America’s Factory Floor

The trade numbers from the first four episodes tell you what crosses the border. The investment numbers tell you why. When General Motors opened its Silao truck plant in 1995, that was a foreign direct investment. When Ford expanded its Hermosillo assembly line in 2005, that was FDI. When BMW built a $1 billion factory in San Luis Potosí in 2019, when Audi invested $1.3 billion in Puebla, when Constellation Brands built the largest brewery in the Western Hemisphere in Nava, Coahuila — every dollar of those investments shows up in the BEA’s direct investment data. And every car, truck, beer, and auto part that those factories subsequently ship north shows up in the trade deficit.

This is the connection that trade statistics alone can never show: much of the U.S. trade deficit with Mexico is American companies importing goods from factories they own. When GM ships a Silverado from Silao to a dealership in Houston, the BEA records that as a Mexican import. The $45,000 truck adds to the deficit. But GM owns the factory, employed the workers, engineered the vehicle in Warren, Michigan, and will book the profit in Detroit. The “deficit” is really an intra-company transaction — America trading with an extension of itself that happens to sit on the other side of the Rio Grande.

In 2024, U.S. companies invested a record $18.2 billion in Mexico — the highest single-year figure in the BEA dataset. This wasn’t an anomaly. The three-year total from 2022 to 2024 was $40 billion, nearly double the $22 billion invested in the preceding three years (2019–2021). The nearshoring boom that began as a whisper during Trump’s tariff wars with China has become a roar. Companies that once debated whether to shift production from China to Mexico are no longer debating. They’re building.

Over the full period covered by BEA data (2003–2024), cumulative U.S. direct investment into Mexico totals roughly $173 billion. That figure understates the true capital deployed, because it measures net annual flows and excludes reinvested earnings that don’t cross the border. The actual stock of American-owned productive assets in Mexico — factories, warehouses, equipment, technology — is estimated at over $250 billion. This capital stock is the physical infrastructure behind the $850 billion in annual trade that we mapped in Episode 1.

Three shocks in six years turned Mexico from a steady FDI destination into the hottest manufacturing market on the planet. Shock One: the China tariffs. In 2018 and 2019, President Trump imposed 25% tariffs on $250 billion of Chinese goods. Overnight, the math changed for every multinational with Chinese factories. A product that cost $10 to make in Shenzhen now carried a $2.50 tariff on arrival in the U.S. The same product made in Monterrey crossed the border duty-free under NAFTA. Samsung was among the first to move, relocating TV production from China to Tijuana. GE Appliances shifted washing machine production from China to its San Luis Potosí facility.

Shock Two: COVID. The pandemic exposed the fragility of 8,000-mile supply chains. When Chinese factories shut down in early 2020, American manufacturers discovered they couldn’t get parts for weeks or months. The lesson was visceral: distance equals risk. Mexico, with its 1,954-mile border and same-day trucking capability, offered what China never could — resilience through proximity. The consulting firm Kearney coined the term “nearshoring” in its 2020 reshoring index, and the concept became a corporate strategy almost overnight. By 2021, industrial real estate vacancy in Mexico’s northern manufacturing corridor had fallen below 2%, and speculative warehouse construction was booming in Monterrey, Juárez, and Saltillo.

Shock Three: geopolitical decoupling. The U.S.-China relationship deteriorated beyond trade into technology controls, semiconductor restrictions, and military tensions over Taiwan. American companies that once viewed China risk as manageable began treating it as existential. Apple’s decision to diversify iPhone assembly to India made headlines, but dozens of smaller manufacturers quietly chose Mexico instead. The advantages were practical: Mexico shares time zones with the U.S., its engineers are trained in American universities, English proficiency is rising, and USMCA guarantees tariff-free access for qualifying goods. For a midsize manufacturer making industrial components, moving from Dongguan to Querétaro made more sense than moving to Vietnam or India, because the customer base — American factories and distributors — was a six-hour truck ride away, not a three-week ocean crossing.

The result is visible in the data. U.S. FDI into Mexico averaged $7.4 billion per year from 2003 to 2019. From 2021 to 2024, it averaged $12.7 billion — a 70% increase. The 2024 figure of $18.2 billion alone is higher than any single year before 2022. And the pipeline is growing: Mexico’s economy ministry reported $36 billion in announced but not-yet-completed investment projects as of late 2024, a backlog that suggests the nearshoring wave has years of momentum left. As we’ll explore in Episode 9, some of this investment is Chinese companies using Mexico as a backdoor to the American market — a development that could fundamentally alter the political dynamics of NAFTA.

Here is the number that should feature in every trade debate but never does: $17.8 billion. That’s how much American companies earned in direct investment income from their Mexican operations in 2024. Profits from GM’s truck plants, from Constellation Brands’ breweries, from Honeywell’s aerospace factories, from Walmart’s 2,700 Mexican stores, from Citigroup’s Banamex banking franchise — all of it flows back to American parent companies and their shareholders.

In 2024, Mexico paid $6 billion in FDI income to its investors in the United States. The net investment income balance was +$11.8 billion in America’s favor. Think about what that means: the United States runs a $181 billion goods deficit with Mexico, but earns back $12 billion of that through investment income alone. Add in the $33 billion services surplus (covered in Episode 7), and the true economic relationship is considerably less imbalanced than the goods deficit suggests.

The investment income has grown in lockstep with the FDI stock. In 2004, U.S. companies earned $7.1 billion from Mexican operations. By 2024, that had risen to $17.8 billion — a 2.5x increase that tracks the growth in installed manufacturing capacity. The factories pay for themselves. A $1 billion auto plant generating $100 million in annual profits earns back its construction cost in a decade, then continues generating returns indefinitely. This is why the FDI keeps flowing: the risk-adjusted return on manufacturing investment in Mexico has consistently outperformed alternatives. Labor costs are one-quarter of American levels. Energy is subsidized by the Mexican government. The customer — the American consumer market — is a truck ride away.

