Rate cut in sight? Walsh's preferred inflation gauge cools significantly!
```
Kevin Walsh's favored inflation indicator is cooling down, but this does not mean the Fed's rate-cut signal is clear. The latest data gives dovish narratives new grounds and brings a more critical question to the forefront: Which inflation thermometer should the Fed ultimately trust?
According to Reuters, the Dallas Fed trimmed mean PCE inflation was approximately 2.3% year-over-year in April, down from 2.4% in March. Nomura’s report shows the 12-month trimmed mean PCE for April as 2.35%. This is Walsh’s explicitly preferred indicator, stated during his confirmation hearing, and supports his judgment that “inflation has improved over the past year.”
But another set of data goes in the opposite direction. Core PCE, released by the US Department of Commerce's Bureau of Economic Analysis, rose to 3.3% year-over-year (Nomura states 3.29%), the fastest since 2023. Fed Governor Lisa Cook said this indicator "is clearly moving in the wrong direction.” This means the cooling trimmed mean PCE is now sending conflicting signals compared to the traditional core inflation metric.
This divergence directly affects the rate-path judgement. If the Walsh-led Fed gives more weight to the trimmed mean PCE, arguments for raising rates are weakened and the narrative for rate cuts gets more support. However, Dallas Fed economists and Nomura both caution that the indicator may currently be underestimating true price pressures. Nomura estimates the trimmed mean PCE year-over-year might have a downward bias of about 48 basis points.
A cooler indicator versus a hotter data set
The Dallas Fed trimmed mean PCE is Walsh’s preferred tool for watching inflation. It tries to remove the most extreme price increases and decreases each month, keeping the middle section and filtering out short-term noise.
The latest April reading shows this indicator continues to cool. The Dallas Fed trimmed mean PCE year-over-year is 2.3%, down from 2.4% in March. Nomura’s report puts April’s year-over-year trimmed mean PCE at 2.35%, 94 basis points lower than the core PCE’s 3.29%.
This is exactly what the market is watching. Walsh has previously criticized the core PCE as merely a “rough swag” during his confirmation hearing, stating he prefers alternative inflation indicators like the trimmed mean. If this indicator gets more weight, the Fed's inflation assessment may not entirely follow the core PCE anymore.
However, core PCE is still the basic metric long-valued by many Fed policy makers. The 4th month’s core PCE rose to roughly 3.3% year-over-year, fastest since 2023. According to Reuters, this result supports rising concerns among more policy makers about price pressures, and even strengthens some beliefs that additional rate hikes may be needed.
Why Walsh prefers the trimmed mean PCE
The logic behind trimmed mean PCE is simple: monthly inflation data is noisy; abnormal volatility in a few items can distort the aggregate reading. Unlike core PCE, which excludes food and energy by default, the trimmed mean PCE does not permanently exclude certain categories. Instead, each month, it removes the items with the highest and lowest price changes.
In April, extreme volatility items included surges in gasoline, airfare, and jewelry prices, and drops in poultry, household fabrics, and haircut prices. The trimmed mean PCE removes these extremes and leaves the more representative middle section.
The Dallas Fed currently removes the bottom 24% and top 31% of price changes. This approach reflects historical patterns: usually, more items experience price drops or slow increases than sharp rises, so more items are trimmed from the high inflation end to adjust for the skew in distribution.
In normal times, this method helps filter out noise and observe inflation trends. The problem is, these may not be normal times.
Dallas Fed cautions: this indicator may now be biased low
Dallas Fed economist Tyler Atkinson told Reuters not to be overly optimistic based on the trimmed mean PCE level. His reasoning is that the skew direction of current price change distributions may have changed.
Over the past year, tariffs imposed by the Trump administration pushed up many goods' prices. This meant price increases were no longer limited to a few outliers but covered a broader range of goods. As a result, the old trimmed rules used to adjust skewness may now be removing too many high inflation items.
In other words, using the existing method to remove the top 31% and bottom 24% could suppress the inflation indicator, underestimating actual price pressures. This isn’t the first time. During the post-pandemic inflation uptrend, the Dallas Fed trimmed mean PCE also sent overly cool signals, suggesting inflation would be milder than it really was.
