America’s inflation problem is becoming harder because the numbers are telling different stories at the same time.

The official May data were uncomfortable. The headline Personal Consumption Expenditures price index rose 4.1% from a year earlier. Core PCE inflation, which excludes food and energy, rose 3.4%.

Those figures are far above the Federal Reserve’s 2% inflation target. They also arrive after several years in which Americans have already absorbed a substantial increase in the overall price level.

Under a conventional reading of the data, the case for cutting interest rates appears weak. Inflation is not near target. Core inflation is moving in the wrong direction. And households are still dealing with higher costs for food, housing, energy, insurance, transportation, and services.

But Kevin Warsh sees another number.

The Dallas Federal Reserve’s trimmed-mean PCE measure showed inflation at 2.4% over the 12 months ending in May. That is not 2%. But it is much closer to the target than either headline PCE or core PCE.

The difference is large enough to change the entire monetary-policy narrative.

One measure says inflation is still running hot. Another says underlying inflation may be approaching normal.

If inflation is 4.1%, the Fed looks trapped. If inflation is 2.4%, the Fed can begin to imagine a path toward easing.

The real issue is not one data point. It is the definition of inflation.

Central banks do not merely react to inflation. They choose how to measure it.

That choice matters because no inflation index is neutral. Every measure includes some prices, excludes others, assigns different weights, uses different assumptions, and interprets consumer behavior differently.

The Consumer Price Index, or CPI, is the measure most visible to households. It tracks the prices consumers pay for a basket of goods and services. It is widely followed because it reflects the lived experience of rising prices.

The Personal Consumption Expenditures price index, or PCE, is the Federal Reserve’s preferred broad inflation measure. It is constructed differently from CPI and tends to capture shifts in consumer spending patterns more flexibly.

Core PCE then removes food and energy. The argument is that these categories can swing sharply from month to month because of weather, oil prices, geopolitical shocks, crop failures, and other volatile factors.

The goal is to see the underlying trend beneath temporary noise.

Trimmed-mean PCE goes further.

Instead of permanently excluding food and energy, it ranks price changes across many categories each month. It then removes the largest increases and the largest declines before calculating the average of what remains.

In simple terms, core PCE says: remove the same volatile categories every month.

Trimmed-mean PCE says: remove whatever is behaving most unusually this month.

That distinction sounds technical. It is not.

It can determine whether inflation looks far above target or nearly under control.

Warsh’s preferred measure is giving him a much easier policy story

Kevin Warsh has made clear that he prefers trimmed-average measures of inflation. His argument is that policymakers should not overreact to one-off price shocks, whether they come from tariffs, energy disruptions, transportation bottlenecks, or isolated category spikes.

That argument has intellectual merit.

A central bank cannot solve every temporary increase in gasoline, airline fares, imported goods, or food prices by raising interest rates. Monetary policy works with long lags. It cannot create more oil, reopen a shipping lane, or reverse a drought.

If policymakers respond mechanically to every visible price spike, they could tighten too much and damage employment, investment, and credit conditions.

This is the strongest case for trimmed means. They may offer a clearer view of persistent, generalized inflation than a headline index distorted by a few extreme categories.

But the timing is politically explosive.

Warsh is taking office at a moment when headline inflation has risen to 4.1% and core PCE has reached 3.4%. A shift toward a lower inflation gauge would naturally be interpreted as a way to justify easier policy despite numbers that remain uncomfortable for consumers.

That does not make the trimmed mean invalid. It does mean the Fed must be extraordinarily careful about how it uses it.

A better inflation measure can improve policy. A newly favored inflation measure can also look like a lower standard when the old one becomes inconvenient.

The Arthur Burns warning still hangs over the Federal Reserve

The historical fear behind this debate is Arthur Burns.

Burns led the Federal Reserve during the 1970s, a decade in which America suffered from rising inflation, weak growth, oil shocks, political pressure, and repeated policy mistakes.

The broad lesson from that era is not that one inflation measure caused the Great Inflation. It did not.

The problem was deeper. Policymakers repeatedly treated inflation as temporary, external, or caused by factors that monetary policy could not address. They were too willing to believe that price controls, wage controls, supply shocks, or one-off events explained the problem.

They were too slow to accept that inflation expectations were becoming embedded.

Richard Nixon imposed wage and price controls in 1971. The policy initially appeared politically useful. But it did not solve the underlying imbalance between demand, money, supply constraints, and inflation expectations.

Then came the oil shock. Energy prices surged. Food prices rose. Inflation broadened. And the United States entered the era of stagflation.

