auction rates follow within days. Cut it and they follow just as fast.

Long bonds, by contrast, price on expectations about growth, inflation, and policy over the coming decades, and respond far more loosely to any single Fed decision. This asymmetry between the short and the long end carries the entire story: whoever borrows short has tied their costs to the central bank’s next move, and whoever borrows long has bought independence from it.

_{The three instruments, schematic}

Chapter 3: The rollover: how the machine works

Let’s get to the part that surprises most people. When a Treasury bill comes due, so essentially everything up to one year from issue date, the government does not pay it down out of tax revenue. It sells a new bill and uses the proceeds to redeem the old one. This creates a cycle where the debt does not shrink at any time but rotates. Old paper out, new paper in, week after week, at whatever rate the auction produces that day.

There is nothing scandalous about this. Every government with its own currency operates exactly this same protocol, and large corporations roll their short-term funding on the same principle. The point is not that rolling debt is illegitimate. The point is what happens to a borrower when a very large share of the balance sheet rolls very often? Quite simply put, the borrower’s costs start tracking current interest rates almost in real time.

The scale is easy to underestimate, so here it is in the auditors’ numbers. In fiscal 2025 the Treasury refinanced 9.1 trillion dollars of maturing securities and borrowed 1.9 trillion on top to cover the deficit, a combined 11 trillion dollars of issuance in a single year, equal to about 36 percent of US economic output. A decade of normal operations used to require issuance around 21 percent of output. The machine has not just gotten bigger with the economy. It has also gotten bigger relative to it.

And it has gotten shorter. The four-week bill, the shortest standard instrument the Treasury sells, averaged 47 billion dollars per auction in 2016. Fast forward ten years, in 2026 it averages 101 billion dollars, which makes it the single largest offering in the entire Treasury lineup. Larger than the famous ten-year note. Larger than the thirty-year bond. The most important funding operation of the United States government is one that has to be repeated every seven days.

But what does this all mean you might ask? In plain terms, the state pays old debts with new debts, and the price of doing so gets reset every week.

Chapter 4: The choice: financing short is a position

Now let’s get to the interesting question. How much of its debt should a government keep in short paper? The Treasury has an advisory committee for exactly this, the TBAC, staffed by the banks and funds that actually buy the paper. Their long-standing guidance is quite straightforward. A sovereign should keep bills at roughly 15 to 20 percent of the total. Below that, you give up the cheapest and most liquid funding there is. Above that, your budget starts swinging with every rate move.

Where does the US stand? Bills currently make up 21.5 percent of marketable debt, which is above the recommended band. The share peaked near 35 percent in the panic of 2008, when the government funded emergency programs with whatever sold fastest. It then fell for years as the Treasury deliberately termed out its debt, reaching 10 percent by 2015. In September 2023 it crossed back above the 20 percent line, and it has stayed above ever since.

_{Bills as a share of marketable debt vs. TBAC band, 2001–2026 · Source: US Treasury MSPD}

The timing of that crossing is worth understanding, because it reveals the logic.

By 2023, the Fed had raised rates at the fastest pace in four decades, and something unusual had happened along the way: The yield curve inverted. That means that short-term borrowing had become more expensive than long-term borrowing.

Normally lenders charge more for longer waits, and the reason is intuition you already have. Nobody would accept the same rate for lending a friend money until next month and until 2056. The longer the money is out of reach, the more can happen to it. Let it be inflation that can slowly but gradually eat the repayment, rates can rise and leave the old loan looking foolish, and life does not pause while you wait.

Time is risk, and risk has a price, which is why the interest curve normally slopes upward from short to long. In 2023 that slope had flipped. For the Treasury that created a strange menu. The cheapest funding on offer was also the longest, but taking it meant signing up for a generation of elevated rates, with no way back once the ink dried.

Issuing bills instead meant paying today’s high short rates only for as long as they lasted, with a free option to refinance cheaper once the Fed cut. Reasonable people can call that prudent flexibility or a rate bet with public money, and in Washington they did argue about exactly that, loudly.

