pith. sign in

arxiv: 2605.30435 · v2 · pith:RJ3OTQVZnew · submitted 2026-05-28 · 💰 econ.GN · q-fin.EC

Global Science Sustains U.S. Innovation

Christopher R. Esposito This is my paper

Pith reviewed 2026-06-28 23:36 UTC · model grok-4.3

classification 💰 econ.GN q-fin.EC
keywords knowledge supply chaininternational scientific collaborationinnovation productivitycitation networksNSF-funded researchborder frictionscritical technologiespatent output
0
0 comments X

The pith

Barriers that slow the cross-border flow of scientific ideas reduce U.S. innovation productivity in priority technologies.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper maps the global supply chain of scientific knowledge that feeds U.S. innovation by following citation paths from NSF-funded research through later papers and into patents. It then simulates frictions that block ideas from crossing the U.S. border and measures the resulting changes in chain connectivity, length, and patent output. The same frictions lower productivity inside the technology classes Congress has designated as critical, including semiconductors, quantum science, and AI. A reader would care because the result indicates that policies restricting international scientific exchange can slow domestic technological progress even in areas treated as national priorities.

Core claim

Tracing multi-generational citation paths from NSF-funded research to downstream patents reveals that the U.S. knowledge supply chain extends globally. Simulated barriers that impede the movement of ideas across the U.S. border reduce the chain's connectivity, increase its length, and lower the rate at which upstream research produces downstream patents. These effects hold inside the technology areas identified by U.S. Congress as critical to national priorities.

What carries the argument

Multi-generational citation paths connecting NSF-funded research to downstream patents, used both to trace the international knowledge supply chain and to stress-test it under simulated border frictions.

If this is right

  • Lowering barriers to cross-border scientific knowledge flows would increase chain connectivity and shorten path length.
  • Innovation productivity would rise in semiconductors, quantum science, and AI if international idea movement faces fewer frictions.
  • Policies that restrict foreign researcher collaboration or data access would reduce patent output even in congressionally prioritized fields.
  • The productivity penalty from border frictions would accumulate across successive generations of citations.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same tracing approach could be applied to other U.S. funding agencies to test whether the global dependence pattern is broader than NSF-supported work.
  • If foreign knowledge sources cannot be fully replaced by domestic ones, restrictions on collaboration would create persistent productivity gaps.
  • Quantifying the size of the productivity loss could inform cost-benefit calculations for export controls or researcher visa rules.
  • Countries that impose similar frictions on incoming scientific knowledge might experience comparable drags on their own innovation rates.

Load-bearing premise

That multi-generational citation paths from NSF-funded research to patents accurately trace and represent the international supply chain of scientific knowledge powering U.S. innovation.

What would settle it

Empirical evidence that U.S. patents in the studied fields draw their key inputs through citation paths that bypass or differ markedly from the traced NSF chains would show the mapping does not capture the actual supply chain.

read the original abstract

Like physical products, new technologies are developed using globally sourced inputs. Yet while the supply chains behind physical goods are well understood, we know far less about the international supply chain of scientific knowledge that powers U.S. innovation, or how vulnerable it may be to disruption. Here, I uncover this supply chain by tracing multi-generational citation paths connecting NSF-funded research to downstream patents, and stress-test it by simulating barriers to scientific knowledge flows across the U.S. border. The U.S. knowledge supply chain extends globally, and frictions impeding the movement of ideas across the U.S. border reduce its connectivity, extend its length, and lower innovation productivity. These impacts extend to technology areas deemed critical to national priorities by U.S. Congress, including Semiconductors, Quantum Science, and AI.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper claims to map the international supply chain of scientific knowledge powering U.S. innovation by tracing multi-generational citation paths from NSF-funded research to downstream patents, then stress-tests the chain by simulating barriers to cross-border knowledge flows; it reports that such frictions reduce connectivity, extend path length, and lower innovation productivity, with effects extending to congressionally prioritized areas including Semiconductors, Quantum Science, and AI.

