Orply.

US Import Prices Rose Nearly One-for-One With 2018 Tariffs

Stephen ReddingHoover InstitutionFriday, November 20, 20266 min read

At the Hoover Institution, economist Stephen Redding argues that the evidence from the 2018 US tariffs points to American businesses and consumers bearing most of the cost. Prices for tariffed imports rose roughly by the amount of the duties, while imports from targeted countries fell and sourcing shifted to other suppliers. Because Redding finds little evidence that foreign exporters cut their prices, he says the tariffs brought no terms-of-trade gain to offset the domestic costs; his estimate of the resulting deadweight welfare loss is $8.2 billion.

The first incidence of the tariff showed up in US prices

When the United States imposed tariffs on washing machines in 2018, prices for major appliances began rising after years of decline. The Bureau of Labor Statistics series that Stephen Redding uses does not isolate washing machines: it covers kitchen appliances including washers, dryers, refrigerators and cookers. The timing is suggestive, he says, but a single product category cannot establish the broader effect.

To look across products, Redding and his co-authors assembled US customs data covering roughly 16,000 products, identified at the 10-digit level of the Harmonized System. For each product imported from each country, the records include value and quantity. Dividing value by quantity gives a unit value: an estimate of the price received by a foreign exporter before US tariffs. The researchers then applied the corresponding tariff rate to estimate the US import price after the duty.

That distinction matters. The tariff-inclusive import price is not necessarily the price paid by the final consumer. A US importer, wholesaler or retailer may absorb some of the cost. But, Redding says, it is a price paid somewhere in the United States.

The researchers tracked 12-month price changes through 2018, comparing goods and countries affected by six tariff waves with products and countries that were not treated. Using year-over-year changes helps account for seasonal patterns, such as the rise in imports around the holidays. They also measured changes relative to the month before each tariff wave, so that the series begin at zero.

For many tariff waves, prices were relatively flat before the announcement, though Redding notes some evidence of anticipation. After tariffs were introduced, import prices rose sharply—roughly in line with tariffs that were typically around 25 percent. Prices for untreated goods and countries remained comparatively flat. That contrast, he argues, makes it less likely that the increases simply reflected another change affecting all imports at the same time.

The price response is the first part of the incidence question: it suggests that much of the tariff was passed through into prices in the United States, rather than absorbed by foreign exporters through lower prices.

Higher prices came with a sharp shift in trade flows

The same comparisons show a marked decline in the value of imports exposed to tariffs. In some waves, import values fell to around 0.6 or 0.5 of their pre-tariff level—a drop of roughly 40 or 50 percent. For several waves, imports were fairly steady before the tariffs, then fell after they took effect. Redding points to more anticipation in the first wave, which included solar panels and washing machines and had attracted speculation before implementation.

Those declines do not necessarily mean that US buyers stopped sourcing the goods altogether. Many early tariffs targeted Chinese products. If the United States imposed a tariff on baseball gloves from China, importers could instead buy them from Vietnam. Such substitution can produce a large reduction in imports from the targeted country even as sourcing shifts elsewhere.

The broader pattern was a reallocation of US imports from China toward other Asian countries, including Vietnam, Cambodia and Thailand. A chart Redding presents shows China’s import share falling as the share of other Asian suppliers rises. He says the change continued beyond the initial tariff waves: China’s share of US imports, which had been around 23 percent, had fallen to less than 11 percent by the time of his discussion.

He cautions that the decline in direct imports from China overstates how far US supply chains have separated from China. Goods imported from Vietnam and other countries can contain Chinese components; trade has also been diverted through Mexico. The shift in sourcing is real, but lower direct imports do not mean that indirect exposure has disappeared.

The conventional model predicts a price shift abroad, too

Redding uses the standard import-demand and export-supply framework to connect the observed price and quantity changes to economic welfare. US import demand slopes downward: when a good becomes more expensive, US consumers buy less and domestic producers have more incentive to make it, reducing imports. Foreign export supply slopes upward: higher prices encourage foreign firms to produce more and foreign consumers to buy less, increasing the amount available for export.

A tariff creates a wedge between the price in the US market and the price received by foreign exporters. In the conventional model, the tariff reduces imports, raises the US price and lowers the price received abroad. The gap between those prices is the tariff payment to the US government.

The framework separates several effects. Higher prices and reduced trade harm US firms and consumers. The government receives tariff revenue. For a country large enough to influence world prices, the lower price received by foreign exporters can also benefit the country as a whole: it is a terms-of-trade improvement. In this version of the model, the net welfare effect is ambiguous. The gain from the terms-of-trade improvement could be larger or smaller than the losses from distorted production and consumption.

But Redding says the measured price increases do not show the corresponding fall in foreign exporters’ prices. The US import price rose roughly by the full amount of the tariff, a result he says has also appeared in other empirical studies, with some caveats. If foreign prices do not fall, the standard model’s terms-of-trade gain disappears.

In that case, the export-supply curve facing the United States is effectively horizontal: foreign exporters continue to receive the same world-market price, while the tariff raises the price paid in the United States. The tariff is passed through domestically, and the welfare effect in the model is unambiguously negative.

The estimated losses were measurable but not large relative to the economy

Under the assumption that foreign exporters’ prices did not change, Redding estimates the welfare loss as the deadweight loss from reduced imports. In the diagram, this is the triangular area associated with trade that no longer occurs even though the United States would value those imports more than their world-market cost. Assuming the import-demand curve is approximately straight, the triangle can be calculated from the tariff increase and the change in import quantities before and after the tariff.

The government’s tariff revenue is not automatically an additional net cost to the United States. If a dollar in government hands is worth as much as a dollar in private hands, the revenue is a transfer between domestic parties. If collecting and administering the tax imposes costs, Redding says, some of that transfer could add to the welfare loss.

$8.2 billion
Redding’s estimated cumulative deadweight welfare loss from the 2018 tariffs

His estimates show deadweight losses and tariff revenue rising as successive waves brought more goods under tariffs and increased tariff rates. The table distinguishes the cumulative figures through December from the December-only values.

Measure2018 cumulative totalDecember value
Deadweight welfare loss$8.2 billion$1.4 billion
Tariff revenue$15.6 billion$3.2 billion
Total cost to importers$23.8 billion$4.7 billion
Redding’s estimated effects of the 2018 tariffs, in current dollars

Redding describes the annual welfare loss as not huge—less than one percent of GDP—but still a reduction in US welfare. He compares the estimate with a welfare gain from NAFTA estimated by economist Lorenzo Caliendo, describing the two as roughly similar in size. For perspective on a much more extreme restriction, he cites economist Doug Irwin’s estimate that the 1807 embargo, which prohibited trade through US ports, cost the United States around 5 percent of GDP.

The frontier, in your inbox tomorrow at 08:00.

Sign up free. Pick the industry Briefs you want. Tomorrow morning, they land. No credit card.

Sign up free