Three Fed Members Just Voted to Hike

Reflation held the top spot in our regime model for a thirteenth week, and the conviction underneath it cracked. Reflation is the setup this letter has described since May: growth firming and dragging inflation up with it, the combination that historically keeps the Fed from cutting. This week the probability the model attaches to it fell to 74 percent from 86, a reading that had sat frozen through nine straight daily runs before breaking on Friday, while Disinflation, the benign case where inflation falls while growth holds and the Fed gets room to ease, climbed to 12 percent from 8. What remains of the label leans heavily on the model's memory. The reading rewards a regime for having persisted; strip that persistence weighting out, and this week's evidence alone would put Reflation near 15 percent. Of the five independent academic frameworks we run alongside our own, only one still votes for it.

Last week's letter set a test: two clocks run on one Treasury curve. Oil drives the front end, the short-maturity yields that track Fed policy; government borrowing weighs on the back, the long maturities that price the state of the nation's finances. If the long end ever rose faster than the front in a week when oil was quiet, the driver had changed. This week ran the experiment. Crude fell 5.5 percent, and the front end that oil had pushed to a seventeen-month high came back hard: the three-month bill fell 13 basis points to 3.82 percent and the two-year fell 14 to 4.23. (A basis point is a hundredth of a percentage point.) The long end kept going the other way. The thirty-year rose to 5.21 percent, touched 5.23 on Friday, its highest in nineteen years, and term premium, the extra compensation investors demand for locking money up in long-dated government debt, rose about six basis points after finishing flat last week. The oil clock stopped. The fiscal clock kept running.

The Fed spent the week confirming its own direction. The committee held its policy rate at 3.50 to 3.75 percent on Wednesday, a fifth consecutive unchanged meeting, on a nine-to-three vote; the dissenters, Hammack, Kashkari and Logan, wanted a quarter-point hike. A hold delivered with three votes for a hike is a message about direction. It arrived in the same week that our alignment reading, which scores the Fed's stance against the stance the data would justify, moved for the first time in fifteen weeks. Since mid-April it had scored the two as trapped together: a boxed-in Fed facing data too mixed to force its hand. It now reads a very hawkish Fed set against data that has turned outright dovish. This week the two sides let go of each other.

Thirteen weeks makes this Reflation run more than three times its normal length. The median run across the 114 episodes in the model's history is four weeks, and the most likely successor remains Disinflation, which has historically taken about 43 percent of the exits. The transition-risk flag we track has now been on for two weeks, and both of the nearest alternative regimes, a financial-conditions squeeze on one side and Disinflation on the other, have already met every threshold condition on their lists. The label is standing; the floor under it has thinned.

The Fed's benchmark scenario, a Soft Landing in which inflation returns to 2 percent without a recession, extended its run. Our daily read on it has come back positive for nineteen consecutive days, the longest and widest positive stretch since this letter began tracking it, and this is the third letter in a row carrying it. Persistence and magnitude are the two things that separate a signal from noise. Both are still present.

The soft-landing evidence keeps improving, and the Fed keeps moving away from it. The Fed's own June minutes flagged persistent inflation as a salient risk, three members just voted to hike, and market pricing of the rate path still builds in roughly three more hikes over the next twelve months, an implied year-end rate of 4.23 percent against an actual policy rate of 3.63. Our pricing model flags that gap as a mispricing: a market positioned for extended tightness sitting on top of data that has swung toward cooling. Last week this letter argued that if the inflation numbers confirmed the cooling, a market priced for tightness would eventually have to reprice toward cuts. No new inflation print has arrived. The front end began repricing anyway, 14 basis points of it, in a week when the Fed's message pointed the other direction. Either the market is early or the data is lying. August 12 begins to settle it.

The biggest IPO in history is live — $1.75T valuation, $135 open, $75B raised. And history says the retail investors who chase day-one hype are the ones who get burned.

The SpaceX run created thousands of new millionaires. But it wasn't the people buying at the open. It was the ones positioned early, in the names set to ride the wave.

