Inflation by Assumption: Fed Tariff Shortcuts

Recently, Federal Reserve Chair Jerome Powell stirred some conversation with his remark regarding new tariffs potentially leading to price increases. He stated that those tariffs would inevitably be passed onto consumers, suggesting a significant jump in prices due to them.

However, this viewpoint seems to reflect Powell’s personal circle rather than the broader data. If his conclusions are drawn from recent Federal Reserve research, there’s a chance that he may have misinterpreted the models involved, which appear to be tighter than they actually are.

A working paper from the Boston Federal Reserve claims nearly all tariffs on Chinese imports from 2018 to 2019 have been transferred to consumers. This paper suggests these tariffs raised core PCE inflation by around 0.3 to 0.4 percentage points annually, implying that similar effects could arise from Trump’s new tariffs.

The catch, though, is that the paper’s findings seem more about self-reinforcing assumptions than actual discovery.

The Model’s Fixed Margins

The authors built a model assuming that company profit margins stay constant relative to marginal costs. So, when tariffs cause input costs to rise, retail prices automatically increase. This model doesn’t actually investigate pricing behaviors; it merely asserts that pass-through will happen.

In essence, this paper doesn’t definitively prove tariffs cause inflation—it’s more about the assumptions it starts with. It reports inflation because that is what its framework requires.

To complicate matters, the authors didn’t compare the pricing behaviors of customs and non-tariff products. They merely looked at categories with significant import shares, attributing any observed price increases directly to tariffs. But it’s important to note that inflation was already occurring during this period, as the Fed was attempting to hit its 2% target. If all prices in the economy were rising concurrently, the authors still point to tariffs as the source of inflation. Their model lacks counterfactuals.

This renders tampering with their findings almost impossible. Rising prices are considered evidence of pass-through, regardless of other influencing factors. If inflation accelerates due to monetary policy, that too counts as pass-through. The methodology effectively dictates the conclusion.

To be fair, while the authors used actual price data at the PCE category level, they neglected to examine company-level pricing or profit margins. They didn’t investigate whether retailers were adjusting margins to keep prices steady, nor did they consider if importers were switching suppliers. Additionally, they omitted how Chinese exporters might adjust their pricing to maintain U.S. market presence. These factors were noted in other studies on 2018 tariffs but are absent here.

A History of Flawed Tariff Research

This type of circular reasoning isn’t new and reflects broader patterns in how the Federal Reserve engages with trade policy.

Back in 2019, a blog post from the New York Fed claimed the tariff burden on U.S. consumers was decreasing. However, the evidence for this claim was solely based on customs data regarding import prices and not retail price or margins. The conclusion hinged on the assumption that changes in import prices would directly affect consumer prices, an untested assumption. Shortly after, even the Fed recognized that valuation disruptions could skew price estimates.

That same year, a report from the New York Fed suggested tariffs led to significant welfare losses for the U.S. economy, based on a short 10-day window of tariff announcements. However, markets rebounded quickly, revealing little strain in macro data. It turned out that if a longer window around the announcements was used, evidence showed inventory levels had actually risen—a narrow focus on a short time frame created an illusion of significant economic costs.

On a more positive note, some recent Fed researchers have avoided these pitfalls. A survey in March 2025 by Atlanta Fed economists used real-world PCE pricing data, approaching the issue without assuming full pass-through. Their findings suggested only a minor, one-off price increase from hypothetical new tariffs, akin to small sales taxes rather than significant inflation. They emphasized that pass-through depends on market structure, competition, and alternatives.

Another paper from 2025 published in AEJ: Insights indicated that much of the 2018 tariffs were absorbed by U.S. retailers rather than passed on to consumers. Consumers often experienced stable prices even as tariffs increased. This reliance on actual retail data stands in contrast to hypothetical modeling.

This context is crucial for understanding the recent Boston Fed entries. They don’t represent a new consensus; rather, they resurrect old assumptions and avoid challenging the data. This doesn’t aid policymakers in grasping the real impacts of tariffs but instead reinforces their pre-existing priorities.

The question of who bears the brunt of tariffs is far too significant to be handled through circular models and assumption-based simulations. If Powell interprets the existing findings as evidence that Trump’s tariffs are inflationary, he might not be engaging with the actual evidence but rather adhering to a predetermined script.