Why tariffs are so confusing

Government data is dead! Long live government data.

Dear Colleagues: Two posts this week. Since the government shutdown, federal agencies have suspended the collection and distribution of government statistics and economic indicators. Private company datasets are imperfect, but they can be informative. This post focuses on tariffs, for which federal data has only been updated through June. Read on to see my look at tariffs using the latest Ramp spend data.

---

Economists were pretty worried when tariffs were first announced, and we had good reason to be. First, the tariffs represented the largest tariff increase in over 100 years. Second, they were applied to nearly all countries, broadening the impact. Third, the deviation from existing policy was so large no one was sure how or when this was all going to be implemented.

This post will focus on the third point, the laginess in tariff enforcement, where Ramp data can shed some light. Consider the frictions in pushing new tariffs. How long does it take for Customs and Border Patrol to create guidelines on how to collect the new tariffs? How long for officers on the ground to understand those guidelines and start enforcing them consistently? How long for suppliers to add tariff charges to invoices?

These frictions have all led to some lag in the actual implementation and payment of new tariffs. Ramp data from manufacturing and retail sector invoices shows a slow and gradual increase in tariff costs. The share of bills and invoices showing a tariff charge has doubled, from 1.4% in 2024 to nearly 3% as of September 2025. Tariff charges have increased virtually every month since they were first imposed in February, with no slowdown in sight (though manufacturing saw a slight dip last month).

Similarly, the Yale Budget Lab analyzed tariff revenues and found they were only 68% of where we would expect them to be based on official announcements. The research group estimates that the effective tariff rate has increased fourfold since the beginning of the year, higher than the twofold increase we see in Ramp data. However, that’s not far off if you consider many tariff-exposed purchases are larger than the rest of a business’s routine purchases.

All to say: it’s hard to evaluate the full impact of tariffs when it’s taken so long for them to move through the economy. At the same time, it’s clear that this landmark policy change neither sparked a manufacturing boom nor sent the industry off a cliff. It’s very possible the final impact of tariffs will be a modest increase in prices for trade-exposed goods, a modest increase in tax revenue, and some supply chains shifting from high-tariff nations to low-tariff nations — though in many cases that won’t mean reshoring to the United States.

Something from my camera roll

A collection of old public health posters on display at my doctor’s office. Stop finger licking and protect yourself NOW!