What nobody disputes is that the shift happened, that it has persisted across two administrations and two Treasury secretaries, and that it shows no sign of reversing. The bill share sits near its post-crisis high today. Do you notice something about the two numbers in this paper so far?

Bills are 21.5 percent of the stock, yet 32.6 percent of the debt matures within a year. The gap consists of notes and bonds sold years ago that happen to come due now. One number is a policy choice being made today. The other is the accumulated consequence of decades of choices. Most coverage blurs them into one scary statistic. They are two different things, and this paper needs both. The stock figure actually understates what is happening, because the flow points harder in the same direction: roughly 85 percent of everything the Treasury currently issues is bills. The outstanding share moves slowly because the stock is enormous, but nearly every new financing decision is a short one.

I do not think this is an accident or an oversight. Financing short is a position. A borrower who locks in thirty-year money is indifferent to the next rate decision. A borrower who rolls a third of the balance sheet every year benefits from every rate cut within weeks and pays for every hike just as fast.

The United States has effectively put itself in a floating-rate mortgage. Officials call it neutral debt management. The balance sheet calls it a position: the government is long Fed cuts. It profits when rates fall, and it is exposed when they rise.

Once you see this, previously confusing things start making sense. It explains why fiscal officials have become loudly opinionated about monetary policy, which used to be taboo. It explains why every move at the short end of the curve is now a budget event. And it also explains why the Treasury and the Fed can no longer be analyzed as independent actors because one of them sets the price of the other’s refinancing, every single week.

Chapter 5: What the machine costs

The bill arrives daily. Net interest on the federal debt runs about 3.5 billion dollars per day, roughly one trillion per year. That makes interest the third largest item in the federal budget, behind Social Security and Medicare and ahead of National Defense. About 22 cents of every tax dollar so essentially a quarter of every penny collected goes to servicing the sovereign debt engine.

But to be honest with you, the speed of change is the real lesson here. Five years ago the average interest rate across all marketable Treasury debt was 1.54 percent. Today it is 3.35 percent. It more than doubled in five years, and the reason is the rollover machine itself. Trillions of dollars borrowed in 2020 and 2021 at rates near zero have already come due and been replaced with paper costing 4 percent and more. A mortgage fixed for thirty years would not care what the Fed did last year. Debt that rotates by a third annually absorbs every rate change within months. Each percentage point of policy rate, applied to 10 trillion of annually maturing debt, is on the order of 100 billion dollars of yearly interest cost arriving within the first refinancing cycle. That is roughly a NASA budget every few months, decided by the auction tape.

The same door swings both ways: each future cut would flow back into the budget just as quickly. Right now, the market is not offering one. Futures currently price the Fed’s next move as more likely a hike than a cut, and the new Fed leadership is signaling it would rather be dragged into easing than rush there voluntarily.

That means the sensitivity runs the painful way first. A government financed short does not get to wait out a hiking phase the way a thirty-year borrower would.

_{Interest on public issues, trailing 12-month sum · Source: US Treasury, Interest Expense}

The line crossed one trillion dollars in late 2025 and sits at 1.05 trillion today. Ten years ago it ran a quarter of that.

_{Average interest rate by security class, 2001–2026 · Source: US Treasury MSPD}

This chart makes the mechanics visible. The bills line (short term treasuries) is the volatile one, collapsing toward zero in the easy years and snapping to 4 percent within months of the hiking cycle. The bonds line barely moves, still digesting decisions from decades ago. The total line, the one the budget feels, tracks closer and closer to the volatile one as the overall general debt changes to shorter maturity. That convergence is the whole paper in a single picture.

Chapter 6: Who is on the other side?

A machine that sells 100 billion dollars of paper per auction only works when somebody reliably buys. So who does take the opposite trade on this?

A decade ago, foreign holders, including central banks like China’s and Japan’s, owned about a third of the US Treasury debt. Today they hold about a quarter. Just by looking at the numbers one might think that this is not even that bad and that this is a slow rebalancing. But this is only partly correct. What this means is that the marginal buyer has become domestic. At the short end the dominant buyer is a money market fund. This is a fund that holds only very short, very safe paper and passes the interest to its investors.

