The labour market in this data is not softening, and the trade being described as a stagflation squeeze is mostly a positioning flush. Payrolls fell by 23.000 thousand while unemployment dropped to 4.1%. Those two moves are only compatible if the labour force is shrinking faster than employment. That is a supply story, not a demand collapse, and it is the least dovish combination the Fed could have received. A shrinking workforce with a falling jobless rate keeps wage pressure alive into an energy spike.

The gold and silver moves are being read backwards. Precious metals down 5.93% and 5.98% over five days while the ten-year breakeven rose 0.040 to 2.35% is not a market retiring its inflation hedge. It is leverage being taken off a crowded trade. Nothing about a five-day drawdown in bullion tells you what the price level does over the next year, and the metals complex has a long history of violent unwinds inside intact trends.

What deserves more attention than either is the front end. Fed funds effective sits at 3.63% and the two-year yields 4.34%. The market is not pricing the easing path. It is pricing the opposite, and the balance sheet agrees: total assets fell to 6,730,912 million dollars, down 14,787 million on the week. Runoff continues without commentary. That is the honest read of the reaction function right now, and it is more restrictive than the statements suggest.

The crude move also needs its inventory context. Crude stocks excluding the SPR rose on the week while gasoline drew 2,536 MBBL and distillate drew 2,228 MBBL. Tightness is in products, not in barrels. Refining capacity and middle distillate cover are doing the work, which is exactly the kind of squeeze that transmits to headline CPI fast and then fades from the core, and exactly what the pandemic-era episode taught policymakers to look through. They will look through this one too, and this time they will probably be right about the pass-through even if they are wrong about the level.

Where the pressure actually sits is the long end. Ten-year at 4.75% against a 2.35% breakeven is a wide real yield, and high yield OAS at 2.63 says credit is not asking for compensation. That combination means the term premium is carrying the adjustment, not default risk. Duration is being repriced for supply.

So: no cut on this data, and the two-year is right to sit above the funds rate. The fiscal problem shows up at the ten-year point first. Sell the notion that soft payrolls buy an easing cycle when unemployment is falling at the same time.

Nine percent on WTI in five sessions is not an inflation impulse, and treating it as one is how the Fed gets the next call wrong.

Look at what the barrels did. US crude stocks excluding the SPR rose by 95 thousand barrels in the week to 21 August. Gasoline drew 2,536 and distillate drew 2,228, which is what the end of August looks like every year. The EIA is publishing pieces about record natural gas production in 2026, longer Permian wells, LNG exports up 23% in the first half, eight new liquids pipelines finished since the start of 2025, and Dangote pulling barrels out of Nigeria. That is not a supply-constrained complex. Brent at 96.524 and WTI at 90.273 came off positioning and risk pricing, and the tell is gold down 5.93% and silver down 5.98% in the same week. Money that genuinely fears an inflation shock does not liquidate metals to buy crude. It buys both.

The breakeven moved 0.040 to 2.35%. Four basis points. If you want to call that a hawkish repricing, you are reading noise as signal.

The core CPI print is a different matter and deserves to be taken on its own terms. Core rose 0.724 on the index against 0.245 for all items. Core running well above headline means whatever is sticky in US prices has nothing to do with energy, which is precisely why a crude spike is the wrong trigger for a policy response. You do not treat a services problem with a demand squeeze aimed at a barrel of oil.

And the labour market is already doing the squeezing. Payrolls fell by 23 thousand in July while unemployment dropped 0.100 to 4.1%. Those two things only reconcile through people leaving the workforce. An unemployment rate that improves because the denominator shrank is not strength, and it is the specific failure mode of reading headline unemployment as the whole employment mandate. The funds rate sits at 3.63%, below the 2y at 4.34% and the 10y at 4.75%. High yield spreads at 2.63 say credit sees nothing worth worrying about. There is no financial stress argument for holding here, only an inflation argument, and the inflation argument now rests on core services rather than on the thing that moved this week.

Cut. The energy move will not be in the year-on-year comparison for long and the participation damage will be.

One more thing the desk narrative skips. The broad dollar index went to 118.7479 and Hungary paid for it: USD/HUF up 2.46% on the week, Poland up 1.07%, South Africa up 0.91%. Nothing changed in Budapest or Pretoria. A US rate expectation moved and Central Europe absorbed it. When those currencies are described next month as needing discipline, remember where the shock originated.

The stagflation read requires the inflation market to believe it, and it does not. WTI up 9.28% in five sessions bought exactly four basis points of ten-year breakeven, to 2.35%. The two-year yield moved zero, sitting at 4.34% against effective funds of 3.63%. The 10s2s spread flattened a basis point to 0.40. A genuine supply-led inflation impulse steepens the curve and repricing starts at the front end. Neither happened. What happened is that a book got flattened.

Look at the metals. Gold -5.93%, silver -5.98%. Silver trades at a multiple of gold's beta in almost every directional move; when the two print the same drawdown to within five basis points, nobody is expressing a view on metals. Someone reduced notional pro rata. And the ten-year yield rose two basis points while breakevens rose four, meaning the real yield fell. Gold falling six percent into falling real yields is the wrong sign under any mapping I use. I cannot make that arithmetic work as a macro repricing. It works as forced selling.

Credit confirms it. High yield OAS at 2.63, wider by three basis points. That is a rounding error. If the market were pricing an energy shock into a labour market that shed 23 thousand payrolls, spreads would be the first thing to go, not the last.

The interesting leg is cross-yen. Crude up nine percent is a terms-of-trade hit to Japan, and the yen firmed anyway: EUR/JPY -0.40%, GBP/JPY -0.52% on the day, USD/JPY down 0.27% against a broad dollar that gained, with USD/CHF up 1.20% on the week. Yen strength on the crosses into a dollar bid and an energy spike is not a Japan story. It is deleveraging in the funding currency. HUF at 318.246, down 2.46% in five days, is the same trade from the other end: the energy-importing carry leg goes first and it goes alone, because MXN moved 0.26% and did nothing.

The inventory mix argues the crude move is not a crude balance story at all. US crude stocks built 95 thousand barrels while gasoline drew 2,536 and distillate 2,228. Products are tight, crude is not. I do not have the futures curve, and without it I will not call this a shortage.

So the position: the energy rally is the trigger, not the content, and long crude is the same book as long gold and short yen. It breaks last, not first. Brent already gave back 0.96% on the day. Anyone adding to inflation hedges here is buying the leg of the trade that has not been margined yet.