Key Takeaways

• The SCOTUS ruling wiped out IEEPA tariffs but left the structural architecture intact – Section 232 (steel, aluminum, autos, copper), Section 301 (China), bilateral deals, and FTA-qualifying goods all survived
• Half of all US imports were already exempt from the headline rates before the ruling – the tariff regime was always a firewall with precisely shaped ports, not a wall
• A clear hierarchy emerges by strategic tier – hemisphere partners pay 1-3% effective rates, European allies 2-5%, Asian allies 4-7%, and China 15-25%
• The tariff isn’t the policy – the tariff is the lever – bilateral deals extract non-tariff concessions (market access, regulatory alignment, defense spending, purchase commitments) that matter more than the rate itself
• The 150-day Section 122 clock and the July 1 USMCA review are the two structural events that will reshape the regime next

On April 2, 2025 – “Liberation Day” – the Trump administration unveiled a poster-board chart of reciprocal tariffs on dozens of countries. China at 54%. Vietnam at 46%. Taiwan at 32%. The EU at 20%. Markets crashed 12.4% in a week. The headlines screamed trade war.

Ten months later, the picture looked nothing like that chart. Eighteen bilateral deals had been signed, half of all US imports had been exempted from the IEEPA tariffs that made up the bulk of the regime, and the gap between what was announced and what was actually being collected revealed something more interesting than protectionism: an access control system with precisely shaped exceptions serving strategic objectives.

Then on February 20, 2026, the Supreme Court struck the whole IEEPA framework down. Six to three. Chief Justice Roberts wrote that IEEPA’s grant of authority to “regulate importation” does not include the power to impose tariffs, and that the power to tax imports is “very clearly a branch of the taxing power” reserved for Congress. Within hours, the administration issued an executive order rescinding all IEEPA tariffs and slapped a temporary 10% global surcharge under Section 122 of the Trade Act of 1974 – an authority that expires in 150 days unless Congress extends it.

The effective tariff rate dropped overnight. Penn Wharton estimated the average fell from about 9.8% to roughly 5.6-6.7%. The Tax Foundation projects the 2026 weighted average at about 12.1% with Section 122 in effect, falling to 6.7% once it expires.

But here’s the thing the headlines are missing: the architecture underneath didn’t change nearly as much as the rate did.

What survived the ruling

The SCOTUS decision wiped out the IEEPA “reciprocal” tariffs (the Liberation Day rates) and the IEEPA “trafficking and immigration” tariffs (the fentanyl tariffs on China, Canada, and Mexico). Those were the headline-grabbing ones. What it did not touch:

Section 232 tariffs remain in full force – steel at 25%, aluminum at 25%, copper at 25%, automobiles at 25%, plus lumber, semiconductors, and medium and heavy trucks. These are the structural tariffs that protect the industrial inputs the domestic buildout depends on, and the administration has active Section 232 investigations into pharmaceuticals, commercial aircraft, robotics, industrial machinery, and wind turbines that could produce additional actions. None of that was affected by the ruling.

Section 301 tariffs on China remain as well. These date back to Trump’s first term, were retained by Biden, expanded in 2025, and cover hundreds of billions in Chinese imports. They were never based on IEEPA authority, so the Court’s decision doesn’t touch them.

The bilateral deals themselves remain operative. The UK framework, the Japan deal, the South Korea deal, the EU framework, Taiwan, Indonesia, Argentina, Switzerland, the CAFTA-DR hemisphere deals – all of those were negotiated against the threat of IEEPA tariffs, but the concessions the other countries made (market access, purchase commitments, regulatory alignment) aren’t contingent on those specific tariffs staying in place. Secretary Bessent stated within hours of the ruling that combining Section 122, Section 232, and Section 301 tariffs “will result in virtually unchanged tariff revenue in 2026.” Whether that’s accurate or aspirational, the signal is clear: the administration intends to maintain the pressure gradient through whatever legal authority survives.

The IEEPA tariffs were the blunt instrument – the deals, the exemptions, and the Section 232 framework were always the durable architecture underneath.

The gradient is the point

Forget the headline rate for a minute. Look at what countries actually pay, and a hierarchy appears.

