Warsh said little about September, but markets heard plenty. Here’s why Treasury yields jumped and why the Fed’s inflation tool may be aimed at the wrong problem.
KEY TAKEAWAYS
Warsh’s Jackson Hole speech contained very little explicit forward guidance, yet markets interpreted his emphasis on broad inflation pressure as decidedly hawkish. The reaction demonstrated how central-bank tone can matter almost as much as actual policy commitments.
Warsh focused on the breadth of inflation across PCE components rather than simply headline or core inflation. His concern is that too much of the inflation basket remains above levels consistent with the Fed’s target.
Markets rapidly repriced September after the speech. Short-term Treasury yields jumped, gold and bitcoin fell, and the probability assigned to a rate hike moved sharply higher.
Much of the inflation currently troubling the Fed comes from areas such as energy, housing, and healthcare. Those categories are influenced heavily by supply constraints, geopolitics, and institutional pricing rather than simply excessive credit-driven demand.
Higher rates can restrain demand, but they cannot directly create more oil, housing, or healthcare capacity. Investors therefore have to distinguish between what the Fed can influence and what remains largely outside monetary policy’s reach.
MY HOT TAKES
Warsh’s communication has improved, but the market is still doing much of the interpretive work for him. The Fed Chair can move asset prices simply by emphasizing risks while refusing to define his actual reaction function.
The market’s September repricing may say as much about investor psychology as it does about monetary policy. No major new economic data was necessary for traders to transform a hike from unlikely into more likely than not.
Bitcoin continues to behave far more like a liquidity-sensitive risk asset than an independent store of value when Fed expectations turn hawkish. Calling it “digital gold” does not change how it actually trades.
The Fed’s traditional interest-rate tool is poorly suited to several important sources of current inflation. Using tighter monetary policy against supply-side inflation risks weakening unrelated parts of the economy without fixing the underlying shortage.
Investors should remain cautious about assets that depend heavily on falling interest rates. Until the Fed provides clearer evidence that easing is coming, maintaining liquidity and avoiding excessive duration exposure makes sense.
You can quote me: “Raising rates to fight that kind of inflation doesn't fix the plumbing. It just turns down the pressure everywhere else in the house while the leak keeps leaking.”
Hall of mirrors. Help wanted: Extremely bright and successful economist or finance expert wanted to run the central bank–must be great with numbers and have the ability to speak for 30 minutes and say nothing–absolutely nothing–while making the world believe that you are credible.
The bar for Warsh’s Friday keynote at Jackson Hole was not all that high. He got off to a rough start with his first spate of public addresses leaving markets and analysts more confused than confident. On Friday, he just needed to tighten up his message. He could have quite easily asked his favorite AI chatbot to look at what all the analysts said about his performance and propose ways to improve the low hanging fruit.
Kevin Warsh stood at a podium in Wyoming on Friday and really did say nothing. No date. No number. No hint of where the next Fed meeting lands. And yet by the time he sat down, the two-year Treasury yield had jumped, gold had dropped 3%, bitcoin had shed a couple grand, and traders had repriced the odds of a September rate hike by twenty points. That's the trick of central banking barely 100 days into the Warsh era–he's figured out how to move markets by describing, in exquisite detail, all the things he refuses to tell you. I've been doing this long enough to admire the craftsmanship even through my squinting eyes.
Warsh himself gave us the name for what just happened. He called it a "hall of mirrors"--the Fed watching market prices for guidance while the market watches the Fed for guidance, each side hoping the other one actually knows something. Nobody does, not really. But everybody has to trade anyway, so Friday became a masterclass in reading tone instead of substance. I'll walk you through what he actually said, what it did to markets, and, because I've never been shy about this, why I think the tool he's reaching for can't fix half the inflation he's worried about.
Let's start with the substance, because there was some, buried under all that studied vagueness. Warsh's preferred measure is the Personal Consumption Expenditures price index, PCE for short–the Fed's official inflation gauge, distinct from the more commonly quoted Consumer Price Index. The Fed has always favored PCE, but he left markets unsure after a vague reference at an FOMC presser a few months back. He told the room he likes to break the PCE basket down into its individual components, 199 of them, and look at how many are running hot. His finding: 54% of those components have shown price increases above 3% over the past twelve months, and 49% over the past six. He was careful to note that's well below the post-pandemic peak, when something like 77% of the basket was running that hot. But it's also well above the roughly 32% share that was typical in the two decades before COVID ever entered the vocabulary. Headline PCE inflation sits at 3.7% year over year. Core, which strips out food and energy, is at 3.3%. Both numbers came in a touch better than expected this summer. Warsh's response, more or less verbatim: that doesn't tell him underlying trends have meaningfully improved. And then, the closest thing to a mission statement he offered all day–the Fed must be confident inflation is moving toward its target, clearly and at sufficient speed, or "we have work to do."