If the U.S.-Mexico FDI relationship is a one-way street, the U.S.-Canada relationship is a freeway with equal traffic in both directions. In 2024, U.S. companies invested $37.7 billion in Canada while Canadian companies invested $37.4 billion in the United States — an almost perfectly balanced exchange. This symmetry reflects the structural similarity of the two economies: both are advanced, high-wage, service-oriented nations with deep capital markets and strong rule of law.

The scale of U.S.-Canada FDI dwarfs the Mexico relationship. In any given year, two to five times more American capital flows to Canada than to Mexico. The reason is that Canadian investment is concentrated in high-value sectors — banking, insurance, mining, technology, and energy — while Mexican investment is concentrated in manufacturing. A bank acquisition or an oil sands joint venture moves billions in a single transaction. A new auto plant in Monterrey is $1–2 billion. The capital intensity per deal is simply different.

But here’s where the income story diverges sharply from Mexico’s. In 2024, U.S. companies earned $30.6 billion from their Canadian operations, while Canadian companies earned $38.8 billion from their U.S. operations. Canada is now earning more from its American investments than the U.S. earns from its Canadian ones — a reversal that occurred around 2017 and has persisted since. Canadian pension funds (CPP, OMERS, CPPIB, Caisse de dépôt) and banks (RBC, TD, BMO, Scotiabank) have become among the largest foreign investors in the United States, owning toll roads, airports, infrastructure, and commercial real estate. TD Bank alone operates 1,200 branches in the eastern United States. The U.S.-Canada investment relationship is genuinely bilateral — a partnership of equals that looks nothing like the U.S.-Mexico patron-client dynamic.

American investment in Mexico is not evenly distributed. It clusters in four industrial corridors that together constitute one of the most concentrated manufacturing zones on the planet.

The Border Corridor — Tijuana, Ciudad Juárez, Reynosa, Matamoros — is the oldest and most established. This is where the maquiladora system began in the 1960s, and it remains the heartland of electronics assembly, medical devices, and wiring harnesses. Tijuana alone has over 1,000 maquiladoras employing 230,000 workers. The advantage is raw proximity: a factory in Tijuana is closer to San Diego than most suburbs. Same-day delivery is not a logistics aspiration — it’s the default.

The Monterrey Corridor — Monterrey, Saltillo, Monclova — is Mexico’s industrial capital, home to the country’s largest conglomerates (CEMEX, FEMSA, Alfa) and an increasingly dense cluster of foreign manufacturers. Monterrey has Mexico’s highest per-capita income and its best technical university (Tec de Monterrey). John Deere, Caterpillar, Whirlpool, and Samsung all operate here. The Monterrey metropolitan area produces 10% of Mexico’s GDP with 4% of its population — a concentration of economic output that rivals Silicon Valley.

The Bajío — Guanajuato, Querétaro, Aguascalientes, San Luis Potosí — is the fastest-growing industrial region and the center of Mexico’s auto industry. GM Silao, Toyota Guanajuato, BMW San Luis Potosí, Honda Celaya, Mazda Salamanca — the cluster includes virtually every global automaker within a 200-mile radius. The Bajío’s advantage is its centrality: equidistant from both the U.S. border and Mexico City, with modern highways and rail connections to the Laredo border crossing. It has been called “Mexico’s Detroit,” but unlike Detroit, it’s still growing.

The Southeast — Puebla, Tlaxcala, Veracruz — is the oldest manufacturing region, anchored by Volkswagen’s historic Puebla plant (producing Beetles since 1964, now building the Taos and Jetta) and Audi’s $1.3 billion Q5 factory. This corridor is smaller and more specialized, focused on premium automotive and aerospace. Bombardier, Safran, and Honeywell all have aerospace operations in Querétaro and Puebla, tapping into a growing pool of Mexican aerospace engineers.

The ultimate question about FDI is whether it creates value or merely shifts it. The skeptic’s view: every dollar invested in Mexico is a dollar not invested in Ohio, and every factory built in Silao is a factory not built in Detroit. The evidence is more nuanced. U.S. manufacturing output — the total value of goods produced in American factories — reached a record $2.9 trillion in 2024, double the level when NAFTA took effect. American factories produce more than ever; they just do it with fewer workers and more sophisticated machinery. The FDI that went to Mexico didn’t replace American manufacturing — it extended it across the border, with U.S. operations specializing in high-value tasks (design, engineering, high-precision components) and Mexican operations handling labor-intensive assembly.

The return on investment has been extraordinary. Cumulative FDI into Mexico of $173 billion is now generating $17.8 billion per year in income — a 10.3% annual return. By comparison, U.S. FDI into Canada generates a 6.1% return ($30.6 billion on roughly $500 billion in cumulative investment). FDI into Europe generates roughly 5%. Into China, roughly 8%. Mexico offers the highest returns in the NAFTA system because the wage differential creates the largest value capture: American companies design and market products at American prices while manufacturing at Mexican costs. The spread between $8/hour and $32/hour is $24/hour per worker, across millions of workers, year after year.

This is why the trade deficit alone is a misleading scorecard. The United States “loses” $181 billion per year in goods trade with Mexico. It “earns back” $18 billion in FDI income, $33 billion in services surplus, and an incalculable amount in lower consumer prices. The factories that American companies built in Mexico aren’t foreign. They’re extensions of American balance sheets, funded by American capital, generating American profits, and selling to American consumers. The trade deficit is not what it seems. It’s the accounting residue of a continental manufacturing system that American companies designed, built, and continue to control.