Why tariffs may reroute trade instead of bringing factories home

Economist Willy Shih says tariff hikes are more likely to push supply chains into new routes than to bring manufacturing back to the United States. In a The Verge interview, the Harvard Business School professor argued that firms respond to higher duties by shifting assembly to third countries, reworking sourcing, or moving final shipments through places such as Mexico and Vietnam.

Tariffs are changing routes, not reversing globalization

Shih’s central argument is blunt: globalization has never been driven mainly by ideology, but by cost. That matters because cost pressure tends to outlast political slogans. When tariffs rise, manufacturers do not simply surrender and rebuild entire factories in the U.S. Instead, they search for the next cheapest legal path through the trade system.

That often means splitting production across borders, using one country for parts, another for assembly, and a third for final export. According to Shih, that pattern is now visible in the way companies are adjusting to tariff pressure.

The result, he suggested, is not full reshoring but rerouting. Mexico and Vietnam have become common workarounds, while Chinese factories remain deeply embedded in key electronics supply chains, including flat-panel TVs.

What is actually happening to supply chains?

What happens when tariffs go up? Companies look for the least expensive path that still works under the rules. Shih said higher duties prompt firms to reconfigure sourcing, move assembly to third countries, or use new shipping routes that keep products moving without absorbing the full cost of the tariff.

The pressure points are already visible in North America. Shih said direct China-to-Mexico shipping routes have expanded over the past two years, a sign that trade flows are adapting rather than disappearing. He also said some goods made in China are entering the U.S. via Mexico under USMCA rules.

That is a reminder that trade policy often works less like a wall and more like a current: if one channel narrows, commerce tends to find another. The channel may be more expensive, more complicated, or more politically awkward, but it rarely closes completely unless governments seal every opening.

Why Mexico keeps coming up

Mexico sits at the center of Shih’s argument because it is the most practical rerouting hub for North American trade. Under USMCA, the region’s trade rules are less fragmented than they were under NAFTA, and Shih argued that specialization across the bloc remains the main case for freer trade.

In his view, North America works best when production is divided according to comparative advantage. That can mean parts made in one country, assembly in another, and final distribution in a third. It is not reshoring in the strict sense, but it can still shift some activity away from China and closer to the U.S. border.

Why Vietnam keeps appearing in trade talk

Vietnam has emerged in recent years as another destination for assembly and sourcing adjustments. In Shih’s telling, it functions as part of the broader response to tariff pressure: firms do not always move back to the United States, but they do move.

That movement matters because it changes who captures the jobs, the investment and the logistics revenue that tariff policy was supposed to reclaim.

Why it matters

Shih’s warning matters because it challenges a common assumption behind tariff policy: that making imported goods more expensive will automatically revive U.S. manufacturing. His view is that the response is usually more indirect and less patriotic than political rhetoric suggests.

For consumers, the implication is straightforward. If exemptions narrow and workarounds close, tariffs eventually show up in prices. Shih said higher tariffs ultimately raise domestic prices once the system runs out of easy escapes. That is one reason tariff policy can be popular in theory and inflationary in practice.

For companies, the message is that supply chains may become more complicated rather than more local. That can create new compliance costs, longer routes and more pressure on margins. For workers, it means that offshoring losses are not automatically repaired just because the policy environment turns hostile to imports.

For the U.S. economy, Shih’s comments point to a deeper structural problem: the trade deficit remains unresolved, and China’s surplus remains intact. Tariffs may change the route map, but they have not yet solved the imbalance.

The inflation angle

Shih linked the period of relatively low U.S. inflation from 2000 to 2014 in part to cheaper imports from China. That is important because it shows how global sourcing can suppress consumer prices even as it weakens domestic manufacturing in some sectors.

The tradeoff is central to the entire debate. Cheaper goods help households, but the jobs tied to those goods do not always stay in the United States. Shih said the U.S. did not do enough to retrain workers who were left behind by offshoring, which left part of the labor force exposed while prices stayed low.

Consumer demand still drives the system

Shih also argued that demand for cheap goods helped keep globalization alive. In other words, the system was sustained not only by executives or trade officials, but by consumers who wanted lower prices and were willing to buy products made through dispersed supply chains.

That point helps explain why tariffs alone have struggled to reverse the trend. As long as shoppers want inexpensive electronics, clothing and household goods, companies will search for the cheapest way to make and ship them.

