As the United States confronts the fallout from a Supreme Court decision striking down President Donald Trump’s sweeping global tariffs, new research from McMaster economist Pau Pujolas offers timely evidence on the lacklustre effectiveness of U.S. tariff policy.
Tariff revenue starts to fall once tariffs cross 25 per cent, Pujolas and his co-author, Jack Rossbach of Georgetown University in Qatar, found.
They developed a global trade model using data from the World Trade Organization and other sources to understand how much tariffs can increase before they stop raising money for the government and start making citizens worse off.
The question became urgent as the U.S. administration floated unprecedented increases.
“When we started hearing proposals for tariffs at 100 per cent or event 140 per cent, we wanted to measure their real impact. At what point do these tariffs stop raising money and start making people worse off?” says Pujolas, an associate professor in the Department of Economics.
Economists labelled the relationship between tax rates and tax revenue the Laffer curve: revenue rises with higher taxes until it hits a peak and then falls when rates become so high that economic activity stops. Pujolas says the same logic applies to international trade but had never been calculated with modern data.
“It’s well known that if you tax nothing, you collect nothing. And if you tax everything, you also collect nothing because economic activity collapses,” he says.
“Somewhere in between, there must be a peak. What hadn’t been studied is where that peak is in international trade.”
Their findings show that tariff revenue peaks at roughly 25 per cent. This was found to be consistent across major U.S. trading partners. Anything above that threshold causes import volumes to shrink so sharply that tariff revenue starts to fall.
But what is best for the treasury is not what is best for people. Tariffs that maximize the well-being of American households are far lower, peaking somewhere in the 0- to 10-per-cent range. Once tariffs exceed that peak, the costs to households mount quickly.
By early 2026, U.S. policy had surged past these limits. The average tariffs rate jumped from 2 per cent to nearly 19.5 per cent, and one-fifth of all tariff lines had crossed their own revenue maximizing peaks, meaning imports were falling enough that tariffs were reducing revenue instead of increasing it. And at those rates, tariffs were harming households.
“When tariffs get too high, they become economically nonsensical. You reduce government revenue and you make your citizens worse off,” Pujolas says. “We saw that happening in the U.S. by early 2026.”
One of the reasons the costs to American households rose is that they accrued on multiple fronts. “The first cost is higher prices on imported goods. But the second cost is just as important: Domestic firms raise prices too because their foreign competitors have been made artificially expensive,” says Pujolas.
With competition weakened, less efficient producers stay afloat and household purchasing power erodes.
The most dramatic case was China, where U.S. tariffs averaged 42.2 per cent, far beyond revenue-maximizing levels.
The model estimates that for every $1 of tariffs the U.S. collected on Chinese imports, it imposed a cost of $1.72 on American households. Tariffs raised import costs, increased prices throughout the economy, and reduced real income — even after accounting for the tariff revenue rebate.
“At those levels, the U.S. chose to harm its own citizens as well as its treasury in order to inflict pain on China. Acting like this must obey political or strategic reasons, but not economic ones,” Pujolas says.
The study also illustrates how global dynamics enabled the tariff escalation: While few trading partners retaliated, they did so in isolated, uncoordinated ways, which Pujolas says allowed the U.S. to continue raising tariffs with limited domestic consequences.
If major trading partners had coordinated, they would have raised the domestic cost of Trump’s tariffs and likely prevented further escalation.
“If other countries had retaliated together, the U.S. would have felt the cost immediately. That would have stopped the trade war,” he says.
“Coordination against a bully works. The numbers show it.”