Significance. If the citation-tracing method is shown to faithfully capture cross-border knowledge inputs whose disruption affects productivity, the results would supply quantitative evidence on the global dependencies of U.S. innovation and the costs of policy-induced frictions, with direct relevance to national-security and R&D policy debates.

major comments (2)
  1. [Abstract and §3] Abstract and §3 (tracing method): the central claim that multi-generational citation paths 'accurately trace and represent the international supply chain of scientific knowledge' is load-bearing yet unsupported by any described validation against independent transfer measures (e.g., licensing data, inventor mobility, or expert-mapped diffusion); citation networks are known to be shaped by strategic, linguistic, and visibility biases that are amplified internationally, and multi-generational chaining compounds measurement error.
  2. [Abstract and simulation section] Abstract and simulation section: no quantitative results, error bars, sample sizes, data sources, or robustness checks are supplied, so it is impossible to evaluate whether the reported reductions in connectivity and productivity are distinguishable from noise or from the mechanical removal of links in a noisy network.
minor comments (1)
  1. [Abstract] The abstract states findings as outputs of an empirical procedure but supplies none of the usual summary statistics or validation steps expected in an empirical manuscript; adding a short results paragraph with key magnitudes would improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which identify key areas where the manuscript can be strengthened. We respond to each major comment below.

read point-by-point responses

  1. Referee: [Abstract and §3] Abstract and §3 (tracing method): the central claim that multi-generational citation paths 'accurately trace and represent the international supply chain of scientific knowledge' is load-bearing yet unsupported by any described validation against independent transfer measures (e.g., licensing data, inventor mobility, or expert-mapped diffusion); citation networks are known to be shaped by strategic, linguistic, and visibility biases that are amplified internationally, and multi-generational chaining compounds measurement error.

    Authors: We agree that the central claim would be strengthened by explicit validation against independent measures. The manuscript relies on citation paths as a standard proxy for knowledge diffusion, consistent with prior work in scientometrics, but does not provide the requested cross-validation. In revision we will add a limitations subsection that explicitly discusses known citation biases (strategic, linguistic, visibility) and the risk of compounded error in multi-generational chains. We will also report any feasible supplementary checks using publicly available inventor-mobility data. revision: yes

  2. Referee: [Abstract and simulation section] Abstract and simulation section: no quantitative results, error bars, sample sizes, data sources, or robustness checks are supplied, so it is impossible to evaluate whether the reported reductions in connectivity and productivity are distinguishable from noise or from the mechanical removal of links in a noisy network.

    Authors: The referee correctly notes that neither the abstract nor the simulation section currently supplies quantitative results, error bars, sample sizes, data sources, or robustness checks. Although the underlying analysis contains these elements, they were not reported in sufficient detail. We will revise both the abstract and the simulation section to include the key quantitative findings, sample sizes, effect sizes with error bars or confidence intervals, explicit data sources, and summaries of robustness checks so that readers can assess whether the reported reductions exceed what would be expected from link removal in a noisy network. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical tracing method is independent of reported outcomes

full rationale

The paper's core procedure traces multi-generational citation paths from NSF-funded papers to downstream patents and then simulates border frictions on that observed network. This is a data-driven measurement and counterfactual exercise whose outputs (connectivity, length, productivity effects) are not defined into the inputs or recovered by fitting parameters to the same quantities. No equations, self-citations, or ansatzes are shown that would make any prediction equivalent to its own construction. The method therefore remains self-contained against external citation data.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that citation paths serve as a valid proxy for knowledge supply chains; no free parameters, new entities, or additional axioms are identifiable from the abstract alone.

axioms (1)
  • domain assumption Multi-generational citation paths from NSF-funded research to downstream patents represent the international supply chain of scientific knowledge
    This premise is invoked to justify the tracing method that uncovers the global supply chain and enables the friction simulations.

pith-pipeline@v0.9.1-grok · 5652 in / 1404 out tokens · 28059 ms · 2026-06-28T23:36:44.599517+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

52 extracted references · 9 canonical work pages

  1. [1]

    Altschul, W

    S. Altschul, W. Gish, E. Myers, D. Lipman, Basic Local Alignment Search Tool.Journal of Molecular Biology215(3), 403–410 (1990)

    1990

  2. [2]

    Tatusov, E

    R. Tatusov, E. Koonin, D. Lipman, A Genomic Perspective on Protein Families.Science 278(5338), 631–637 (1997)

    1997

  3. [3]

    Benson,et al., GenBank.Nucleic Acids Research41(D1), D36–D42 (2013)

    D. Benson,et al., GenBank.Nucleic Acids Research41(D1), D36–D42 (2013)

    2013

  4. [4]

    Mojica, C

    F. Mojica, C. Diez-Villasenor, E. Soria, G. Juez, Biological Significance of a Family of Reg- ularly Spaced Repeats in the Genomes of Archaea, Bacteria and Mitochondria.Molecular Microbiology36(1), 244–246 (2000)

    2000

  5. [5]