Hemisphere partners pay the least. Before the SCOTUS ruling, El Salvador’s headline rate was 10%, but CAFTA-DR exempted roughly 87% of what they actually export to the US – textiles, coffee, sugar, fruit – putting their effective rate somewhere around 1-2%, and Guatemala was the same story. Chile’s copper, about 30% of their exports to the US, was exempt from reciprocal tariffs entirely under the Annex II exemption list. Argentina’s new deal, signed February 5, quadrupled the beef quota to 100,000 metric tons (80,000 new tons duty-free in 2026) and exempted 1,675 resource products from reciprocal tariffs. Post-SCOTUS, the CAFTA-DR zero-rate treatment and the FTA-qualifying goods pathways are unchanged, which means these countries are still inside the wall.

Allied partners pay a moderate rate. Japan, South Korea, Taiwan, the UK, the EU – all negotiated framework deals that brought their rates down from the Liberation Day shock numbers to manageable levels, but the real story was always the exemptions, not the rates. Taiwan’s semiconductors are over 55% of their exports to the US and were completely exempt, Ireland’s pharma is about 65% of their exports and also exempt, and in both cases the headline rate was cosmetically applied to a sliver of actual trade. The Section 232 tariffs on steel, aluminum, and autos hit everyone including allies – that’s the cost of admission – but the product-level exemptions shaped the effective rate downward for partners willing to align.

China pays the most. Even during the truce period, the stacking of Section 301 tariffs (first term, retained by Biden), the now-replaced IEEPA fentanyl tariffs, Section 232 on steel and aluminum, and the reciprocal tariff generated massive duties on $463 billion in imports. Penn Wharton found that China’s effective tariff rate reached 33.4% in December 2025, and the de minimis closure that hit Shein and Temu at effectively 120% added to the burden. China accounted for roughly 52% of all IEEPA tariff revenue despite being only about 14% of imports. Post-SCOTUS, Section 301 and Section 232 still apply and the new Section 122 surcharge stacks on top, so while China’s rate has come down from the peaks, it remains structurally higher than any other major partner by a wide margin.

When you lay it out by tier, the hierarchy is unmistakable:

If the visualization above doesn’t render, you can view it directly.

The gradient is the architecture. Countries that align with US strategic priorities – resource access, supply chain decoupling from China, defense cooperation, market access for US agriculture and energy – get the lower rates, and countries that resist or hedge get the higher ones.

The exemption structure tells you the strategy

The product-level exemptions have been remarkably consistent across the various tariff authorities, and they survived the SCOTUS ruling because most were implemented through Annex II/III proclamations separate from the core IEEPA orders.

Annex II exempts energy and energy products, semiconductors, pharmaceuticals, copper, lumber, critical minerals, and gold/bullion from reciprocal tariffs. These are the goods the US either can’t produce domestically at scale or needs to keep cheap to fuel its own industrial buildout. The exemption list reads like a bill of materials for the infrastructure and manufacturing renaissance the administration is trying to run.

FTA-qualifying goods under USMCA, CAFTA-DR, the Chile FTA, and the Colombia FTA enter at zero. Penn Wharton found that 88.2% of imports from Canada and Mexico entered duty-free under USMCA by December 2025 – a number that had been climbing sharply as importers aggressively leveraged the rules of origin to avoid the tariff wall hitting everyone else. These are the pre-existing architecture that the new regime layers on top of.

The bilateral deal carve-outs are where the strategy gets explicit. Argentina gets 100,000 tons of beef and 1,675 product exemptions in exchange for opening its market to 221 categories of US goods, committing to regulatory alignment, IP protections, and investment facilitation. Taiwan’s deal includes $44.4 billion in LNG and crude oil purchase commitments alongside a pledge to push defense spending above 3% of GDP.

In every case, the concessions run in both directions, and in every case, the non-tariff commitments matter more than the tariff reduction itself.

The tariff isn’t the policy. The tariff is the lever.

What the SCOTUS ruling actually changes

The ruling matters, but maybe not the way most coverage suggests.

The immediate effect is a reduction in the effective rate – Yale Budget Lab estimates the post-SCOTUS tariff regime raises about half what the full IEEPA structure would have, roughly $1.2 trillion over ten years versus $2.4 trillion. The household cost burden drops significantly too, with Yale estimating the price level increase falls from about 1.2% to 0.6% without IEEPA tariffs.