_{Foreign & international holdings as a share of total public debt · Source: Fed Financial Accounts (via FRED), US Treasury}

When a broker pays 4 percent on uninvested cash, a money market fund holding Treasury bills is usually the reason. The industry has swollen to 7.9 trillion dollars, because savers finally got paid for cash again, and it holds 3.4 trillion dollars of Treasury securities on the Fed’s own accounting. The rollover machine found its counterparty in the ordinary saver’s brokerage account.

It also further found a newer counterparty in a place traditional finance rarely looks at.

Stablecoins are digital tokens on public blockchains designed to always be worth exactly one dollar, and the companies that issue them hold real reserves so every token is backed. Those reserves sit overwhelmingly in Treasury bills. Tether, the largest issuer, reports 141 billion dollars of bill exposure, which would rank it as the seventeenth largest holder of US government debt in the world, ahead of Germany.

Circle, the second largest, parks roughly 80 percent of its reserves in a fund that buys nothing but bills and overnight loans. Since last year this is not even a preference anymore but law. US stablecoin legislation requires issuers to hold reserves in cash or bills maturing within 93 days. Every new tokenized dollar is, by statute, demand for exactly the paper the rollover machine needs to sell most. A technology built to route around the banking system ended up financing the government at its most rate-sensitive point, and Washington wrote the arrangement into law.

I find this the single most underappreciated link between the two worlds of finance.

And quietly, a third buyer has re-entered the room: the Federal Reserve itself. The Fed holds a large portfolio of Treasuries as a byproduct of past crisis programs, and as that portfolio is maintained and reinvested, its purchases have shifted toward bills.

The Treasury’s own advisory materials now model Fed bill buying as a structural source of demand at the front end. Let’s read that back slowly: the institution that sets the price of short money is also becoming a steady buyer of the short paper whose price it sets.

Nothing about that is hidden or improper, but it tightens the loop one more turn. Is this kind of demand healthy? One useful tool is the bid-to-cover ratio, which measures how many dollars of bids arrive per dollar of paper sold.

On four-week bills it sits near 2.7, meaning nearly three dollars chase every dollar available. The machine is not straining. What has changed is who keeps it running, from foreign governments toward domestic funds, crypto reserves, and the central bank itself. Composition like that moves slowly, until it does not, which is exactly why it belongs on a monthly watchlist rather than in a one-time headline.

Chapter 7: The honest counterargument

Never trust a story that hides its strongest objection. So here it is, in its best form.

Measured differently, US debt has rarely looked so patient. The measure in question is the weighted average maturity, and it is less technical than it sounds: take every outstanding security, ask how many years it has left, and average those lifetimes, giving big securities more weight than small ones. The answer today is about 71 months, almost six years, and close to the highest level in 26 years. On top of that, roughly a third of the debt needs no refinancing for at least five years. None of this is an accident either. The Treasury of the 2010s deliberately used the era of near-zero rates to lock in long, cheap money, and that cushion is real. Anyone pointing at it is pointing at a true thing.

_{Weighted average maturity of marketable debt · Source: US Treasury MSPD, own calculation}

The second objection is just as fair. Ten trillion dollars of maturing debt sounds like ten trillion dollars that has to be found somewhere, and it is not. The investors holding that paper already own it, and when it matures, most of them put the money straight back into the next auction, almost automatically. Think of the money market fund whose bill matures on Tuesday: it does not take the cash and leave, because holding safe short paper is its entire job. It buys another bill on Tuesday. Refinancing is mostly this, the same money rotating from old paper into new. The true nightmare scenario, an auction where nobody shows up, is remote for the country that issues the asset the whole financial world treats as its safest parking spot.

So which is it? Is the debt patient or is it urgent? The honest answer is both, because the two claims measure different things, and seeing why is the most useful thing this paper can teach.