The Dallas Fed currently has no plans to change the method. Tyler Atkinson says if tariff-induced price pressures fade as expected, the issue may resolve itself in coming months. Until then, other indicators of underlying inflation may be more worth watching.
Trimmed mean PCE may underestimate inflation by 48 basis points
Nomura gives a more systematic explanation of this issue in its May 29 report. The core finding is that the trimmed mean PCE tends to be slow in responding to changes in inflation trends, and may now underestimate inflation due to changes in price distribution skewness.
The trimmed mean PCE's advantage is fewer false positives, rarely mistaking short-term noise for new trends. But the same trait also makes it slow to respond when inflation dynamics change. In the early stages of the post-pandemic inflation surge, the trimmed mean PCE lagged behind multiple "true inflation" trend metrics. By comparison, core PCE has tracked these trend metrics better in recent years.
More importantly, since the Dallas Fed updated its trimming methods in 2009, the distribution of US PCE component price changes has become more frequently right-skewed. If the distribution becomes more right-skewed, the trimmed mean PCE over-ignores high-end items, thereby underestimating underlying inflation.
Currently, official trimmed mean PCE year-over-year is 2.35%, but adjusted for bias, it is about 2.8%. This means the official indicator may underestimate underlying inflation by around 48 basis points. Though the adjusted reading is still lower than the core PCE’s roughly 3.3%, the gap is clearly narrowed.
Changes in goods prices are undermining reliability of the indicator
Nomura argues that a key reason for the current bias in the trimmed mean PCE is changing goods price dynamics.
Before the pandemic, core goods prices were long a drag on inflation. In the 2010s, core goods PCE prices were almost always negative, with quality adjustments for electronics, computers, TVs, cars, and appliances driving prices lower. But after the pandemic, core goods prices no longer provide stable deflation. Post-pandemic, core goods prices kept pushing inflation up, and the price change distribution is now more prone to right skew.
The AI investment boom and demand for key electronic components like semiconductors could continue supporting core goods prices. Meanwhile, businesses’ pricing behavior is changing. Richmond Fed surveys show a higher share of manufacturers now adjust prices at least annually compared to before the pandemic, and infrequent price changers make up a smaller share.
This means goods prices are now more sensitive to inventory and supply-demand shifts, and are more cyclical. If rising goods prices become more common, the problem of over-trimming high-end price changes in the trimmed mean PCE will become more pronounced.
The rate-cut narrative gets support, but the evidence is not clean
The cooling in Walsh’s favorite indicator undoubtedly weakens the argument that “inflation is out of control and tighter policy is needed.” It also provides data support for Walsh’s statements about improving inflation.
But this is not a one-way signal. Core PCE rose to its fastest since 2023, and Lisa Cook’s comments show some policy makers still see current inflation as a risk. Nomura warns, If monetary policy relies on biased inflation indicators, the Fed could risk falling behind the inflation curve, leading to more aggressive rate hikes later.
Standard Chartered Bank analysts Steve Englander and Dan Pan are also skeptical about the cooling inflation signals from the trimmed mean PCE. They argue it’s hard to prove the disinflation shown by this indicator is genuine, and historically it has been less effective in forecasting future inflation compared to the core PCE.
Harvard economist Jason Furman commented on X that the trimmed mean PCE isn’t the only inflation indicator showing softer numbers lately—Cleveland Fed’s PCE median has also declined, but its 2.8% level is higher than the Dallas Fed trimmed mean. He thinks Walsh’s favored indicator isn’t unreasonable in itself, the real concern is whether it’s being cherry-picked after the fact.
Thus, the rate-cut narrative does gain a more favorable data point, but it’s not enough to form a definitive rate-cut signal. For markets, the next key is not just whether the trimmed mean PCE keeps cooling, but whether Walsh’s Fed formally changes the inflation signal weights, and if other policy makers accept this framework.
Risk Warning and DisclaimerThe market has risks and investments require caution. This article does not constitute personal investment advice and does not take into account individual users' specific investment objectives, financial situations, or needs. Users should consider whether any opinions, views, or conclusions in this article are suited to their own circumstances. Investment decisions based on this are at your own risk. ```