The danger was not simply that officials excluded volatile prices. The danger was that they kept finding reasons not to respond to the inflation that households were actually experiencing.

That is why the Burns analogy is relevant today.

If the Fed uses trimmed-mean PCE as one of several diagnostic tools, that is normal central-bank practice. If it uses trimmed mean as a reason to dismiss sustained inflation in essential household expenses, the comparison becomes more serious.

PCE was not created as a trick, but measurement choices still matter

Critics often argue that the Federal Reserve’s shift toward PCE was a way to make inflation appear lower than CPI. That criticism contains a real political intuition, but the full story is more complicated.

PCE has legitimate methodological advantages. It captures a broader set of expenditures. It reflects changes in what consumers buy when relative prices change. It includes spending made on behalf of households, such as employer-provided health care. And it adjusts weights more frequently than CPI.

These features can make PCE useful for understanding the economy’s overall consumption pattern.

But it is also true that PCE has often run lower than CPI. That means a 2% PCE target is not identical to a 2% CPI target in the public imagination.

For consumers, this distinction can be frustrating. Households do not experience inflation as an econometric concept. They experience it when groceries cost more, rent rises, insurance premiums jump, and utility bills become harder to pay.

A central bank can be technically correct and politically disconnected at the same time.

That is why moving one step further—from core PCE toward trimmed-mean PCE—requires careful communication.

The Fed cannot simply say: inflation is lower because we used a measure that removed the prices hurting you most.

It must explain why those prices are temporary, why they will not spread, and why monetary policy should not react to them.

Why trimmed mean can be useful

The case for trimmed-mean inflation is not cynical by definition.

Inflation data are noisy. Individual categories can move violently for reasons that have little to do with broad inflation momentum. A single month can be distorted by a jump in gasoline, airline fares, jewelry, medical services, hotel prices, or imported consumer goods.

If the Fed reacts to every extreme movement, policy may become unstable.

Trimmed mean seeks to identify the middle of the price-change distribution. It asks whether inflation is widespread across the economy or concentrated in a small number of unusual categories.

That can be particularly useful when policymakers are trying to distinguish a supply shock from a demand-driven inflation problem.

For example, a sudden oil shock can raise headline inflation quickly. But if most other prices are stable, the central bank may not need to react as though the entire economy has overheated.

In that sense, trimmed mean can prevent policy mistakes.

It can help policymakers avoid treating every temporary price move as a permanent change in the inflation trend.

The problem is that no measure can identify the future with certainty.

What looks like an outlier today can become the beginning of a broader inflation cycle tomorrow.

Why trimmed mean can also fail at the worst possible moment

The weakness of trimmed mean is also obvious.

Inflation often begins in a narrow set of categories before spreading through the rest of the economy. At the beginning of an inflation surge, the categories moving first may look like outliers. A trimmed-mean measure may remove them.

That can make the inflation picture look calmer than it really is.

The danger is greatest at turning points.

If energy rises sharply, then transportation costs rise. If transportation costs rise, goods prices rise. If goods prices rise, businesses raise prices. If workers demand compensation, wage pressure increases. If expectations change, inflation can become broader and more persistent.

By the time the change appears in the middle of the distribution, the central bank may already be late.

This is the main criticism of relying too heavily on trimmed mean.

It can be excellent at filtering noise. It can also be slow to recognize that the noise is becoming the new signal.

The central-bank mistake is not always reacting too much to volatility. Sometimes it is trimming away the first evidence that inflation is spreading.

May’s numbers show the exact problem

The May inflation data illustrate why this debate matters.

Headline PCE inflation was 4.1%. Core PCE inflation was 3.4%. Trimmed-mean PCE inflation was 2.4%.

These are not small differences. They imply three very different interpretations of the U.S. economy.

At 4.1%, inflation looks dangerous and politically visible. It suggests that the cost of living is rising too fast and that rate cuts could reignite pressure.

At 3.4%, inflation still looks clearly above target. It suggests that even after excluding food and energy, broad price pressure remains too high.

At 2.4%, inflation looks like a problem that is close to being solved. It suggests that the central bank should avoid overreacting to temporary shocks.

The policy consequences are enormous.

One number supports holding rates high or even considering further tightening. Another supports patience. A third creates room to begin discussing rate cuts.

This is why the choice of metric is not a technical footnote. It is a policy decision in disguise.

Warsh may be a hawk in language and a dove in framework

Kevin Warsh is often described as a hawkish figure because of his criticism of past Federal Reserve policy, his concern about institutional credibility, and his willingness to challenge the post-pandemic policy regime.

But a central banker’s label depends on more than speeches. It depends on the framework used to interpret the economy.