The average is long. The front is heavy. Both at once. Here is how: the average gets pulled up by a relatively small amount of very old, very long paper, thirty-year bonds sold decades ago that still have decades to run. A few far-away dates can stretch an average a long way. Meanwhile the bulk of the money sits at short maturities. My own calculation from the Treasury’s records puts 67.7 percent of all marketable debt inside five years. So when someone quotes the near-record average maturity, they are describing the tail. When this paper describes a third rolling over every year, it is describing the front. Both use real numbers. They just answer different questions. The question that matters for the budget is the second one, because the front is where debt gets repriced, and repricing is what moves the interest bill.

That is why my conclusion is deliberately not a crisis call. I do not think the United States faces a funding crisis. What I think is this: the country has wired its budget directly to the Fed’s decisions, so that every rate move flows into government interest costs within months. A wire is not a crisis. A wire is a connection, a transmission channel, and knowing where the transmission channels run is most of what reading markets actually means.

One more thing, because pretending not to have a view would be its own kind of dishonesty. I will tell you where I personally lean. I think the political pressure to solve this tension with easier money, lower rates sooner than inflation alone would justify, will grow with every refinancing cycle. Understand that this is a conviction statement. The data in this paper shows the pressure exists. It cannot show how the fight ends.

And a conviction that cannot be wrong is worthless, so here is what would weaken mine. The bill share drifting back inside the advisory band would weaken it, because it would mean the Treasury is unwinding the short position rather than deepening it. The interest line flattening while rates stay firm would weaken it, because it would mean the budget is absorbing the pain without political rescue. Auction demand holding through a full hiking phase without official support would weaken it too, because it would mean the machine does not need easier money to keep running. The monitor below tracks exactly these things, which means it exists partly to keep me honest. Month by month, the same five numbers will show whether that pressure is turning into policy, or whether the credibility camp holds the line.

Chapter 8: What you can do with this

This is not investment advice, and it deliberately contains no forecast. What it gives you is a lens, and a lens is only worth something when you know where to point it. Four uses.

Read the Fed differently. At the next rate decision, ask the question this paper forces: who is being financed at this rate, and at what horizon? A central bank whose sovereign rolls a third of its debt annually is not setting rates in a vacuum. It is setting its own government’s refinancing cost. Every debate about central bank independence has this concrete, dollars-per-week mechanism underneath it, and every press conference reads differently once you hold that in mind.

Understand your own cash. The yield on a money market fund, a broker’s cash sweep, even a decent savings account, all of it sits downstream of the bill auctions described here. Anyone earning 4 percent on parked cash is, through one or two intermediaries, one of the buyers keeping the rollover machine running. That is the actual plumbing, and it also means the income on your cash and the interest bill of your government are the same number seen from two sides.

Place the crypto story correctly. The next time someone calls stablecoins a sideshow, you have the number: the largest issuer alone holds more US government debt than Germany. The next time someone calls crypto disconnected from the real economy, you know the connection is written into federal law. Both maximalist stories, crypto as pure escape and crypto as pure casino, break on this fact.

Filter the noise. You will keep encountering two kinds of coverage. Panic pieces shouting that a third of the debt is about to hit a wall, and reassurance pieces pointing at record-long average maturities. Both quote real numbers, both are incomplete, and you now know which number answers which question. That filter alone puts you ahead of most of the commentary.

Chapter 9: The monitor

A frame is only useful when you can hold events against it, so this paper comes with a fixed monthly check. The Treasury publishes its debt statement on the fourth business day of each month. On that day, Rollover Watch updates five numbers. Always the same five, always in the same order.

The bill’s share of marketable debt, currently 21.5 percent, read against the advisors’ 15 to 20 percent band. The share of debt maturing within one year, currently 32.6 percent. The average interest rate on the debt, currently 3.35 percent. The interest bill itself, which just crossed one trillion dollars over the past twelve months and sits at 1.05 trillion today. And the bid-to-cover on four-week bills as the pulse of demand, currently near 2.7. Every figure is computed from the Treasury’s published security-level data with the same pipeline behind every chart in this paper, and a sixth element each month is one sentence pulling the month’s main event through this lens.