Our analyst pinpointed 3 stocks positioned to ride the SpaceX wave — with entry guidance and price targets, a bonus 4th pick (the most undervalued name in the sector), and a 3-phase playbook for when to buy and when to sell.

Over 2,500 investors have already read the report. Claim your free copy here.

Get the Free Report

Thursday's second-quarter GDP report managed to disappoint both sides of the Fed's mandate at once. Growth came in at 1.5 percent annualized against 2.1 expected. The report's broadest price gauge, covering everything Americans buy including imports, accelerated to 5.7 percent from 3.6, and the PCE price index, the inflation measure the Fed formally targets, rose to 5.1 percent from 4.6. Yet inside the same report, core PCE, which strips out food and energy to show where inflation would settle on its own, decelerated a full point to 3.4 percent from 4.4 in the first quarter. The oil and tariff shock is arriving in the headline numbers while the inflation that demand generates cools underneath it.

One accounting note keeps the growth figure honest. The BEA attributes much of the quarter's decline in federal spending to crude sales out of the Strategic Petroleum Reserve, the government's emergency oil stockpile, and those sales count against government consumption. Part of the slowdown is bookkeeping: the oil buffer drawdown passing through the federal ledger on its way to market. The consumer carried what growth there was. Private domestic final sales, consumer spending plus business investment and the cleanest read on underlying private demand, accelerated to 3.9 percent from 1.7, and consumption contributed more than the entire net change.

Brent crude spent two days above 100 dollars, closing at 105.32 on July 23 and 100.31 on July 24. Last week this letter, working from press quotes, described that as a touch of 100 and a settle below; the primary price series corrects the record, and so do we. The confirmation we set for a supply squeeze, two consecutive weeks above 100, still has not fired, and crude has since backed away from the line, trading near 90 to 93 through the back half of the week. Even after the 5.5 percent weekly decline, crude is up 25 percent on the month and 87 percent this year. Copper went the other way, gaining 3.2 percent in the week oil fell. It is now the only asset class we track doing better than equities over the past three months.

Japan defended its currency at record scale. The Bank of Japan held its policy rate at 1.00 percent on Friday, eight votes to one, with the dissenter proposing a hike to 1.25, and warned that core inflation could run clearly above 2 percent from the second half of fiscal 2026. The day before the decision, Japanese authorities bought yen at what central bank account data implies was record single-day scale, roughly 8.45 trillion yen, between 53 and 59 billion dollars, enough to hand the dollar its biggest one-day fall against the yen since 2022. Japan's household and institutional savings are among the world's largest pools of lendable money. A finance ministry willing to spend at that scale to defend the currency is a new fact for every market that borrows from those pools.

Beneath a calm index, the equity market kept de-risking. Microsoft rose 16 percent on Thursday, its best session since 2008 and the largest single-day gain in market value on record, roughly 480 billion dollars, on the strength of its cloud results, and the S&P 500 finished the week just 1.4 percent below its June high. The Nasdaq-100 sits 7.7 percent below its own. Underneath, the market's steadiest stocks have beaten its most aggressive by 9.5 percentage points over the past month, and momentum, the strategy of owning whatever has been winning, is trailing the index by nearly 9 points. That combination is the signature of large investors cutting risk without leaving the market. It began only this month.

The oil buffer convergence is eleven days out. Between August 13 and 15, three cushions that have been absorbing the supply shock go away at once: the 172-million-barrel Strategic Petroleum Reserve exchange completes its deliveries, the International Energy Agency's coordinated 400-million-barrel release runs out, and the sixty-day clock on the nuclear talks lapses. What sits underneath is thin. The reserve stands at 307.7 million barrels, its lowest since March 1983 and about 58 million above its operational floor. Commercial crude inventories are 7 percent below their five-year seasonal average, and refineries are running at 97.2 percent of capacity, which leaves almost no slack anywhere in the system.

