Tracing the Genealogy of American Innovation: How Foreign Ideas Shape U.S. Patents and Labor Markets

The United States has long cultivated an image of domestic ingenuity, celebrating home-grown inventors who pioneered breakthroughs from the light bulb to the semiconductor. However, a comprehensive new study released in October 2026 under Working Paper 35900 (DOI: 10.3386/w35900) challenges this insular narrative. By examining the full text of 13 million patents filed between 1887 and 1889 and stretching through 2024, researchers have constructed a massive, directed genealogy of technologies. This input-output network of technological recombination reveals that American innovation is deeply and fundamentally reliant on foreign ideas.

The research paper, which maps out how earlier technologies appear in the founding patents of new ones, reports four primary findings that fundamentally reshape our understanding of technological progress. Foreign ideas are not marginal contributors to the American landscape; they are pervasive structural inputs. More than half of all U.S. technologies—specifically 54 percent—build directly on at least one foreign input. Furthermore, the correlation between global input and domestic success is striking: the most fertile and prolific U.S. technologies rely most heavily on these international knowledge transfers.

Beyond mere correlation, the study establishes a powerful causal relationship between foreign ideas and domestic inventive output. Utilizing the historical settlement patterns of immigrants as a source of exogenous variation in local ancestry, the authors demonstrated that U.S. counties with higher exposure to knowledge from specific foreign nations absorb significantly more technologies from those countries. Consequently, these regions invent substantially more U.S. technologies that directly build upon those imported foundations.

The research also underscores how foreign intellectual capital directs the geographic and thematic trajectory of technical change. American counties are vastly more likely to pioneer specific technologies if the relevant foreign inputs are already present in their local "idea stock." Moreover, the shifting origins of these foreign ideas mirror the broader geographic migration of American innovation itself, tracking the evolution of the nation’s inventive hubs from its historical industrial heartland toward the coastal technology corridors. Finally, these early exposures leave a permanent imprint on local economies. Decades down the line, counties that successfully absorbed an origin’s ideas host a greater concentration of jobs working directly with those concepts, alongside an even larger workforce dedicated to the domestic U.S. technologies descended from them.

A Century and a Half of Data: The Historical Evolution of Patent Genealogies

To map out this intricate web of global innovation, the authors of Working Paper 35900 undertook a monumental empirical task. Rather than relying on superficial citation metrics or self-reported patent categories, the research team analyzed the complete textual corpus of 13 million patents filed over a 137-year span, starting from the late 19th century and running through the mid-2020s.

The historical timeline covered by the study captures the radical transformation of global industrialization. In the late 1800s, during the second Industrial Revolution, American inventors were rapidly scaling mechanized production, drawing heavily upon European metallurgical and chemical advancements originating in the United Kingdom, Germany, and France. As the decades progressed, the genealogy of U.S. patents shifted to incorporate electro-technical designs from Central Europe, followed by mid-century aerospace and computing concepts, and eventually the sophisticated manufacturing and semiconductor innovations originating from East Asia in the late 20th and early 21st centuries.

By tracing the textual lineage of these patents—identifying when a fundamental concept first articulated abroad reappears as a core component of a domestic patent application—the researchers built an unprecedented input-output matrix. This matrix treats ideas much like traditional economic commodities, tracking how raw intellectual materials are imported, processed, recombined, and ultimately exported as finished domestic technologies.

The empirical methodology relies heavily on natural language processing and machine learning tools applied to historical texts. By parsing millions of pages of patent applications, specifications, and claims, the algorithm tracked the linguistic and conceptual evolution of technological building blocks. This established a rigorous, verifiable timeline of how foreign concepts migrated across borders long before the advent of the internet, typically traveling through channels of human migration, international trade, and multinational enterprise.

Empirical Evidence and Statistical Breakdown

The statistical insights unveiled in the working paper provide a rigorous mathematical backbone to the otherwise abstract concept of globalized knowledge diffusion. The headline figure—that 54 percent of U.S. technologies build on at least one foreign input—shatters the myth of domestic technological autarky.

Even more revealing is the intensity of this reliance among high-performing sectors. The study shows that the most fertile, high-impact U.S. technologies are precisely the ones that integrate foreign ideas most aggressively. This indicates that radical innovation rarely happens in a vacuum; rather, it is the product of diverse intellectual recombination, where mixing distinct domestic and international knowledge streams yields exponentially greater scientific and commercial value.

