Warsh's Fed holds rates but 3 dissents and a muddled press conference rattle markets
A 9-3 vote to hold rates steady was contentious enough — then Kevin Warsh's second press conference left investors unsure whether a rate hike is coming and which inflation number the Fed even trusts.
Source: fortune.com
The Federal Reserve held interest rates steady after its July 2026 meeting, but almost nothing else about the decision was steady. Three policymakers dissented in favor of a rate hike — the most in nearly a decade — and Fed Chair Kevin Warsh's press conference left bond markets, currency traders, and professional analysts struggling to agree on what the central bank had just told them.
The confusion started with the vote itself: 9-3 is not a comfortable majority for a central bank that prizes consensus. It deepened when Warsh fielded questions about inflation — and the answer to what sounds like the simplest possible question, 'How high is inflation right now?', turned out to depend entirely on which of several official government gauges you consult.
Here is why that question is harder than it looks, and why the answer matters far beyond the Fed's meeting room.
The Fed's preferred gauge — the Personal Consumption Expenditures index, or PCE — showed headline inflation at 3.7% year-over-year in June 2026, down from 4.1% in May. Strip out food and energy and that same PCE report yields 3.3%. Use the Dallas Fed's 'trimmed mean' version of PCE, which Warsh has publicly favored, and the number drops to 2.2% — barely above the Fed's 2% target. The Atlanta Fed's 'sticky' inflation reading, which tracks prices that change slowly, clocked in at 2.8%; its 'flexible' reading, for prices that move quickly, hit 5.1%. All of these numbers describe the same U.S. economy in the same month.
None of them is wrong. They are answers to subtly different questions — and understanding why illuminates both the Fed's dilemma and why Warsh's press conference sent such mixed signals.
Every inflation index starts with a deceptively simple task: figure out what a typical household buys and in what proportions. From there, the methodological choices multiply fast. If a laptop costs the same as last year's but runs twice as fast, official statistics count that as a price cut — a 'quality adjustment' that makes intuitive sense to economists but looks like a flat price to the person buying it. Housing is measured not by what homeowners actually pay each month but by a hypothetical estimate of what they would pay to rent their own home — a calculation that affects roughly two-thirds of American households who own rather than rent. And when steak gets expensive and shoppers switch to chicken, should the index track the steak price or the consumer's workaround?
The widely used Consumer Price Index and the Fed's preferred PCE index handle these choices differently. PCE takes a broader approach that incorporates producer prices alongside consumer spending, and it updates its spending weights more frequently to reflect how people actually shift their buying habits. That is why PCE tends to run slightly lower than CPI — and why the two can sometimes point in different directions.
The 'trimmed mean' takes yet another approach: it discards the biggest price swings in both directions — the volatile outliers driven by, say, a spike in oil prices after a geopolitical shock — and averages what remains. The logic is that extreme moves are often noisy and temporary, so they should count for less when setting long-term interest rate policy. Warsh has argued for this approach, which is why his preference for the Dallas Fed's 2.2% reading matters: it suggests inflation is much closer to target than the headline PCE figure implies.
At his press conference, Warsh confirmed the Fed is sticking with PCE as its formal benchmark for now — but added a caveat that unsettled markets. Asked whether the 2% target itself might change, he said that observers would have to wait until after next January to know what the committee might say about strategy. Investors heard a possible softening of the target. His prepared remarks told a different story.
There is no soft inflation target, there is no soft implicit target — not on this Committee's watch. There is only a target, and it is 2 percent.— Kevin Warsh, Federal Reserve Chair
That gap — between the careful language of his prepared statement and the murkier answers he gave reporters — is what CNBC described as a 'muddled performance' that 'scrambled many investors' view of the Federal Reserve.' Long-term Treasury yields rose, the dollar fell, and gold climbed. A chorus of market analysts said the performance called his credibility into question.
We understand that the five-plus years of inflation above target cannot be cured in nine weeks — or by a single month of modest price decreases.— Kevin Warsh, Federal Reserve Chair
That line was Warsh's response to a rare piece of good news: the Consumer Price Index had shown a 0.4% month-over-month price decline in June — an outright drop. Rather than celebrate it, Warsh waved it off, a signal that CNBC reads as more hawkish than markets initially registered. The outlet notes that when Jerome Powell said 'we will not hesitate' to raise rates in May 2022, he followed through with a three-quarter-point hike the following month. Warsh used nearly identical language Wednesday: 'Where necessary and appropriate, we will not hesitate to act.'
Beyond the rate question, Warsh signaled the FOMC has begun discussing its balance sheet — the roughly $7 trillion in bonds the Fed accumulated through years of asset purchases. Shrinking that balance sheet would tighten financial conditions without a formal rate hike, giving Warsh a tool that doesn't require a committee majority to vote for higher rates. 'We discussed monetary policy tools and strategies for achieving stable prices,' he said, adding that the committee examined 'how much accommodation are we getting from the balance sheet?'
Warsh has also formed internal task forces to review the Fed's inflation frameworks, data sources, and communications strategy. Those groups are not expected to report back before the end of 2026, which means the current ambiguity — over which inflation number to trust, over whether a hike is coming, over what the Fed's strategy actually is — may persist for months. U.S. borrowing costs hit a 19-year high this week even without a rate increase, a sign that markets are already pricing in the uncertainty.
For households, the stakes of this methodological debate are concrete. The inflation gauge the Fed officially adopts shapes not just interest rates but also Social Security cost-of-living adjustments, federal income tax brackets, and wage negotiations. A reading of 2.2% tells a very different story about how much relief workers need — and how urgently the Fed must act — than a reading of 3.7%. With tariffs continuing to push up goods prices and energy markets swinging on geopolitical news, the gap between those numbers is unlikely to close on its own.
Why it matters — The inflation gauge the Fed officially adopts shapes not just interest rates but mortgage costs, Social Security adjustments, tax brackets, and wage negotiations — so the choice between a 2.2% and a 3.7% reading has direct consequences for every American household budget.
⚠ Not yet confirmed
- Markets interpreted Warsh's press conference remarks as dovish, suggesting he might soften the 2% inflation target.
- Warsh's task forces reviewing inflation frameworks and Fed communications will not report back before end of 2026.
Sources differ on Whether Warsh's press conference signaled a dovish or hawkish stance on inflation: Markets and many analysts read Warsh as dovish, driving bond yields higher and the dollar lower. (cnbc.com) vs A close reading of his prepared remarks suggests he remains hawkish and may be preparing to raise rates or shrink the balance sheet. (cnbc.com)
Reported by fortune.com, cnbc.com, bloomberg.com
- Jerome PowellAmerican financier (born 1953)