    Jansen, J

    R. Jansen, J. van Embden, W. Gaastra, L. Schouls, Identification of Genes that are Assoicated with DNA Repeats in Prokaryotes.Molecular Microbiology43(6), 1565–1575 (2002)

    2002

  6. [6]

    Mojica, C

    F. Mojica, C. Diez-Villasenor, J. Garcia-Martinez, E. Soria, Intervening Sequences of Regularly Spaced Prokaryotic Repeats Derive from Foreign Genetic Elements.Journal of Molecular Evolution60, 174–182 (2005)

    2005

  7. [7]

    Barrangou,et al., CRISPR provides acquired resistance against viruses in prokaryotes

    R. Barrangou,et al., CRISPR provides acquired resistance against viruses in prokaryotes. Science315(5819), 1709–1712 (2007), doi:10.1126/science.1138140

    work page doi:10.1126/science.1138140 2007

  8. [8]

    Zhang, CRISPR-Cas Systems and Methods for Altering Expression of Gene Products (2014)

    F. Zhang, CRISPR-Cas Systems and Methods for Altering Expression of Gene Products (2014)

    2014

  9. [9]

    Doudna, M

    J. Doudna, M. Jinek, C. Emmanuelle, K. Chylinski, Methods and Compositions for RNA- Directed Target DNA Modification and for RNA-Directed Modulation of Transcription (2019)

    2019

  10. [10]

    P. M. Romer, Endogenous Technological Change.Journal of Political Economy98(2), S71– S102 (1990)

    1990

  11. [11]

    Mokyr,The Lever of Riches: Technological Creativity and Economic Progress(Oxford University Press, New York) (1990)

    J. Mokyr,The Lever of Riches: Technological Creativity and Economic Progress(Oxford University Press, New York) (1990)

    1990

  12. [12]

    Weitzman, Recombinant Growth.The Quarterly Journal of Economics113(2) (1998)

    M. Weitzman, Recombinant Growth.The Quarterly Journal of Economics113(2) (1998). 13

    1998

  13. [13]

    Fleming, O

    L. Fleming, O. Sorenson,Technology as a complex adaptive system: evidence from patent data, Tech. rep. (2001), publication Title: Research Policy Volume: 30

    2001

  14. [14]

    B. F. Jones, The burden of knowledge and the ”death of the renaissance man”: Is innovation getting harder?Review of Economic Studies76(1) (2009), doi:10.1111/j.1467-937X.2008. 00531.x

    work page doi:10.1111/j.1467-937x.2008 2009

  15. [15]

    Wagner, L

    C. Wagner, L. Leydesdorff, Network structure, self-organization, and the growth of international collaboration in science.Research Policy34(10), 1608–1618 (2005)

    2005

  16. [16]

    J. P. Nagar, S. Breschi, A. Fosfuri, ERC Science and Invention: Does ERC Break Free from the EU Paradox?Research Policy53(8), 105038 (2024), doi:10.1016/j.respol.2024.105038, https://www.sciencedirect.com/science/article/pii/S0048733324000878

    work page doi:10.1016/j.respol.2024.105038 2024

  17. [17]

    J. Wang, X. Li, Y. Wang, The Contribution of Chinese Science to US Technological Advance- ment: Evidence from Patent Citation of Academic Papers.SSRN Working Paper #4802379 (2024)

    2024

  18. [18]

    R. Wu, C. Esposito, J. Evans, China’s Rising Leadership in Global Science.arXiv: 2406.05917 (2024)

    work page arXiv 2024

  19. [19]

    Chatterji, F

    A. Chatterji, F. Murray, How Geopolitics Is Changing the Economics of Innovation, inEn- trepreneurship and Innovation Policy and the Economy(University of Chicago Press), vol. 5 ofNBER Book Series(2025)

    2025

  20. [20]

    J. Lind, D. G. Press, Strategies of Prioritization: American Foreign Policy After Pri- macy.Foreign Affairs(2025),https://www.foreignaffairs.com/united-states/ strategies-prioritization-lind-press

    2025

  21. [21]

    D. of Homeland Security, Harvard University Loses Student and Exchange Visitor Program Certification for Pro-Terrorist Conduct (2025),https://www.dhs.gov/news/2025/05/22/ harvard-university-loses-student-and-exchange-visitor-program-certification-pro

    2025

  22. [22]

    National Institutes of Health, Updated NIH Policy on Foreign Subawards, NIH Guide Notice NOT-OD-25-104 (2025),https://grants.nih.gov/grants/guide/notice-files/ NOT-OD-25-104.html, accessed April 2026. 14