To prove causality rather than mere coincidence, the researchers leveraged historical migrant settlement data as an instrumental variable. Because immigrants historically tended to settle in specific U.S. counties based on historical railroad routes, labor demands, and early diaspora communities, these settlement patterns created long-lasting local cultural and informational ties to specific foreign nations.

The data confirms that these historical diasporas acted as conduits for knowledge transfer. U.S. counties with deep historical roots tied to a particular country were significantly better equipped to decode, absorb, and build upon technological patents originating from that same country generations later. This localized knowledge absorption capacity translated directly into heightened patent productivity. Counties with high exposure to a specific foreign idea stock did not just passively consume foreign research; they actively transformed it into new, proprietary American inventions at rates far exceeding the national average.

Geographic Shifts and Economic Implications

One of the most compelling dimensions of the study is its analysis of how foreign ideas dictate both the level and the direction of technical change. For decades, economic geographers and policymakers have tracked the spatial drift of American innovation, noting the decline of traditional manufacturing and inventive centers in the Midwest Rust Belt and the concurrent explosive growth of coastal technology clusters in Silicon Valley, the Pacific Northwest, and the Boston-New York corridor.

Working Paper 35900 suggests that these geographic shifts were not merely driven by venture capital availability or local tax incentives, but by the evolving geography of foreign idea integration. As the primary sources of global technological leadership shifted over the 20th century—moving from Western European industrial powerhouses to East Asian electronics and software hubs—coastal U.S. counties were often better positioned to absorb these new international knowledge stocks due to immigrant demographics, port access, and global trade networks. Consequently, the regional specialization of American innovation followed the shifting currents of international intellectual inputs.

The implications of this dynamic extend far beyond patent offices and corporate boardrooms, leaving a lasting imprint on local labor markets. The study’s generational analysis reveals that the economic benefits of absorbing foreign ideas persist for decades. Counties that successfully integrated a foreign nation’s technological inputs in the mid-20th century exhibit structurally different labor markets today.

Specifically, these regions host a higher density of jobs directly utilizing those imported concepts. Even more significantly, they support a massive secondary workforce employed in industries and roles tied to the domestic U.S. technologies that descended from those original recombinations. This demonstrates that local economic resilience and job creation are intrinsically linked to a region’s historical openness and capacity to process global knowledge.

Expert Perspectives and Economic Analysis

While Working Paper 35900 is an academic study published through the National Bureau of Economic Research (NBER), its findings carry profound implications for contemporary economic policy, industrial strategy, and immigration debates. Economists and innovation experts analyzing the paper point out several key takeaways for policymakers navigating an increasingly fragmented global trade and technology landscape.

In recent years, geopolitical tensions and rising economic nationalism have prompted various nations to pursue policies of technological decoupling, "reshoring," and strict supply chain localization. Critics of such insular industrial policies argue that data from the NBER paper serves as an empirical warning sign: attempting to seal off a national innovation ecosystem from foreign intellectual inputs may inadvertently starve domestic industries of the vital recombination materials required for breakthrough technological progress.

"Innovation is fundamentally a combinatorial process," notes one labor economist familiar with the study who spoke on condition of anonymity regarding ongoing policy advisories. "If you restrict the inflow of foreign ideas—whether through restrictive high-skilled immigration policies, limitations on international academic collaboration, or the erection of digital and scientific barriers—you are effectively cutting off the raw inputs that power more than half of all advanced U.S. technologies."

Furthermore, regional economic development officials are likely to take note of the study’s findings concerning local labor market persistence. For decades, local governments have competed fiercely to attract corporate headquarters and manufacturing plants through subsidies. However, the NBER research indicates that long-term regional prosperity is deeply rooted in "knowledge absorptive capacity"—the local cultural, educational, and demographic infrastructure that enables a community to understand, translate, and build upon global ideas. Building this capacity requires sustained investments in higher education, open scientific exchange, and welcoming demographic policies that foster diverse intellectual communities.

Conclusion and Future Research Directions

As the global economy continues to navigate the complexities of digital transformation, artificial intelligence, and shifting geopolitical alliances, the findings of Working Paper 35900 offer a timely corrective to overly nationalist views of technological supremacy. By mapping 13 million patents across nearly 140 years, the research provides indisputable proof that American innovation has never been a purely domestic enterprise.

Instead, the historical engine of U.S. prosperity has relied on a vast, interconnected network of technological recombination, where foreign ideas serve as essential catalysts for domestic breakthroughs. As policymakers weigh the future of trade, immigration, and research funding, the message embedded within this extensive patent genealogy is clear: the health, direction, and job-creating power of the American economy remain inextricably bound to its continued engagement with the global marketplace of ideas.