Tariffs can’t explain rising goods inflation

The price index for core personal consumption expenditures (core PCE) continues to grow at a rate well above the Federal Reserve’s 2 percent inflation target. This note assesses the claim that tariffs account for most of core PCE inflation’s overshoot of that target. We argue that the pattern of inflation within disaggregated PCE goods categories is inconsistent with the impact of tariffs predicted by a widely used accounting framework. If tariffs have not (yet) shown up in core PCE inflation, then other factors may be keeping inflation above target.

Core goods inflation has been rising

Through January 2026, core PCE inflation was at 3.1 percent year over year. Figure 1 shows core PCE inflation broken down into three categories: core goods, housing, and core services excluding housing. Housing inflation has returned to its pre-pandemic rate, and core services excluding housing has largely remained stable since late 2024.¹ Core goods inflation, however, has been rising steadily; as of January, core goods inflation was 1.9 percent year over year, well above its pre-pandemic average of -0.6 percent. Based on a 2015–19 baseline, excess core goods inflation is adding 0.6 percentage points to core PCE.

The predicted contribution of tariffs to goods inflation

What accounts for elevated core goods inflation? The dominant narrative has been that core goods inflation has been mostly or entirely driven by tariffs. Estimates suggest that tariffs are adding between 0.5 to 1 percentage points to inflation. Put another way, core PCE inflation would be 2.1–2.6 percent without tariffs—much closer to target.

Assessing the contribution of tariffs to consumer price inflation requires assumptions about how tariffs are passed through to wholesale and consumer prices. When large tariffs were first introduced in the spring of 2025, Federal Reserve economists proposed accounting frameworks to assess how country- and industry-specific tariffs would ultimately impact consumer prices.²

The general approach was to use input-output tables to convert country and industry tariffs into cost shocks that have direct and indirect effects on industry prices. The change in industry prices could be converted to changes in PCE prices using PCE bridge tables. The effect on core PCE inflation, which depends on the assumed size of the tariffs and assumptions about markups and pass-through, was forecast to be around 1 percentage point.

Since more time has elapsed, we can use this approach with realized effective tariffs instead of announced tariffs. Realized tariffs have generally been lower than announced tariffs due to substitution, various exemptions and postponements, and imperfect enforcement. For instance, a large tariff was announced on pharmaceutical imports but has yet to come into effect, with realized effective tariffs on pharmaceutical goods remaining close to zero.³

Following earlier studies, we use realized tariffs at the industry level through December 2025 as the cost shock.⁴ Using the input-output tables and the PCE bridge table, this cost shock at the industry level is translated into price changes at the disaggregate PCE goods level. Using the input-output tables accounts for the direct effect of tariffs and the indirect effect as price changes to intermediate goods (like steel) pass through to final goods (like cars).

Realized inflation does not match predicted inflation in core goods

This accounting framework finds that realized tariffs are adding 2 percentage points to core goods inflation, equivalent to an additional 0.5 percentage points to core PCE inflation—in line with but at the lower end of other estimates of the tariffs’ effects. However, this accounting framework also produces predictions for inflation of the many underlying consumption categories in core goods. Crucially, we find that actual inflation within these categories does not line up with the predicted inflation of the tariff accounting framework.

Figure 2 shows the predicted and actual contributions to inflation from PCE core goods categories. Actual contributions are the change in prices by category (relative to 2024) multiplied by the category’s weight in core personal consumption. If tariffs behaved exactly as the accounting framework would predict, then predicted and actual contributions should lie close to the 45-degree line.

As the simple weighted regression line shows, predicted and actual contributions appear to be negatively correlated. Put another way, many categories of core goods with the biggest contributions to inflation in 2025 have faced low tariffs, and some categories (like new motor vehicles) with high predicted tariff exposure have contributed little to inflation.

Does the negative correlation mean tariffs have had no effect on inflation? No—in some categories, predicted and actual inflation contributions do line up. For categories like furniture and home furnishings, there have been clear increases in prices with a magnitude commensurate with their tariff exposure. Those categories close to the 45-degree line whose actual inflation lines up with predicted inflation account for at most 0.2 percentage points of core PCE inflation. However, it seems challenging for tariffs to explain much more of core goods inflation when major categories like pharmaceuticals, clothing, and motor vehicles do not show excess inflation consistent with their tariff exposure.

Where is the tariff inflation?

How does this analysis help us understand the relationship between tariffs and inflation? One possibility is that price changes in core goods may reflect the anticipation or announcement of tariffs rather than realized cost increases. This may explain high inflation in PCE goods categories that have not faced significant tariffs (like pharmaceuticals or video equipment). More generally, the anticipation of tariffs may have allowed manufacturers and retailers to raise prices in early 2025 prior to any imposition of tariffs.⁵ Artificial intelligence–induced demand for electronics may also account for some of the outsized price increases in video and information processing equipment.

Another possibility is that tariff-induced price increases are still in the pipeline. Our framework assumes complete pass-through, but it may take longer for price increases from indirect channels to show up in consumer prices. Firms may wait to pass through price increases as pre-tariff inventories are drawn down or wait for contracts to reset. Firms may perceive a lower ability to pass through price increases and instead absorb tariff costs via lower profits. In a 2026 article, Peter Orszag and Adam Posen argue that price increases from tariffs are still making their way through to consumers.

Our findings do not contradict studies that suggest nearly full pass-through of tariffs to prices paid by importers. Recent work from Gita Gopinath and Brent Neiman finds evidence of pass-through of tariffs to importers but only limited evidence of pass-through of tariffs to consumer prices. The February consumer price index (CPI) and producer price index (PPI) readings provide some suggestive evidence that tariff effects are still in the pipeline. For example, inflation for apparel—one category heavily exposed to tariffs with relatively modest price increases through January—jumped 1.3 percent month over month in February. February PPI also rose 0.7 percent and has progressively increased in each of the last five months. Recently released March ISM prices paid for both manufacturing and services moved sharply higher relative to the prior month.

In short, accounting-based estimates of the contribution of tariffs to core PCE inflation may be overstating the impact of tariffs on core goods inflation. The underlying pattern of inflation within core goods is inconsistent with the predicted pattern of price increases from tariffs.

Endnotes

1 See the Federal Reserve Bank of Minneapolis research note “Despite cooling prices for new leases, overall housing inflation could remain elevated into 2025” for a discussion of the lags in shelter inflation.

2 See Federal Reserve Bank of Boston brief by Barbiero and Stein and Board of Governors note by Minton and Somale. See also economics letters from the Federal Reserve Bank of San Francisco and the Federal Reserve Bank of St. Louis.

3 A 100 percent tariff on imported pharmaceutical products was announced on September 25, 2025.

4 Our methodology note and replication file are available at github.com/tradewartracker/tariffs-inflation.

5 Core goods prices on a monthly basis rose 0.4 percent in January and February of 2025.