The Factory That Moved Three Times: Mexico's Nearshoring Experiment

In February 2024, María Elena Garza noticed something strange in the parking lot of Foxconn's newest facility in Apodaca, just outside Monterrey, Mexico. The industrial park's eight-lane access road, built the previous summer, was already cracking under the weight of eighteen-wheelers that ran through the night. 'We poured that concrete to Mexican specifications,' she told me, standing in her hard hat as the third shift clocked in. 'But these are Chinese logistics rhythms.' Garza is a civil engineer who has spent twenty-two years building factories in Nuevo León. She has never seen anything like the past eighteen months.

The facility she was helping expand had made the same product — power management chips for data centers — in three different countries since 2019. First in Shenzhen, then briefly in Vietnam, now here. Each move left behind workers, supply chains, and institutional knowledge. Each move was driven not by efficiency, but by tariffs, export controls, and the tectonic collision of American and Chinese industrial policy. The chips themselves are unchanged. Everything else has been upended.

What the Numbers Told Them

The scale of what is happening in Mexico defies easy summary. Foreign direct investment in Mexican manufacturing hit $36.1 billion in 2024, according to Mexico's Economy Ministry — a 12 percent increase over 2023 and nearly double the 2019 figure. But here is what those headline numbers obscure: much of this investment is not new productive capacity. It is the same factories, the same machines, sometimes the same managers, simply relocated under immense geopolitical pressure.

The thing is, nobody expected this transition to be cheap. What surprised economists was how much of the cost falls on workers and local governments rather than the multinationals orchestrating the moves. In Monterrey, the municipal government has spent $890 million since 2022 upgrading roads, water systems, and electrical grids to accommodate the manufacturing surge — capital expenditure that Mexico's federal budget did not anticipate and has only partially reimbursed.

'We are subsidizing American supply chain security,' Monterrey's economy minister, Iván Rivas, told me with a mixture of pride and frustration. 'The question is whether we get a permanent manufacturing base, or whether we're just a waystation until the next tariff regime.'

The Uncomfortable Data

The prevailing narrative in Washington and Brussels holds that nearshoring represents a permanent structural shift — the end of hyper-globalization and the dawn of regional trading blocs. But the data tells a more complicated story. A March 2025 analysis by the Inter-American Development Bank found that 62 percent of new manufacturing investment in Mexico still depends on intermediate goods imported from China. The supply chains have not decoupled. They have simply added a node.

In the electronics sector — where decoupling pressures are most intense — Mexican factories source 71 percent of their components from Chinese suppliers. Some of those suppliers have themselves opened Mexican subsidiaries, creating what economists call 'nested nearshoring': Chinese-owned facilities producing in Mexico to qualify for USMCA trade preferences, using inputs shipped from Guangdong.

This is not what American policymakers intended when they designed the Section 301 tariffs or the CHIPS Act. But supply chains are not patriotic. They follow cost gradients and regulatory arbitrage with the moral neutrality of water seeking its level.

Who Carries the Weight

Back in Apodaca, the human cost of this grand restructuring is visible in the labor market. Wages for electronics assembly workers have risen 34 percent since 2021 — good news, on its face. But inflation has consumed most of those gains, and the nature of the work has changed. The new facilities demand higher skills but offer the same maquiladora employment structures: six-month contracts, limited benefits, irregular hours.

Rosa Hernández spent twelve years at an automotive supplier before taking a job at one of the new semiconductor packaging facilities in 2024. 'They told us this was the future,' she said during a lunch break, eating tacos from a food truck that has followed the factory migration north. 'The future pays the same as the past, but they make you take a two-hour bus ride to get there.'

The housing crisis has become the most visible symptom of Mexico's nearshoring boom. Workers are streaming into cities whose infrastructure was built for a different era. In Saltillo, two hours west of Monterrey, informal settlements have expanded by 23 percent since 2022. Local officials lack the fiscal tools to respond; property tax collection rates remain below 40 percent.

The Parallel Experiment in Vietnam

Mexico is not the only laboratory for this restructuring. Vietnam received $23.2 billion in foreign direct investment pledges in 2024, much of it in electronics and solar panel manufacturing. But the Vietnamese experiment reveals the limits of the nearshoring model. Power grid capacity has become a binding constraint; factories in the north faced 127 days of electricity rationing last year. Water stress in the Ho Chi Minh City corridor threatens semiconductor fabrication, which requires ultrapure water in vast quantities.

The World Bank's latest assessment of Vietnam's infrastructure gap estimated that the country would need $25 billion annually in infrastructure investment through 2030 to sustain current manufacturing growth rates. The government's capital expenditure budget provides roughly one-third of that.

Here is what this means: the destinations that policymakers have chosen for supply chain diversification — Mexico, Vietnam, India, Indonesia — all face severe infrastructure deficits. Decoupling from China means coupling to countries whose governments cannot build fast enough to absorb the capital flowing their way.

What the Economists Disagree About

The academic debate over nearshoring has fractured along predictable lines. Trade economists at institutions like the Peterson Institute argue that the efficiency losses are modest — perhaps 2 to 3 percent of final goods prices — and worth paying for supply chain resilience. Caroline Freund, the dean of the UC San Diego School of Global Policy, told me that 'the real question is not whether nearshoring is efficient, but whether the alternative — concentrated supply chains in geopolitically unstable relationships — is sustainable at all.'

But critics, including economists at the World Trade Organization, warn that the proliferation of bilateral trade arrangements and industrial policy subsidies is fragmenting the global trading system beyond repair. A February 2025 WTO report found that trade-restrictive measures now cover 18 percent of global merchandise trade, up from 9 percent in 2019. The organization's dispute settlement mechanism remains paralyzed; the United States has blocked judicial appointments since 2019.

'We have entered a period of trade architecture without architects,' said Ngozi Okonjo-Iweala, the WTO Director-General, in a March 2025 speech in Geneva. 'Everyone is building walls, and no one is drawing blueprints.'

What We Still Don't Know

María Elena Garza is still pouring concrete in Apodaca. The facility she is helping build will eventually employ 4,200 workers — a significant addition to the regional economy. But she has watched enough factories arrive and depart to know that permanence is never guaranteed. 'They moved here because of tariffs,' she said. 'What happens when the tariffs change?'

It is the question that haunts the entire nearshoring experiment. The policy architecture driving this migration — Section 301 tariffs, CHIPS Act subsidies, EU Carbon Border Adjustment Mechanism — rests on political foundations that could shift with any election cycle. Companies are making fifteen-year capital allocation decisions based on regulatory environments that could change in two.

The deeper uncertainty is whether supply chain resilience is even achievable through geographic diversification. The pandemic revealed that disruptions propagate through networks in ways that simple relocation cannot fix. A 2024 McKinsey analysis found that companies with diversified supply chains experienced only marginally fewer disruptions than those dependent on single sources — they were simply disrupted in different places.

For now, the factories keep moving. Workers adapt, cities strain, governments borrow. The power management chips that left Shenzhen and passed through Vietnam now emerge from Nuevo León, marked as Mexican exports, qualifying for duty-free entry into the United States. The supply chain has not broken. It has just become longer, more expensive, and no more resilient than before. The thing is, nobody knows yet whether this is a transition or a trap.