A policymaker can sound hawkish while adopting a measurement system that makes inflation appear closer to target. That can create a more dovish policy path without ever requiring the chair to announce a softer inflation objective.

The official target can remain 2%. The public language can remain focused on price stability. The Federal Reserve can insist that it has not changed its mandate.

Yet the practical reaction function can shift.

If the Fed places greater weight on trimmed-mean PCE, it may conclude that current inflation is less threatening than headline or core measures imply. That could make it easier to tolerate inflation above 2% for longer. It could make it easier to cut rates when the labor market weakens. It could make it easier to argue that tariff or energy-driven inflation should be “looked through.”

That is not necessarily an abuse of power. It may be a coherent view of how monetary policy should work.

But it would be a major shift in emphasis.

Investors should not dismiss it as a technical adjustment.

The task forces matter more than they first appear

Warsh has repeatedly emphasized internal Federal Reserve task forces. That deserves attention.

Large institutions rarely change their operating framework through one dramatic announcement. They create committees. They review data. They redefine communications. They publish research. They adjust staff processes. Then, gradually, the new framework becomes normal.

A task force focused on inflation measurement does not automatically mean the Fed will abandon core PCE. But it can create the intellectual foundation for changing how policymakers describe inflation risks.

The first shift may be rhetorical.

Officials may begin mentioning trimmed mean more frequently in speeches. They may refer to it as a better measure of trend inflation. They may place more emphasis on dispersion, outliers, and temporary price shocks. They may argue that headline inflation overstates the policy problem.

The second shift may be institutional.

Staff forecasts could give more prominence to alternative measures. FOMC discussions could use trimmed mean as a key benchmark. Forecast documents could place more weight on it.

The third shift would be political.

If the Fed begins cutting rates while headline inflation remains high, trimmed mean could become the intellectual explanation for why that decision is justified.

That is why markets should watch the framework review closely.

The risk is not only inflation. It is credibility.

The Federal Reserve’s most important asset is not its balance sheet. It is credibility.

Businesses, workers, investors, and households make decisions based partly on what they think the Fed will tolerate. If they believe the Fed will protect price stability, inflation expectations remain more contained. If they believe the Fed will redefine inflation whenever the data become inconvenient, expectations can become less anchored.

This is especially sensitive after several years of elevated inflation.

Americans have already experienced a large cumulative increase in prices. Many households do not care whether the Fed calls a price shock temporary. They care whether the price level ever becomes more manageable.

Of course, monetary policy cannot reverse every past price increase. The Fed’s target is inflation, not deflation.

But that makes communication even more important.

If the public hears that inflation is “under control” while its household budget says otherwise, the gap between official policy language and lived experience will widen.

A lower technical measure cannot replace public trust.

The Federal Reserve can choose a better statistic. It cannot choose a different lived experience for households.

What markets should watch next

The first signal is language. Investors should watch whether Warsh and other officials increasingly describe trimmed-mean PCE as the best guide to “underlying” inflation.

The second signal is the policy statement. A shift in the FOMC’s language away from broad inflation pressures and toward temporary or relative-price shocks would matter.

The third signal is the Summary of Economic Projections. If the Fed gives more prominence to alternative inflation measures, it would indicate that the framework is changing internally.

The fourth signal is wage growth and services inflation. If wage-sensitive services remain elevated, it will be harder for the Fed to argue that inflation is merely a collection of outliers.

The fifth signal is inflation expectations. If market-based expectations, consumer surveys, or business pricing plans rise, then the Fed will have less room to “look through” high headline readings.

The sixth signal is whether the trimmed mean remains low. In May it stood at 2.4%. That is close enough to target to be politically important. But it could move higher quickly if price increases broaden across more categories.

Conclusion: the Fed may not be changing the target, but it may be changing the path

Kevin Warsh may not formally abandon the 2% inflation target. He may not announce that core PCE has been replaced. He may not openly say that the Federal Reserve is preparing to tolerate higher inflation.

But a shift toward trimmed-mean inflation measures could still change policy in practice.

It could make the current inflation problem appear less severe. It could make supply shocks look more temporary. It could make a future rate cut easier to defend. It could move the Fed’s reaction function in a more dovish direction without changing the official target.

That is why the debate is so important.

The question is not whether trimmed-mean PCE is a legitimate economic tool. It is.

The question is whether the Fed uses it as one useful lens among many, or as the preferred lens when every other measure is sending an unwelcome message.

The simplest way to understand the Warsh debate is this: he may not be moving the 2% target, but he may be looking for a different ruler that makes the distance to that target appear shorter.