Tariffs Are Now a Cost Line, Not a Headline: 0.9 Points of PCE and Counting

Tariff announcements have stopped producing the market gyrations they caused in 2025, according to Kent Jones, professor emeritus of economics at Babson College, writing at The Conversation ^[1][20]. What they move instead is the price level: the Dallas Fed estimates the Fed's preferred inflation measure ran at an annual 3.2% in March, against 2.3% without tariffs ^[2].

That gap is 0.9 percentage points ^[3], roughly 28% of the reported number ^[4]. It is no longer an event to be traded around. It is a recurring input cost that has already been passed through part of the way and will be passed through further.

The structural change came in February, when the Supreme Court overturned Trump's emergency tariffs, an outcome many business groups had read as the end of the tariff wars ^[5]. Five months later the administration announced a replacement set of import taxes ^[6]. These cover nearly all US imports and, in the administration's view, rest more firmly on existing US trade law and outside the reach of Supreme Court review, with more such measures envisioned ^[7]. The 2025 "Liberation Day" levies and their immediate successors turned out to be temporary; the bulk of the new ones are designed to be permanent ^[8].

The consequence for planning is stacking. Jones argues that the total cost burden is likely to rise even if headline rates never change, because the new tariffs sit on top of older ones ^[9]. The authorities in use include Section 301, Section 232 national security tariffs, and Section 338, a discrimination provision dating to the Smoot-Hawley Tariff Act of 1930 ^[15]. The new global Section 301 rates run 10% to 12.5% and can be raised at the president's discretion ^[16]. Section 301 also supports country-specific rates, including 25% on Brazil, while Section 338 was cited for an extra 50% on certain Canadian goods ^[17]. Product-specific levies of 25% to 50% cover steel, aluminium, automobiles, copper, timber, lumber and pharmaceuticals ^[18].

On incidence, the mechanics are not in dispute. Treasury sends the tariff invoice to the US business doing the importing ^[10]. Trump has deflected price criticism by claiming, erroneously, that foreigners pay ^[11], and has said US tariffs "aren't high enough" ^[12]. Importers can absorb some of it to defend share and run down inventory, but Jones expects most of it to reach consumers eventually ^[10]. The Yale Budget Lab puts current pass-through at anywhere from half to the entire cost, depending on the goods ^[13].

The margin question is the interesting one for operators. Yves Smith of Naked Capitalism notes that with corporate profits at a record share of GDP, companies could afford to carry a good deal of the tariff bill, and observes that short-termism and customer squeezing are the norm instead ^[14]. The Yale range says some categories are indeed eating part of it. The Dallas Fed number says the consumer is eating the rest.

Watch the Section 301 rate band, since it moves at presidential discretion rather than through any process a supplier can plan around ^[16]. Watch the stacking arithmetic in your own landed-cost model, because effective rates can climb with no announcement at all ^[9]. And watch sentiment: tariffs remain unpopular, economic sentiment is souring ahead of the November 2026 midterms, and approval on the economy is low ^[19].