‘We literally calculated tariff’: White House defends math but critics puzzled as penguins make the list

When White House Deputy Press Secretary Kush Desai posted a mathematical formula to justify new US tariffs, it sparked more confusion than clarity. The formula, presented with Greek letters and academic flourish, was meant to signal analytical rigor. 

But economists and journalists quickly broke it down—and what they found resembled back-of-the-envelope math rather than sound economic modeling. "No we literally calculated tariff and non tariff barriers," Desai posted on X, sharing a link on the calculations.

The White House used a simplified economic model to set tariffs aimed at eliminating bilateral trade deficits. Based on 2024 U.S. Census data, it assumed a price elasticity of -4 and a passthrough rate of 0.25. The resulting formula calculated the tariff needed to zero out the trade gap, ignoring broader market effects. Tariffs were capped at a minimum of zero, but the wide range—from 0% to 99%—prompted concerns about inconsistency. On average, tariffs were 50% across deficit countries and 20% globally (unweighted), or 45% and 41% respectively when weighted by imports.

Despite Desai’s assertion, the formula used to compute tariff rates boiled down to a basic ratio: the trade deficit divided by imports. While dressed up in symbols and jargon, it was quickly deconstructed by observers. “It’s like what creationism is to biology,” former Treasury Secretary Lawrence Summers reportedly said, referring to its lack of scientific merit.

China, for example, received a 34% tariff based on a $295 billion trade surplus divided by $438 billion in exports, halved from 67%. Indonesia’s 32% tariff followed the same method: a $17.9 billion deficit divided by $28 billion in imports. South Korea’s 25% tariff came from a $66 billion deficit and $133 billion in imports.

The White House said the approach was designed to compute the tariff rate that would balance bilateral trade. “Reciprocal tariffs are calculated as the tariff rate necessary to balance bilateral trade deficits between the U.S. and each of our trading partners,” it said, arguing that “persistent trade deficits are due to a combination of tariff and non-tariff factors.”

But critics say the formula ignored real-world complexities—such as services trade, where the U.S. often runs surpluses, and WTO-reported tariff rates, which differ significantly. The European Union, for instance, with an average tariff of 5%, was assigned a 20% tariff. Even uninhabited territories like Heard and McDonald Islands, home only to penguins, were hit with a 10% rate.

Adding to the intrigue, some analysts speculated that the formula’s origin might lie in AI-generated suggestions. Multiple large language models—including ChatGPT, Claude, Gemini, and Grok—reportedly proposed similar frameworks when asked to design reciprocal tariff policies. 

This led to questions of whether the White House’s model was influenced by “government by vibe coding.”