    2025

  23. [23]

    European Commission, Choose Europe for Science: EU Announces =C500 Mil- lion Package to Attract Researchers, Speech by President von der Leyen, Sor- bonne University, Paris (2025),https://www.science.org/content/article/ europe-pledges-600-million-lure-foreign-researchers-vows-protect-scientific-freedom, accessed April 2026

    2025

  24. [24]

    Canadian Association of University Teachers, Advisory on Travel to the United States, CAUT Advisory Statement (2025),https://www.timeshighereducation.com/news/ risk-us-isolationism-canadian-academics-urged-not-travel, accessed April 2026

    2025

  25. [25]

    Polanyi,The Tacit Dimension(University of Chicago Press) (1966)

    M. Polanyi,The Tacit Dimension(University of Chicago Press) (1966)

    1966

  26. [26]

    A. B. Jaffe, M. Trajtenberg, R. Henderson, Geographic Localization of Knowledge Spillovers as Evidenced by Patent Citations.Quarterly Journal of Economics108(3), 577–598 (1993), doi:10.2307/2118401

    work page doi:10.2307/2118401 1993

  27. [27]

    Balsmeier, S

    B. Balsmeier, S. L¨ uck, L. Fleming, Science Knowledge Localizes.Research Policy54(10), 105333 (2025), doi:10.1016/j.respol.2025.105333,https://www.sciencedirect.com/ science/article/pii/S0048733325001623

    work page doi:10.1016/j.respol.2025.105333 2025

  28. [28]

    Balsmeier, L

    B. Balsmeier, L. Fleming, S. L¨ uck, Isolating Personal Knowledge Spillovers: Co-inventor Deaths and Spatial Citation Differentials.American Economic Review: Insights(2023)

    2023

  29. [29]

    Y. Lin, C. B. Frey, L. Wu, Remote Collaboration Fuses Fewer Breakthrough Ideas.Nature623, 987–991 (2023)

    2023

  30. [30]

    Ahmadpoor, B

    M. Ahmadpoor, B. Jones, The Dual Frontier: Patented Inventions and Prior Scientific Advance. Science357(6351), 583–587 (2017)

    2017

  31. [31]

    Samila, O

    S. Samila, O. Sorenson, Venture Capital as a Catalyst to Commercialization.Research Policy 39(10), 1348–1360 (2010)

    2010

  32. [32]

    Samila, O

    S. Samila, O. Sorenson, Venture Capital, Entrepreneurship, and Economic Growth.Review of Economics and Statistics93(1), 338–349 (2011). 15

    2011

  33. [33]

    M. Marx, A. Fuegi, Reliance on Science: Worldwide Front-Page Patent Citations to Scientific Articles.202041(9), 1572–1594 (2020)

    2020

  34. [34]

    Z. Lin, Y. Yin, L. Liu, D. Wang, SciSciNet: A Large-Scale Open Data Lake for the Science of Science Research.Scientific Data10(215) (2023)

    2023

  35. [35]

    Trajtenberg, Innovation in Israel 1968–1997: A comparative analysis using patent data

    M. Trajtenberg, Innovation in Israel 1968–1997: A comparative analysis using patent data. Research Policy30(3), 363–389 (2001)

    1968

  36. [36]

    117th United States Congress, CHIPS and Science Act of 2022, Public Law 117-167 (2022), https://www.congress.gov/bill/117th-congress/house-bill/4346

    2022

  37. [37]

    Arora, S

    A. Arora, S. Belenzon, L. Cioaca, L. Sheer, D. Shvadron, DISCERN 2.0: Extending and Enhancing the DISCERN Dataset.Working paper(2023)

    2023

  38. [38]

    Fleming, Recombinant uncertainty in technological search.Manag

    L. Fleming, Recombinant uncertainty in technological search.Manag. Sci.47(1), 117–132 (2001), publisher: INFORMS Inst.for Operations Res.and the Management Sciences, doi: 10.1287/mnsc.47.1.117.10671

    work page doi:10.1287/mnsc.47.1.117.10671 2001

  39. [39]

    Teece, Profiting from technological innovation: Implications for integration, collaboration, licensing and public policy.Research Policy15(6), 285–305 (1986)

    D. Teece, Profiting from technological innovation: Implications for integration, collaboration, licensing and public policy.Research Policy15(6), 285–305 (1986)

    1986

  40. [40]

    Furman, M

    J. