Foreign Firms Hired 8.66M Americans in 2023 — Share Slipped to 6.2%
BEA benchmark: foreign-owned U.S. affiliate jobs grew 1.9% to 8.66M, but their share of private payrolls slipped to 6.2%. Capex tells a different story.
The Bureau of Economic Analysis released its annual benchmark on foreign-owned American operations late on a quiet release calendar, and almost nobody read it. The release covers data for 2023, the most recent full year for which complete benchmark figures are available, and was published by BEA in December 2025. The headline number — 8.66 million U.S. workers employed by majority-owned U.S. affiliates of foreign multinational enterprises, up 1.9 percent from 8.50 million in 2022 — is the one that will make the wires.
The more interesting number sits in the second paragraph of the same release. Foreign affiliates accounted for 6.2 percent of total private-industry employment in 2023 — measured against BLS data on total nonfarm private payrolls — down from 6.3 percent in 2022. That is a tenth of a percentage point, which sounds like nothing. It is not nothing. It means domestic employers added jobs faster than foreign-owned employers did, even in a year when foreign-owned headcount grew at a respectable clip. The composition of the U.S. private labor force tilted, marginally, toward American-headquartered firms.
What the BEA actually counts
The BEA dataset is a specific cut of the FDI universe. It tracks “majority-owned U.S. affiliates” — operations where a foreign parent owns more than 50 percent of the voting equity. So Toyota’s Kentucky plants count. Volkswagen’s Tennessee assembly counts. Novartis’s New Jersey research campuses count. A German hedge-fund stake in an S&P 500 name does not, because that is portfolio investment, not direct.
The benchmark is annual, lagged about two years, and considered the authoritative source for inbound foreign direct investment data — methodology details live in BEA’s published methodologies catalog for international economic accounts. The companion dataset — Direct Investment by Country and Industry, also from BEA — pegged the total FDI position in the United States at $5.39 trillion at the end of 2023, a $227 billion increase from 2022. Manufacturing alone accounted for 41.2 percent of that position.
The capex-jobs gap is the real signal
Look at the four headline activity metrics together:
| Metric | 2023 | 2022 (revised) | Year-over-year |
|---|---|---|---|
| Employment (millions) | 8.66 | 8.50 | +1.9% |
| Value added | $1.47T | $1.38T | +6.3% |
| Capex on PP&E | $322.7B | $303.5B | +6.3% |
| R&D spending | $87.8B | $86.8B | +1.1% |
Capital expenditure rose more than three times faster than employment. Value added — the BEA’s measure of direct contribution to U.S. GDP — grew at the same 6.3 percent clip. That gap is consistent with a continued shift toward capital-intensive operations: semiconductors, pharmaceutical manufacturing, automotive electrification, data center build-out. Foreign multinationals are putting more money into American plant and equipment per worker than they were a year earlier, which lines up with the broader picture in our take on what the manufacturing PMI is signaling.
R&D growth of 1.1 percent is the laggard. After several years of foreign R&D in the U.S. growing faster than the broader economy, 2023 was a flat year for innovation spend by foreign-owned operations. That deserves a footnote, not a headline. Some of it is base effects — the 2022 number was revised upward from $80.3 billion to $86.8 billion, which made the 2023 comparison harder. Some of it may reflect a one-year pause as European and Asian parent companies rethought where to locate next-generation R&D in light of U.S. tax policy, semiconductor export controls administered by the Bureau of Industry and Security, and the aftermath of the IEEPA tariff regime that the Supreme Court partially upheld this year.
Where the money comes from
The companion BEA release on FDI by country shows the geography of foreign capital in the U.S. By foreign parent at year-end 2023, the Netherlands led with a $717.5 billion position, followed by Japan at $688.1 billion, Canada at $671.6 billion, and the United Kingdom at $630.1 billion. By ultimate beneficial owner — which strips out holding-company chains and points to where the real money lives — Japan moved into first place at $783.3 billion, with Canada at $749.6 billion, Germany at $657.8 billion, and the U.K. at $635.6 billion. The companion benchmark on outbound activity by U.S. parents, BEA’s release on activities of U.S. multinational enterprises in 2023, provides the mirror view for anyone modeling cross-border corporate footprints.
Two implications worth thinking about. First, the U.S. footprint of Japanese, German, and Korean industrial capacity matters more than the headlines about specific companies suggest. The same dynamic that animates a TSMC quarterly readthrough or an ASML capex print shows up in this benchmark — foreign multinational capex on U.S. soil is a distinct funding channel for the domestic capital stock. Second, the share decline from 6.3 percent to 6.2 percent looks small until you compare it to FDI position growth of $227 billion in the same year. Capital is still flowing in. Headcount is just not following at the same rate.
What investors should actually take from this
The release is a slow-moving structural data point, not a tradable catalyst. It does not move markets when it drops, and it should not. But it sits in the background of three live conversations.
The first is the manufacturing renaissance narrative. Foreign affiliate capex of $322.7 billion in a single year, concentrated heavily in manufacturing, is a real number. It is consistent with what the tariff and reshoring story has produced over the past year — capital is locating in the U.S. that previously located elsewhere. The labor-share decline does not contradict that. It just says the capital is more efficient per worker than it used to be.
The second is the productivity question that animates the debate over Fed policy heading into the late-April FOMC meeting. Higher capex per worker is, by definition, productivity-enhancing investment. If foreign affiliates are contributing meaningfully to U.S. productivity gains, that softens the inflation-from-wage-growth concern that has kept the FOMC divided.
The third is the geographic concentration of foreign exposure. A few states — Texas, California, New York, Illinois, North Carolina, South Carolina, Georgia, Michigan — host disproportionate shares of foreign affiliate employment. State-level economic exposure to foreign parent decisions is higher than the 6.2 percent national share suggests, particularly in industrial corridors that depend on a small number of large foreign-owned employers. The clustering has held even as the national share has drifted lower.
How this is likely to evolve
A few things bear watching as the 2024 figures arrive next December. The R&D number could rebase higher if the 2022-to-2023 flatness reflected a one-year transition rather than a structural slowdown. Capex is likely to continue running above employment growth as semiconductor and battery projects funded by the CHIPS Act program at the Commerce Department and the Inflation Reduction Act move from announcement to production. The country mix could shift if the Netherlands position — partly an artifact of holding-company structures — gets re-attributed to ultimate beneficial owners in future revisions.
For a portfolio context, none of this changes allocations on its own. It does refine the lens you read other releases through. Foreign capital is still arriving. It is just arriving in a different form than ten years ago — more capex, less labor — and concentrating more heavily in industries the U.S. has decided are strategic.
The full BEA dataset, including state-level and industry-level breakouts that the news release does not reproduce, lives at the BEA’s foreign affiliate data hub. Anyone modeling state-level economic exposure or industry-level inbound capital flow should start there.
This article is for educational and informational purposes only and does not constitute investment advice, a recommendation, or an offer to buy or sell any security. Ferrante Capital LLC is a Registered Investment Adviser; nothing in this post is tailored to any individual’s financial situation. Data referenced is from the U.S. Bureau of Economic Analysis. Forward-looking statements reflect Ferrante Capital’s current analysis and are subject to change without notice; actual outcomes may differ materially. Past performance is not indicative of future results. Please consult a qualified financial professional before making investment decisions.