The refund question is enormous and unresolved. Penn Wharton estimates approximately $165-175 billion in IEEPA tariff collections through January 2026 that may be subject to refund, and over 1,000 businesses had already filed refund claims before the ruling came down. The administration has signaled it will fight refunds in court, which could drag on for years.

But the structural effect may be less dramatic than the rate change suggests. The Section 122 temporary surcharge at 10-15% replaces much of the IEEPA baseline, the Section 232 tariffs on metals, autos, and industrial goods were always the durable architecture (they survived the Obama and Biden administrations and they survived this ruling), and the bilateral deals were negotiated under IEEPA threat but the concessions won’t be walked back. No country is going to un-open its market because the legal basis for the tariff that motivated the deal changed.

The real question is what happens when the 150-day Section 122 authority expires, because Congress has to vote to extend it and that vote happens right before the 2026 midterms with tariff costs as a live political issue.

If Congress doesn’t act, the weighted average applied rate drops to roughly 6.7% – still more than triple the 2022 level of 1.5%, but a very different regime from what existed in late 2025.

Two other structural events to watch: the USMCA review kicks off July 1, 2026, which will tighten hemisphere integration with stricter rules of origin. And the Section 232 investigations into pharma, robotics, and aircraft could produce new tariff actions under authority the Court explicitly left intact.

What this means for the architecture

I’ve spent this series tracing a hemispheric system organized around convergent incentives – energy from Guyana, infrastructure cleared of Chinese influence, manufacturing in Mexico, automation at home, icebreakers in the Arctic – and the tariff regime is the economic boundary of that system, the mechanism that makes the cost gradient between “inside” and “outside” large enough to move capital, supply chains, and political alignment.

The SCOTUS ruling didn’t eliminate that boundary – it shifted the legal foundation from IEEPA to a combination of Section 232, Section 301, Section 122, and bilateral agreements. The hierarchy remains intact: hemisphere partners get the best terms, aligned allies get moderate terms, and China gets the worst terms. Countries choose which tier they want to be in by the concessions they’re willing to make.

The 10% universal baseline – whether under IEEPA or Section 122 – functions as the firewall, a default deny state where everything is taxed unless you have an exception. The bilateral deals, the FTA-qualifying goods, and the Annex exemptions are the open ports: surgically defined pathways that let specific goods through for specific strategic reasons.

Countries that get generous ports are hemisphere partners and strategic allies willing to align on supply chains, defense, and anti-China containment. Countries that stay at high rates are adversaries, the insufficiently aligned, or those being disciplined for geopolitical choices.

The headline number has bounced around all year – 13.5% in early 2026, then dropping after the ruling, potentially settling around 6-7% if Section 122 expires. But the headline describes the firewall, not the traffic flowing through it. The traffic tells a different story entirely.

Data as of February 26, 2026. Sources: Tax Foundation (tariff tracker, weighted average and effective rate estimates), Penn Wharton Budget Model (effective tariff rates, IEEPA revenue estimates, USMCA compliance rates), Yale Budget Lab (post-SCOTUS estimates, distributional analysis, price level effects), Congressional Research Service (tariff action timeline, bilateral deal tracker), Council on Foreign Relations (trade deal tracker, trade calendar), USTR (bilateral agreement fact sheets, Argentina/El Salvador/Guatemala/Taiwan/Indonesia deal texts), PIIE, US Census Bureau, SCOTUSblog (Learning Resources v. Trump ruling summary), Holland & Knight / Ropes & Gray / Sidley Austin / Troutman Pepper Locke (SCOTUS analysis and refund guidance), Brookings Institution (expert roundtable on ruling), CBS News / NPR / UPI (Argentina deal reporting), CBP Quota Bulletin 26-223 (Argentina beef TRQ implementation), Supreme Court opinion 24-1287 (February 20, 2026). Effective rates are analytical estimates based on reported exemption structures, FTA compliance rates, deal terms, and the models cited above. Country-level effective rates involve estimation and should be treated as directional rather than precise.

This is a standalone analysis. For the broader series on hemispheric architecture, start with Post 1: Follow the Pipes.