338 And the Vessel Clause: A Tariff You Can Drive Around

In this part two of our three-part article on Section 338, we explore the “imported in a vessel of…” clause.

Now that the SCOTUS has struck down the IEEPA tariffs, what can we expect as a reaction from the Administration?  Two words: section 338 (OK, one word and one number.  Sue me.)

Section 338 of the Tariff Act of 1930 contains a provision that sounds, at first glance, like a powerful, modern tool. It allows the President to impose duties not only on goods that are the “growth or product of” a country, but also on goods “imported in a vessel of” that country. Translated into 2025 terms: not just what the goods are, but how they arrive.

That sounds potent. It is also unusually easy to work around. Well, in theory.  Truckers: you’re welcome?  Yes. You’re welcome.

The Numbers Look Impressive

Let’s start with a simple estimate.  I’ve kept the assumptions and estimates conservative.

Total U.S. goods imports are roughly $3.3 trillion annually (U.S. Census Bureau, recent trade data). If approximately 11% of those goods arrive on Chinese-flagged vessels, as reported in recent maritime analysis (MarineTraffic, 2025), that implies on the order of $363 billion of imports tied to Chinese-flagged carriage.

At a 50% ad valorem rate, that produces a theoretical tariff exposure of approximately $181 billion.

That is a serious number. It is also, in practice, not a real one.

The Tax Base Is Not the Goods — It’s the Route

The key feature of Section 338 is that it ties duty not only to country of origin, but to the vessel used for importation. That creates an unusual dynamic.

Country of origin (shameless promotion: stay tuned for my book) is sticky. It takes factories, capital, and time to change.  

Shipping routes are not.

Containerized cargo is highly fungible. The average value of goods in a container is often estimated in the range of $40,000–$43,000, depending on the mix of goods. That same container can be loaded onto a different vessel, routed through a different port, or discharged in a different country with relatively little friction.

Which means the taxable base here is not production, it’s logistics decisions.

That is a much weaker foundation.  

For those of you in supply chain management and logistics: you’re welcome.

The First Response: Reflagging

Section 338 uses the phrase “vessel of” a country. In maritime law, that means flag state, not ownership.

A Chinese-owned vessel registered in Liberia is not a Chinese vessel. A Liberian-owned vessel flying the Chinese flag a “vessel of” China.  

Changing the flag of a vessel is not trivial—but it is far easier than moving a factory. If a tariff attaches to the flag, there is an immediate incentive to reflag the fleet.

A statute written in 1930 assumed that ships carried national policy. In 2025, they often carry tax advantages.

The Second Response: Rerouting

Even without reflagging, there is a simpler option.

Section 338 applies when goods are imported into the United States on a vessel of the targeted country. It does not apply when a foreign-flagged vessel is used somewhere earlier in the supply chain.

So, consider two scenarios:

• Cargo arrives in Los Angeles on a Chinese-flagged vessel → vessel clause applies
• The same cargo arrives in Manzanillo, Mexico, and is trucked into Laredo → vessel clause does not apply

Same goods. Same origin. Different route.

One is hit with a 338 duty at 50%. The other is not.

If that sounds like an invitation to reroute cargo through Mexico and Canada, it is.

(Trade policy rarely eliminates behavior. It usually just moves it. Often quite literally.)

The Effect of the Second Response: HMF

At a rate of 0.125%, the Harbor Maintenance Fee generates roughly $2 billion annually. If approximately $300–350 billion of imports currently arrive on Chinese-flagged vessels, that volume alone represents roughly $400–450 million in HMF revenue. A vessel-based tariff that incentivizes rerouting through Mexico or Canada would not eliminate the trade — it would eliminate the fee.

Section 338 won’t just shift routing, it will also reduce federal revenue elsewhere from decreased Harbor Maintenance Fees.  

And let’s remember that, unlike the Merchandise Processing fee which has a minimum and maximum that are adjusted annually, the HMF is a flat rate with no ceiling.

Ports Lose, Trucks Win

Rerouting has second-order effects. Cargo that enters through a U.S. seaport is subject to the Harbor Maintenance Fee (HMF). Cargo that crosses the border by truck or rail is not. A vessel-based tariff therefore does not just shift carriers—it shifts mode of entry.

The likely result is less volume at U.S. sea ports, more volume at land crossings, and a set of congestion and infrastructure issues that will be someone else’s problem.

Tariffs are supposed to influence trade flows. This one may mostly influence traffic patterns.

True store: one time, as I was in a car being driven from Juarez to El Paso on a Friday afternoon,  traffic was so bad I got out of the car and walked across the bridge.  I told people “I walked from Mexico to the US!”  Good times.

Customs broker on the southern border: you’re welcome.  Buckle up, sharpen your pencils, and get ready to rumble.

The Third Problem: Two Tariffs for One Shipment

Modern tariff systems are built on country of origin. Section 338 adds a second variable: the vessel used for importation.

That opens the door to a layered structure:

• One duty based on origin
• Another based on the flag of the importing vessel

In other words, the same product could face different duty treatment depending on which ship carried it across the border.

That is not how the system is currently designed to operate. It is also not especially easy to explain to a CFO.

A 1930 Tool in a 2025 Market

Recent industry reporting has already explored proposals to impose significant fees on “China-linked” vessels—defined by flag, ownership, or construction—and warned that such measures could disrupt container shipping markets just as they are stabilizing post-pandemic.

That debate highlights the core issue. In modern shipping, “Chinese” can mean several different things, none of which align neatly with the statutory language of Section 338. The statute assumes those concepts overlap. The market does not.

Conclusion

The vessel clause looks like a powerful lever. In practice, it is a lever attached to something that moves.

If roughly $300–350 billion of imports are potentially exposed, a 50% tariff produces a large theoretical number. But the relevant variable is not trade volume, it is behavior. Ships can be reflagged. Routes can be changed. Cargo can move. A tariff that depends on those variables is less a tax on trade than a tax on choices that can be changed.

For an interesting show on narcotraffic that shows a vessel changing identity in the middle of the open sea, watch ZeroZeroZero: outstanding writing.

And when the choices change, the revenue tends to follow.