[amluto]

The entire new tariff scheme is napkin math. Although it could also be napkin math done by some LLM.

If it was real math, it wouldn’t assume that there is some magical trade elasticity that is completely linear in the tariff rate. And real math might notice that the prices of goods that are subject to tariffs are an utterly absurd measure of value, cost, movement of money, or anything else.

Consider:

A US company does a bunch of R&D and designs a widget. They pay $10 each to a Chinese factory to manufacture it. They warehouse the widgets in Hong Kong. Each widget purchased by a US customer results in a “$100” item being imported. $90 stays in the US. $10 goes to China.

The same company does exactly the same thing except they ship in bulk to a US warehouse. The imported item is now “$10”. The tariff is 1/10 as much, the napkin math sees 1/10 as much trade imbalance, but the economic effect of the import is identical.

Or maybe they ship from Hong Kong to a French customer. This should be seen as an export from the US to France with $90 and an export from China to France worth $10. But I think it’s invisible to the napkin math.

Now consider that the US is home to some wildly successful companies with names like AMD and Nvidia. They sell chips for thousands of dollars each, worldwide. They pay TSMC quite a lot less to make them. If they warehouse in the US, they may be screwed now! If they ship from Taiwan to a buyer somewhere else, the US has, in effect, exported quite close to the full sale price of that chip, but no trade goods ever touched US soil. Can the napkin math sees that?

You can bet that several other countries use brains instead of napkins and will have no difficulty thinking that they could retaliate by restricting or taxing of these US-designed goods even if they’re imported from elsewhere. And China is working very hard to make their own alternatives, and they will surely be willing to export them.

(Don’t forget: The UK and Israel have CPU design expertise. ASML is dependent on tin zapping tech from San Diego, but they’re an EU company. And it looks like the successor to that tin zapping tech might be free electron lasers, and that technology come from US national labs and universities, but other countries also have FELs, and the nerdy physicists who fiddle with them are not happy with the US government right now.)

[walterbell]

> The entire new tariff scheme is napkin math.

See the 2024 paper (40 pages) by Stephen Miran, current chair of the Council of Economic Advisers, which has influenced tariff policy, https://news.ycombinator.com/item?id=43589350

https://financialpost.com/news/stephen-miran-economist-trump...

> Miran.. points to Trump’s application of tariffs on China in 2018-2019, which he argues “passed with little discernible macroeconomic consequence.” He adds that during that time the U.S. dollar rose to offset the macroeconomic impact of the tariffs and resulted in significant revenue for the U.S. Treasury.. “The effective tariff rate on Chinese imports increased by 17.9 percentage points from the start of the trade war in 2018 to the maximum tariff rate in 2019,” the report said. “As the financial markets digested the news, the Chinese renminbi depreciated against the dollar over this period by 13.7 per cent, so that the after-tariff USD import price rose by 4.1 per cent.”

[kristiandupont]

Whatever paper may have "influenced" the policy, here is the official statement and it is very much napkin math and nothing else.

https://ustr.gov/issue-areas/reciprocal-tariff-calculations

[walterbell]

That "napkin math" guide is like "enterprise pricing"—a starting point—for negotiation defined by the official White House Executive Order, https://www.whitehouse.gov/presidential-actions/2025/04/regu...

  (c)  Should any trading partner take significant steps to remedy non-reciprocal trade arrangements and align sufficiently with the United States on economic and national security matters, I may further modify the HTSUS to decrease or limit in scope the duties imposed under this order.

Lobbyists and trade negotiators can read 700 words of strawman "napkin math", or 40 pages by CEA chair, or interviews with administration officials, to inform their negotiating position.

[amluto]

This is nonsense. The “non-reciprocal trade arrangements” mentioned don’t obviously exist at all, and, to the extent they exist, they are certainly not measured in any meaningful respect by the formulas going into the tariffs.

Having spent a whopping ten minutes finding official data (and I have no idea how good this data is as a whole — probably mediocre but far better than whatever nonsense the USTR is doing):

(a) The EU seems to thing that they run a trade surplus with us in goods and a deficit in services, and they’re close to balancing out.

(b) The US’s own data shows a net surplus with some countries and a net deficit with others. See here, page 28:

https://www.census.gov/foreign-trade/Press-Release/current_p...

If the US was trying to negotiate sensibly and to identify anything remotely non-reciprocal, they would be rewarding the countries with positive numbers in that table! The US should be delighted to trade with Australia, Brazil, Belgium, the Netherlands, etc! By all means, we should buy more widgets and sell more fancy services!

As a silly analogy: if your housemate goes on a wild drunken rampage and starts trashing everything in sight while screaming “you need to reciprocate and wash the dishes more often and I will continue trashing things until you wash the dishes reciprocally,” the situation is not a viable negotiation tactic.