Initially delayed, the pass-through of tariffs to consumer prices has arrived

In an April article, we argued that the impact of tariffs could not fully explain the high levels of goods inflation. At the time, goods inflation was pushing core inflation above 3 percent. We argued that high goods inflation mostly reflected elevated price increases in electronics and information technology related to the AI investment boom, while potential tariff effects might still be in the pipeline.

Here we provide an update using personal consumption expenditures (PCE) price index data through July. We find that recent data seem consistent with both effects: material AI impacts on goods inflation and delayed pass-through of tariffs. Tariff impacts have become more visible in recent months. But they continue to account for only part of the overshoot in core inflation, adding around 0.4 percentage points to core PCE inflation.

Inflation data through July: Goods and core inflation higher

Core PCE inflation through July is up 3.3 percent year over year—the highest level since 2023, and outside of the pandemic, the highest level since the early 1990s. The goods component of core PCE—core goods—is up 2.3 percent year over year. Compared with its 2015–19 pre-pandemic average, core goods are currently adding more than 0.7 percentage points to core PCE inflation.¹

Tariff impacts have become more visible

In our April article, we provided an accounting framework for predicting tariff impacts on core goods categories using effective tariff rates.² Figure 2 shows scatterplots of observed excess annual inflation (relative to 2015–19 baseline) in core goods subcategories against the predicted tariff effects. The predicted tariff effect assumes full pass-through of tariffs into prices over the last year. Importantly, to keep our focus on tariff effects, we exclude video and information processing equipment, a category that has largely been free from tariffs but shows substantial excess inflation. We discuss this category separately below.

Through December 2025, the fitted line is flat, indicating little relationship between inflation in core goods categories and tariffs (Figure 2A). By July 2026, however, that relationship has turned markedly positive (Figure 2B). Both the weighted and unweighted least squares lines show positive coefficients, though less than one.

This is driven by a handful of heavily tariffed categories that are showing greater excess inflation, especially in apparel. In the category of clothing and footwear, year-over-year inflation rose from 0.3 percent in December 2025 to 3.5 percent in July 2026. Prior to the pandemic, inflation in the category was close to zero.

Is further tariff inflation likely to materialize in the coming months? Some heavily tariffed categories, such as new motor vehicles, have yet to see sizable increases in inflation. Moreover, recent additional tariffs on auto parts have been announced. A recent survey conducted by the Federal Reserve Bank of New York finds that firms are planning to pass through further tariff-related price increases.

Tariffs versus AI in goods inflation

To calculate the contribution of tariffs to core inflation, we use the regression coefficients from Figure 2B to find the fitted values across goods and take the sum, weighted by their share of core personal consumption. This calculation finds that tariffs account for 0.2 to 0.4 percentage points of core PCE inflation as of July.³ Our analysis matches a recent tariff-effect estimate by the Federal Reserve Bank of St. Louis.

Our calculation means that absent tariffs, core PCE inflation would still be about 1 percentage point above the Fed’s 2 percent target. It also suggests that some of the worsening in core PCE inflation in the first half of 2026 was due to delayed tariff pass-through.

As we noted above, in these calculations we have excluded the category of video and information processing equipment, where AI-driven demand—not tariffs—is the primary story. This category has a weight of 2.4 percent in core PCE and has displayed abnormally high inflation rates in recent months. Through July 2026, prices are up 12.2 percent for this category year over year; from 2015 to 2019, prices were falling in this category by 6.5 percent per year.

This swing is enough to add another 0.4 percentage points to core PCE inflation. As the AI investment boom drives demand for memory and other computer hardware, the spillovers to goods prices appear to be at least as large as tariffs in keeping core inflation high.

Endnotes

1 A FEDS note from April estimated that tariffs were adding 0.8 percentage points (pp) to core PCE inflation as of February 2026. Cavallo et al. (2025) found a similar contribution for overall consumer price index (CPI) inflation as of September 2025. Amiti et al. (2026) find a tariff contribution of 2.2 pp to CPI goods through February 2026, or about 0.8 pp to overall CPI.

2 The accounting framework uses input-output tables to convert country and industry tariffs into cost shocks that have direct and indirect effects on industry prices. For a detailed explanation and the Fed research underlying this method, see our article “Tariffs can’t explain rising goods inflation,” April 2026.

3 More precisely, using the ordinary least squares estimate in Figure 2 we obtain a contribution of 23 basis points. Using the weighted least squares estimate we find a contribution of 38 basis points.