When the President Says 50%, Read the Fine Print

Section 338 Is Not the Lever You Think It Is

Part Three of Our Three-part Article on Section 338

In December 2025, we published “Section 338 and the Ghost of Smoot-Hawley,” the first of our three-part series on Section 338.

In February, 2026 we published “338 and The Vessel Clause: A Tariff You Can Drive Around,” the second part of our three-part series.

This is the third part of our three-part series.  Trilogy.  This is my trilogy (makes me sound like an author).

Yesterday, President Trump posted a threat on Truth Social: a “Country supplying Military Weapons to Iran will face a 50% tariff on all goods sold to the United States, effective immediately, there will be no exclusions or exceptions!”

Anyway, the number got attention. It was supposed to.

Politico reported Wednesday morning that while Trump has often wielded tariffs to pressure other countries, the Supreme Court this spring took away his main legal tool, and that his legal path on this threat is murky. Politico noted that Trump could try to use Section 338 of the Tariff Act of 1930 (thanks Politico!), which allows the president to impose tariffs of up to 50% — but flagged it as “a legal stretch,” given that the statute is designed to combat foreign discriminatory trade practices against U.S. goods, not weapons sales to third countries.

Where 50% Comes From

Section 338 of the Tariff Act of 1930 does authorize the President to impose additional duties of up to 50% ad valorem on goods from countries that discriminate against U.S. commerce. The number is real. The statute is real. It remains on the books at 19 U.S.C. § 1338.

But the statute has some problems that are not getting mentioned in today’s coverage.

• First, Section 338(d) — the broad retaliatory tariff provision — was almost certainly superseded by the Reciprocal Tariff Agreements Act of 1934. 
• Then in 1947, The General Agreement on Tariffs and Trade, a 1947 agreement between countries intended to lower trade barriers (tariffs, quotas, licenses…) and promote growth was enacted.  GATT was not designed in a vacuum. Its architects had watched Smoot-Hawley trigger a retaliatory spiral that deepened the Great Depression and destabilized international commerce. The multilateral trading system that emerged after World War II was, in no small part, a deliberate rejection of that model. That assessment is correct. It also understates the problem considerably.

• Next, in 1962, Section 252 of the Trade Expansion Act, covered the same subject. Legal scholars raised this during the IEEPA litigation earlier this year. 

• Finally, the Trade Act of 1974 gave us sections 301, 122, etc. that we all know so well today.  Maybe we don’t know them so well, but we know them.

No court has formally declared Section 338(d) dead because no president has ever actually used it to impose a tariff. The coffin has no lid, but the body is in it.

Even if 338(d) survived a legal challenge, the statute is not a same-day lever. The President must find as a fact that discrimination exists, issue a proclamation, and wait 30 days before the duty takes effect. “Effective immediately” is not how Section 338 works.

The Implied Repeal Problem

This is the part most coverage skips.

When Congress passed the Trade Expansion Act of 1962 and later the Trade Act of 1974, it created new retaliatory tariff mechanisms — Section 252 and Section 301 — that covered the same ground as Section 338(d). Under the legal doctrine of implied repeal, when a later statute covers the whole subject of an earlier one and is clearly intended as a substitute, the earlier provision is superseded even without an explicit repeal.  In the nomenclature of inventory control: LIFO.

Section 338(d) has never received a ceremonial burial because no administration has ever tested it in court. The current administration’s own lawyers appear to know this, which is why they reached for Section 301 and Section 122 after IEEPA was struck down — not Section 338.

What 50% Actually Gets You

Here is the analytical problem nobody is talking about.

China currently faces stacked duties under Section 301, Section 232, and the Section 122 global surcharge. On many product categories, the effective duty rate already exceeds 50%. A Section 338 retaliatory tariff, capped at 50%, would in some cases represent a reduction in effective duty burden on Chinese goods, not an escalation.

Remember: because 301 was intended to replace 338, then you cannot (we presume) invoke both 338 and 301 simultaneously.  In other words, these tariffs cannot be stacked.

If the goal is maximum economic pressure on China, Section 338 is not the instrument. It was designed for a different trade architecture, with a ceiling that made sense in 1930 and looks modest against today’s stacked duty regime.

What Would Actually Deliver Maximum Pressure

Moving China to Column 2 of the HTSUS would. Column 2 rates — the statutory rates set in 1930 and never negotiated down through GATT — run substantially higher than MFN rates across nearly every product category that matters. The delta on electronics and machinery alone, the two largest import categories from China, is potentially 20 to 35 percentage points above current MFN rates, before any stacked duties.

But Column 2 requires an act of Congress. Simple majority in both chambers, presidential signature, and realistically 60 Senate votes to survive a filibuster. No proclamation gets you there. No emergency authority gets you there. The President cannot do it alone.  And this path seems pretty much impossible in today’s environment.

The Bottom Line

When the President posts a 50% tariff threat and the number matches Section 338’s statutory cap, the connection is not coincidental. Politico was right to name 338 as the likely candidate. They were also right to call it a legal stretch — though the stretch is longer than a single paragraph can convey.

Section 338 may be available in limited form. It may not survive an implied repeal challenge. It does not work immediately. And its ceiling may already be below what importers from China are paying today on many product lines.

The threat is real as a signal. As a legal instrument, it is considerably more complicated than a Truth Social post suggests.