Background: how we got here

The broader backdrop to Shih’s comments is the long effort by U.S. policymakers to bring manufacturing back from abroad, especially from China. Tariffs have been one of the main tools used to apply that pressure, but the record has been mixed.

USMCA replaced NAFTA and was supposed to modernize North American trade. Shih’s view is that it is preferable to the older NAFTA-era fragmentation because it supports more coherent production across the region. That does not mean a return to fully domestic production. It means a more integrated North American system in which specialization can work more efficiently.

One of the clearest symbols of failed reshoring hopes is Foxconn’s once-promised LCD project in Wisconsin. The Verge first covered that promise in 2019, when the plant was still being framed as a landmark U.S. manufacturing comeback. In practice, the example became shorthand for how hard it is to recreate electronics manufacturing on U.S. soil at scale.

Shih returned to that theme with a pointed line about the project’s limits: “I told you it’s not going to be a panel factory.” The remark underscores his broader skepticism about political promises that ignore the economics of industrial location.

By the numbers

The timeline below captures the key reference points from Shih’s remarks and The Verge’s coverage.

How supply chains work around tariffs

Shih’s argument is easiest to understand if you think of modern manufacturing as a chain of separate decisions, not a single factory. Design may happen in one country, component production in another, assembly somewhere else, and final shipment through a different route altogether.

That structure gives firms flexibility. It also gives them room to dodge the impact of tariffs without moving production fully back to the U.S. If one step becomes too expensive, the company can shift that step while keeping the rest of the chain intact.

Shih said this is why tariff hikes rarely produce the dramatic reshoring that policymakers promise. The chain adjusts, and the adjustment often happens outside the United States.

China remains central

Despite the push to reroute trade, Shih said Chinese factories remain central to flat-panel TV supply chains. That detail is important because it shows that even when final assembly shifts elsewhere, the industrial base in China is still deeply embedded in the production of major consumer electronics.

The implication is that reshoring can be partial, not total. Even if a product is assembled in Mexico or Vietnam, many of its underlying inputs may still depend on Chinese manufacturing.

What the tariff debate misses

Shih’s remarks also point to what tariff debates often leave out: industrial policy is not just about where the final product is assembled. It is about who controls component supply, logistics, financing and compliance. A country can regain some activity without regaining the full industrial ecosystem.

That is why higher tariffs can create the appearance of action without delivering a wholesale manufacturing revival. Companies may alter invoices, reroute containers and redesign procurement plans, but the global system does not disappear.

Shih said the last time the U.S. had a trade surplus and the dollar moved sharply was in the 1980s, a reminder that today’s trade patterns are the product of decades of structural change, not just one administration’s policies.

What U.S. workers lost

Shih did not frame offshoring as costless. He said some U.S. workers were left behind and that retraining was not adequate. That matters because the political appeal of tariffs often rests on the promise that manufacturing jobs can simply be restored. His critique is that the damage was real, but the remedy has been too narrow and too slow.

In other words, tariffs may punish imports without rebuilding the workforce that lost ground when production moved away. That gap helps explain why the policy debate keeps returning to the same unresolved questions.

What happens next?

The key question is how far tariff workarounds can continue before policy closes the remaining loopholes. If governments tighten rules around transshipment, country-of-origin labeling or USMCA treatment, companies may have fewer places to send production.

But the likely outcome, based on Shih’s analysis, is still not a clean return to U.S. manufacturing. More production may shift to Mexico or Vietnam instead. The geography could change again, while the basic logic of global cost competition remains in place.

Another question is how much prices rise if tariffs expand further. Shih’s view is that the cost eventually reaches consumers once exemptions and workarounds are exhausted. That would put tariffs back into the center of the inflation debate.

Finally, policymakers still have to decide whether retraining and industrial policy can do what tariffs alone have not. Shih’s comments suggest that without a broader workforce strategy, the U.S. will keep chasing a manufacturing comeback that supply chains are built to evade.

Bottom line

Shih’s message is not that trade policy is powerless. It is that tariffs do not work the way their supporters often claim. They can reroute supply chains, raise costs and shift some production. They have not, so far, forced a broad return of manufacturing to the United States.

Mexico is emerging as the most important rerouting hub, Vietnam remains an attractive alternative, and China still anchors key electronics supply chains. That is the story of globalization in 2026: not retreat, but adaptation.

“It’s cheaper.”

That short line from Shih captures the entire logic of the system. Until policy changes the economics more than the routing, the containers will keep moving — just by different paths.