These five are not a dashboard for its own sake. They are chosen so that the argument of this paper, including my own stated lean, can be checked against them and fail. The bills share tests whether the short position deepens or unwinds. The interest line tests whether the pressure is real. The bid-to-cover tests whether the machine needs help. Charts that move slower, like the average maturity of the whole stock, will reappear here whenever they have something new to say, but the monthly core stays fixed.

A government that rolls a third of its debt every year answers the question of every Fed meeting before it is asked. From next month on, we will watch it precisely and update on a recurring basis. Stay tuned.

_{Data: US Treasury, Monthly Statement of the Public Debt and auction records, as of June 30, 2026. Money market figures from the Federal Reserve’s financial accounts and ICI. Foreign holdings from the Federal Reserve Financial Accounts via FRED, as of Q4 2025. Stablecoin reserve figures from issuer attestations, which are point-in-time statements rather than full audits. Corroborating estimates from Apollo Global Management, the GAO and the Treasury Borrowing Advisory Committee. All calculations reproducible from published security-level data. Nothing here is investment advice. It is a lens.}

This was a week where the plumbing dictated everything. A thirteenth straight night of US strikes on Iran, plus Houthi attacks on two Saudi oil tankers in the Red Sea, sent Brent crude (the global oil benchmark) storming toward $100 a barrel by Thursday.

Higher oil feeds directly into headline inflation, and inflation is the one thing that keeps a central bank from cutting interest rates. So bond investors did the logical thing: they sold Treasuries, pushing the US 10-year yield to 4.70%, its highest since January 2025, and began pricing a real chance the Federal Reserve actually raises rates next week rather than cutting.

Rising yields make future company profits worth less today, and they landed at the exact moment two of the largest US firms spooked the market over how much they are spending on artificial intelligence. Alphabet and Tesla both fell hard Thursday, dragging the Magnificent Seven megacaps down by roughly $800 billion in a single session.

Friday brought relief: reports that Pakistan, backed by China, was trying to revive US-Iran talks knocked oil back below $96, letting stocks steady into the close. The S&P 500 ended the week almost flat, the tech-heavy Nasdaq around 2% lower, and the bond market walked away having quietly changed its mind about where the Fed is heading.

The Scoreboard

A basis point (bp) is one hundredth of a percentage point, so +15 bp means a yield rose from 4.55% to 4.70%. Yields and bond prices move in opposite directions: when a yield rises, the price of the bond you already own falls, and the cost of new borrowing for governments, companies and mortgage holders goes up. Equities, currencies and commodities are shown in percent. US equities and Treasury yields are the 22:00 CET close; European indices and the Bund settle earlier at 17:30 CET; foreign exchange and crypto are a fixed 22:00 CET snapshot. Levels marked ≈ are approximate where the official settlement was not yet confirmed.

Reading the Board

Equities split along a clear line. US large caps held up because strong company earnings offset the oil and rates pressure, leaving the S&P 500 barely changed. The damage sat in technology, where the Nasdaq’s 2% weekly drop came almost entirely from chipmakers and the AI-spending scare. Europe drifted lower as higher energy costs raised the odds of more rate increases from its own central bank.

Rates were the story. Both the 2-year yield (the market’s best summary of where the Fed is heading over the next two years) and the 10-year yield climbed together, a sign investors now expect tighter policy for longer. The German Bund, the euro area’s benchmark government bond, touched its highest level since 2011 for the same reason: oil-driven inflation forcing central banks toward the exit.

Foreign exchange stayed quiet by comparison. The dollar firmed slightly, helped by its role as a safe haven during Middle East conflict, and the euro slipped below $1.14, near its weakest in a year, as traders weighed Europe’s exposure to the energy shock.

Commodities were the pressure source. Brent’s near 9% weekly jump captures the entire chain of events, even after Friday’s pullback. Gold added a modest 0.8%, doing its usual job as insurance when geopolitics turns dangerous.

Crypto went nowhere fast. Bitcoin drifted just below $64,150 and finished slightly down, caught between improving internal demand and the same macro storm hitting every risky asset.

