The Tariff Shell Game, Part II: 25 States in the Audience

The Audience Grows

In our last article, we asked a question that still has no answer: if the tariff policy is working, why won’t they let the courts and Congress see it? Three weeks later, the question has gotten bigger. The Canadian tariffs do not take effect until August 19, the legal challenges were already being prepared before that date, and twenty-five states have sued over a different tariff program entirely.

Let us be fair to the administration first, because fairness is where this has to start. A defender of the policy would say it is being resourceful. The courts keep blocking one legal path, so the executive finds another, each time choosing a statute that fits the moment. IEEPA failed, so Section 122. Section 122 expired, so Section 301 and Section 232. Now Section 338, a provision of the Tariff Act of 1930 that, as the Independent Institute notes, no president has ever used to impose tariffs before this year.

That reading was plausible in February. It is harder to maintain in August, and the reason is the calendar.

The 50 percent tariffs on Canadian cars, cheese, and a long list of other goods take effect on August 19 - fourteen days from today. The legal challenges were already being mapped out on August 3, before a single dollar of the new tariff had been collected. And this week, twenty-five states filed suit in the US Court of International Trade over a separate tariff program covering sixty trading partners, calling it a pretext for replacing the tariffs the Supreme Court struck down in February, as The Guardian reported.

A policy that needs a new legal costume every few months is not a policy. It is a shell game. This is the sequel to the first article in this series, and the game has grown a second table.

The Sequence

Walk the sequence and the pattern is not subtle. It is a march.

It began with IEEPA. The Supreme Court ruled on February 20, 2026, that the International Emergency Economic Powers Act does not authorize the president to impose tariffs. The decision was 6-3, and the Penn Wharton Budget Model summarized what followed: the ruling did not even explicitly order refunds, but the Court of International Trade did. The government then appealed the refund order, arguing that refunds for finally liquidated entries require importer-specific court orders. The appeal is still moving through the courts. Money collected under a struck-down law is still being fought over, line by line.

Then came the stopgap. Section 122 of the Trade Act of 1974 permitted a 15 percent surcharge for 150 days, the story we told in the first article. That clock expired in July and the administration said the tariffs would continue under other authorities. That was the plan all along, and it was not a secret.

Then came Section 338. It is a 1930 statute designed to deter foreign retaliation: if a country discriminates against American commerce, the president may raise duties against it by up to 50 percent, the statutory maximum, as Holland & Knight explained. The administration used it to impose 50 percent tariffs on Canadian goods in three separate findings covering roughly $20 billion in annual imports, with no USMCA exemption. The Volokh Conspiracy published the prospective legal challenges on August 3, before the tariffs take effect.

And the shell game has crossed the Atlantic. On July 23, the European Commission fined Google 890 million euros, as France 24 reported. The president responded by threatening a Section 301 trade investigation against the European Union. Le Monde quoted the threat: the EU would pay a very big price. A competition fine becomes a trade investigation. A trade investigation becomes a tariff. The costume changes, the impulse does not.

Two Tables, One Policy

There are now two tariff programs, two legal fights, and one policy goal. Put them side by side and the structure becomes visible.

The forced-labor program was imposed last month on goods from sixty trading partners, at rates of 10 to 12.5 percent. The states’ complaint says what everyone watching already suspected: the forced-labor rationale is a pretext for recreating the tariffs the Supreme Court invalidated. If the motivation were genuinely about labor conditions, the tariff schedule would look different. It looks like a duplicate.

The Canadian program is the novel one. Section 338 has sat in the statute book for nearly a century, used for threats and negotiations but never for actual tariffs. The Independent Institute called out the novelty plainly: a law no president has ever used, deployed now against the country’s largest trading partner, fourteen days before it takes effect.

One policy goal connects both tables: keep the tariffs in place by any means necessary. The legal authorities are interchangeable. The tariffs are not.

The Narrowing Tell

Watch the authorities narrow. That is the tell.

IEEPA was a sweeping emergency-powers statute. Section 122 was a narrow stopgap: 15 percent, 150 days. Section 301 requires a finding of unfair trade practices. Section 232 requires a finding about national security. Section 338 requires a finding that a country discriminates against American commerce. Each new authority is more specific, more awkward, more contestable than the last.

A defensible policy survives on one authority. It is enacted by Congress, upheld by the courts, and left alone. The tariff program has gone through four authorities in eighteen months, and each one is narrower than the one before. That is not the signature of a policy that works. It is the signature of a policy that cannot stand still long enough to be examined.

The Keystone: Pre-Fighting Tariffs That Do Not Exist Yet

The keystone of the shell game is timing. Lawyers and twenty-five state attorneys general are pre-fighting tariffs that have not yet taken effect.

The Canadian tariffs do not begin until August 19. The prospective challenges were published on August 3. The twenty-five states filed their suit on August 3 as well, against a program that has been in effect for only weeks. In both cases, the legal fight is running ahead of the policy.

Nobody pre-fights a policy they expect to survive. Litigation before implementation is what you do to a policy you expect to be temporary. The states are not suing because the tariffs are working. They are suing because everyone can see the authorities cannot hold, and the only question is which one fails first.

Notice also what the policy never engages with. Tariffs are a tax paid by importers and consumers, not by foreign governments. Trade, as we explained in Trade - Why Both Sides Win, makes both sides richer, and the reason goes back to comparative advantage, the insight we explored in The Deepest and Most Beautiful Insight in Economics. A tariff is a tax on that exchange. And the previous trade war did not stop trade; it redirected it, most profitably to Vietnam, as we documented in Eighty Percent of Something.

If the policy is working, none of this should be necessary. Congress could enact a tariff law tomorrow and settle the question in public. That is the constitutional path, and it is the one path the administration has never taken. Why? Because Congress would have to weigh the costs and the benefits in the open, and the policy cannot survive that weighing.

The Lens

Next time someone says the tariffs are working, ask: why does the policy need a new legal costume every few months?

Not as an accusation. As a question. If the policy were working, one authority would be enough. It would be enacted, upheld, and left alone. Instead, the authorities keep expiring, getting struck down, or getting sued before they take effect. The refunds from the last struck-down program are still being litigated. The challenges to the next program are already written.

The audience keeps growing. First the courts. Now Congress is watching, and twenty-five states have taken their seats in the front row. The policy does not change. Only the costume does.

When a policy needs a new legal justification every few months, it is not being refined. It is being hidden. And the question from the first article still stands, with more witnesses in the room: if the policy is working, why won’t they let the courts and Congress see it?

This is the second in a series on the tariff shell game. Start here: The Tariff Shell Game