The buffer arithmetic is the high-confidence part of this story. The price direction stays deliberately uncalled, because the testing keeps cutting against the intuitive read. Across six modern strategic releases, the median outcome twelve months out was a 16 percent price decline, with prices high enough that households and industry simply burned less fuel. This week a second tested precedent fired on the same side: producers in the WTI futures market are net long by a record margin, and across the eight comparable extremes since 2006, crude fell a median of 7 percent over the following six months and 8 percent over twelve, though it typically kept rising through the first three. A market bracing for a supply cliff and a positioning record that historically precedes weakness are both in the data at once.

Credit widened for a second week, and last week's caution still applies. High-yield spreads, the extra yield investors demand to lend to the weakest corporate borrowers, moved to 284 basis points from 277, and the gap between what the weakest and the strongest corporate borrowers pay stretched to 204 basis points from 198. The level is still historically tight, the direction has been widening for two weeks, and our read is that this is broad de-risking: none of the mechanisms that mark a genuine credit break, fund redemption gates, drawn bank credit lines, failed refinancings, moved this week. Monday's Senior Loan Officer Survey, the Fed's quarterly poll of bank lending standards, is the first scheduled reading that could change that.

The third warning is the backdrop, because it sets the depth available to any decline. The equity risk premium, the extra return stocks offer over safe Treasuries, has been negative for over two months. Across the six comparable episodes since 1882, markets corrected more than 10 percent within six months in only two of them; in 2003 and 2022 they rallied straight through. What the backdrop does measure is depth. With prices at 40.9 times a decade of smoothed earnings, the 99th percentile of readings back to 1881 against a long-run average of 17.4, two-thirds of comparable months saw a real drawdown of more than 10 percent within a year, and the arithmetic implies a ten-year forward real return of 1.7 percent. The reading measures the size of the air pocket. It stays silent on the date.

The credit warning gets all three of its live tests inside five days. The Senior Loan Officer Survey lands Monday afternoon, job openings Tuesday, and initial jobless claims Thursday, from a base of 197,000 against the sustained 300,000 level that has historically marked recession territory. No Fed meeting intervenes; the next decision is September 16, forty-five days out, which leaves the data to speak for itself.

Friday's July payrolls report is the week's only high-impact release, and its composition will matter more than its headline. June's 57,000 gain leaned on education, health care and government, which together supplied more than the entire number, while leisure and hospitality lost 61,000 positions, one of the most extreme readings in our payroll data. Composition like that means a solid headline can sit on top of a cooling private economy. A weak print, if one comes, deserves its own discipline: our testing of past negative payroll months found no reliable recession signal in any single print. The reading that matters more arrives the following Wednesday. The August 12 CPI is the first fresh monthly inflation print in four weeks, and in the gap the only new inflation inputs anyone has had are the oil price, the tariff schedule, and the quarterly averages buried in Thursday's GDP report. The buffer convergence window opens the day after.

The week delivered a clean resolution and a harder puzzle. The curve test resolved: oil went quiet, the front end fell, the long end rose to a nineteen-year high, and the driver of the yield move changed exactly as the framework said it would. The puzzle is harder. The soft-landing evidence is at its strongest reading in this letter's history at the very moment the regime label that contradicts it is at its weakest, and the Fed just produced its most hawkish vote of the cycle into data our model scores as calling for the opposite. Someone in that triangle is wrong. The market has started picking a side, taking back 14 basis points of front-end tightening in a single week without a shred of new inflation data to justify it.

August 12 is the arbiter. A cool CPI print validates the front end's move, confirms the turn the model made on July 31, and strands a market still priced for three more hikes. A hot one vindicates the dissenters and restarts the repricing that ran through July. And the seventy-two hours that follow it empty the oil buffers that have been absorbing the supply shock since spring. The data that decides the argument and the event that stress-tests it land inside the same five trading days.

This report is published by Benjamin Capital Research for educational and informational purposes only. It does not constitute investment advice, a recommendation, or a solicitation to buy or sell any security. All positioning commentary reflects historical patterns and educational analysis, not personal recommendations. Past performance does not guarantee future results. Always consult a qualified financial advisor before making investment decisions.