That's it. That's the whole hawkish turn. No reaction function, no dot plot preview, no promise. Just a chairman disaggregating a spreadsheet in public and letting everyone else do the emotional labor of deciding what it means.
And they did. The two-year yield, which tends to move fastest on rate expectations, jumped more than twelve basis points to 4.35%. Side note, in case you don’t know, a basis point is the Wall Street way of slicing a percent into hundredths, so twelve of them is a real move for a single speech–especially in the two-year. 😉 The ten-year rose about five basis points to 4.73%, and the thirty-year ticked up two to 5.21%. Gold, which had been pushing toward three-month highs on hopes for a gentler Fed, fell roughly 3% as the dollar caught a bid. Bitcoin slid close to 2%, down to around $79,200, proving once again that crypto still trades like a leveraged bet on Fed dovishness no matter how many people insist it's digital gold. Equities were more muted with the Nasdaq off about six-tenths of a percent, the S&P down a couple ticks, which tells you the stock market wasn't nearly as spooked as the bond and currency desks were. Stocks have heard hawkish Fed speeches before. They've learned to wait for the actual meeting.
And the meeting is where this gets interesting. Before Warsh opened his mouth, futures markets were pricing something like a 35% to 40% chance of a rate hike at the September 16 decision. Within the hour, that jumped past 56%. By this morning, it's sitting at 60% on the CME's FedWatch tool. In three trading sessions, "probably not" became "more likely than not"--without a single new data point arriving in between. That's the hall of mirrors working exactly as advertised.
My long-time followers know that I'm not convinced raising rates does what Warsh needs it to do. Look at where his own 199 components are running hottest, and you'll notice a pattern: the categories driving the overrun aren't the ones a higher Fed Funds rate can touch. Energy has been jerked around by the Strait of Hormuz situation all month and as recently as this morning WHILE YOU SLEPT–that's a geopolitical supply shock, not a case of Americans borrowing too cheaply to buy gasoline. Housing costs are a function of a chronic shortage of units built over the better part of two decades; you don't build more homes by making construction loans more expensive, you build fewer. 💡 Healthcare pricing is administered through insurers, hospital systems, and Medicare reimbursement formulas that don't so much as glance at the Fed Funds rate.
Rate hikes work on one thing: the cost of borrowed money. That's a genuinely useful lever when inflation is a demand problem, too many dollars chasing too few goods. It's a much duller instrument when inflation is a supply problem, wrapped in policy formulas, wrapped in geography. Raising rates to fight that kind of inflation doesn't fix the plumbing. It just turns down the pressure everywhere else in the house while the leak keeps leaking. I'll leave you with the question rather than the verdict, because I think it's the right one to sit with heading into September: is the Fed's tool actually aimed at the problem it says it's solving, or is it mostly available for the problem it knows how to solve?
For your portfolio, the practical takeaway is less philosophical. A September hike, or even the market's current coin-flip pricing of one, argues for continued pressure on rate-sensitive assets –growth stocks trading on distant earnings, long-duration bonds, anything priced off the assumption that cheap money was coming back soon. It's a reason to keep some dry powder rather than chase every rally, and a reason mortgage shoppers and credit card holders shouldn't assume relief is imminent. For everyday households, the more useful lesson might be the one buried in my skepticism above: watch your own grocery bill and your own rent renewal separately from what the Fed decides in September, because those two clocks aren't running on the same schedule, and they never really were.
Which brings us back to that job posting I opened with. Give Warsh this much–Friday was tighter than his first few outings, and judging by the move in yields, gold, and bitcoin, tighter was all the job required. He didn't need a new number or a new promise. Just thirty disciplined minutes of describing the problem without naming the fix, and letting everyone else do the work of deciding what it meant. Turns out that's the whole skill set. He's watching the same 199 numbers we now are. He still hasn't told us what number would satisfy him. Until he does, we're all just watching each other watch him–pricing our futures off the reflection, and hoping the next occupant of that office learns the lesson a little faster than this one did.
FRIDAY’S MARKETS
On Friday, the S&P 500 slipped 0.25%, the Nasdaq Composite fell 0.52%, and the Dow Jones Industrial Average edged down 9 points, or 0.02%, to 53,560. The 10-year Treasury yield rose more than 5 basis points to 4.73%, while the 2-year, more sensitive to Fed policy expectations, jumped over 12 basis points to 4.36%. Gold futures fell roughly 3% to settle near $4,504 an ounce as the dollar strengthened. Bitcoin dropped 3.0% to $77,838, with roughly $488 million in leveraged crypto positions liquidated.
NEXT UP
Dallas Fed Manufacturing Activity (August) may have improved to 1.6 from 1.3.
The week ahead: still some important earnings you don’t want to miss in addition to PMIs, JOLTS Job Openings, Factory Orders, ADP Employment Change, Challenger Job Cuts, Nonfarm Payrolls, and the Unemployment Rate. Make sure you check back often so you can stay ahead of the wave.