The New 'Preference' of the Federal Reserve's New Chairman: Trimmed Mean Inflation

Similarly measuring inflation, different yardsticks can produce results that differ by nearly one percentage point. And this one percentage point gap may directly determine whether the Federal Reserve has room to lower interest rates.

Warsh's "Preference": Trimmed Mean Inflation

Kevin Warsh, who is expected to take over as the Chairman of the Federal Reserve in mid-May, recently expressed his views on inflation measurement during a hearing at the Senate Banking Committee.

He believes that the current data used to judge inflation is "quite imperfect," and the Federal Reserve should instead use "trimmed mean" to assess "underlying, widespread price changes" in the economy, rather than considering one-time price changes triggered by "geopolitical changes or fluctuations in beef prices."

According to this yardstick, over the past 12 months ending in March this year, the inflation rate is about 2.4% — almost within reach of the Federal Reserve's 2% target, enough to declare the war on inflation as already won by some.

In comparison, the Federal Reserve's current official preference, core PCE (Personal Consumption Expenditures Price Index, excluding food and energy), is running at an annual rate of 3.2% in March, significantly higher than the target level. The two differ by nearly one percentage point.

Two Measures, Different Logics

The logic behind core PCE is fixed exclusion: regardless of how food and energy prices fluctuate, they are excluded. The logic of trimmed mean is different: it calculates the price changes of all consumer categories each month, then "trims" the categories with the most extreme price changes from both ends, retaining the middle part as the inflation reading.

Taking March this year as an example, the Dallas Fed excluded computer software and accessories with an annual increase of 60.2%, and tax return services with an annual decrease of 27% when calculating the trimmed mean — both of which were included in the calculation of core PCE.

The Dallas Fed's research shows that the trimmed mean can better penetrate short-term noise and capture the underlying trend of inflation, making it a superior real-time inflation indicator.

Potential Blind Spot: Serious Distortion Under Tariff Environment

However, this yardstick is not without flaws, and its limitations are particularly worth being vigilant in the current macro environment.

The core mechanism of trimmed mean is to exclude extreme values. But when widespread tariffs simultaneously push up the prices of a large number of imported goods, the price increases of many categories will collectively move towards the top of the distribution — which is exactly where the trimmed mean "takes action." The Dallas Fed itself has warned about this: if tariffs trigger a one-time price shock, the trimmed mean can provide a cleaner reading; but if the price pressures continue to spread, the trimmed mean may significantly underestimate the real level of inflation.

Historical precedents are also worth being vigilant: during the initial spread of inflation in 2021, the readings of trimmed mean continued to be lower than core PCE, missing the early signals of the comprehensive spread of price pressures.

Other institutions' indicators also point to higher inflation levels: the Cleveland Fed's median CPI is close to overall inflation; the New York Fed's underlying inflation model gave a reading of 3.2% in March, consistent with core PCE.

Risk of "Moving Goalposts" Under Political Pressure

Even if Warsh intends to push the Federal Reserve towards trimmed mean, he cannot decide alone. The Federal Reserve's monetary policy is collectively formulated by the committee, with the chairman responsible for chairing meetings and setting the agenda, but voting rights are distributed among the members. Last year, the incumbent chairman Powell tried to promote a reform of the communication framework, but ultimately gave up after being dissuaded by colleagues and staff. Warsh may face similar resistance.

The thornier political risk is that in the context of high expectations for rate cuts and new chairman facing political pressures as soon as they take office, switching to an indicator that just shows inflation close to the target can easily trigger accusations of the "Federal Reserve moving the goalposts," damaging the credibility of monetary policy.

The next release of the Dallas Fed's trimmed mean data will be on May 28th. If the readings are again significantly lower than core PCE, Warsh may continue to use this indicator to back up his policy stance. However, whether the market and other Fed officials can accept this new yardstick remains an open question.