Three Brews That Mattered

1. Oil is now the Fed’s problem.
The mechanism is direct. When a barrel of Brent jumps roughly 30% above its pre-conflict level, petrol and diesel prices follow within weeks, and that shows up in the inflation numbers the Fed watches. A central bank cannot cut rates into rising inflation without losing credibility, so every dollar oil gains pushes rate-cut hopes further out and pulls a rate hike closer. That is why a war thousands of miles away moved your mortgage rate this week: the 10-year yield, which sets the tone for US home loans, rose because oil made a Fed cut harder to justify. Goldman Sachs flagged a “VaR shock” risk into the weekend, meaning the size of the rates move was large enough to force leveraged funds to cut positions to control their measured risk, which can feed on itself. Why it matters: the path of oil, more than any Fed speech, now decides the path of rates.

2. The AI bill is coming due.
Alphabet and Tesla did not disappoint on profits so much as on spending. Alphabet signalled capital expenditure (the money a company pours into data centres and chips) far above what analysts expected, with some estimates running toward $325–375 billion against a Street figure nearer $250 billion. Investors reacted by asking the uncomfortable question of when all that hardware starts earning its keep. The answer for now is later than hoped, which is why the Magnificent Seven shed around $800 billion on Thursday. The people who ultimately pay are shareholders, whose future profits are being reinvested today at the exact moment higher yields make those distant profits worth less. Why it matters: the market has started charging megacaps for AI ambition rather than rewarding it.

3. A jobs number rewrote the Fed math.
Weekly jobless claims, the count of Americans filing for unemployment benefits for the first time, fell to 187,000, the lowest in nearly 60 years and far below the 212,000 economists expected. A labour market this tight gives the Fed no reason to rush to support the economy, and every reason to worry that low unemployment plus rising oil keeps inflation sticky. Within hours, the probability of a July rate increase priced by futures traders jumped past one in three, from near zero only weeks ago. Why it matters: the Fed’s excuse to cut has evaporated, and the meeting on July 28–29 just became live.

Data Check

“Consensus” is the average forecast economists submit before a release. Markets move on the gap between the actual number and that expectation, not the number on its own.

The services survey was the standout, running well ahead of forecast and pointing to growth of roughly 2% on an annualised basis. One caveat worth keeping: S&P Global noted July activity was flattered by hospitality spending around the FIFA World Cup and the USA 250 anniversary, so some of the strength may fade. The European Central Bank left its main rate unchanged as expected, yet the message underneath was the harder part, with policymakers including Bundesbank chief Joachim Nagel warning that inflation risks from energy remain elevated.

The Onchain Pour

Bitcoin spent the week doing something quietly interesting while everyone watched oil. Its price sat close to its “realized price,” the average level at which all coins last moved on-chain, which acts as a rough cost basis for the whole market and often marks where committed holders step in. Long-term holders, wallets that have not sold in months, accumulated at the fastest pace in six years, a sign the strongest hands are buying weakness rather than selling it. The demand is not only retail: exchange-traded products that hold Bitcoin for investors have pulled in around $5 billion more than their gold equivalents since March, a genuine rotation from the old safe haven into the new one. Against that, real risks remain. Analyst André Dragosch called Bitcoin the “canary in the macro coal mine,” pointing to the Strait of Hormuz, rising yields and stress in the credit that funds AI data centres as the things that could still drag it lower, while fresh worries about quantum computing weighed on sentiment. On the policy side, the CLARITY Act, the bill that would finally define which US regulator oversees which crypto asset, lost some momentum after last-minute amendments, even as exchanges pushed the Senate to pass it. The setup is a market with improving foundations sitting inside a hostile macro backdrop, waiting to see which one breaks first.

Last Drop

Next week is the one that counts. The Fed decides on July 28–29 with a rate hike no longer unthinkable, and the numbers in this Scoreboard are the ones traders will be pricing off. Sunday’s Weekly Steep will go deeper on what a hiking Fed would mean for a government that has quietly turned itself into a floating-rate borrower, and why the bond market’s mood this week is the real signal to watch. Until then, keep an eye on the oil tape. It is running the show.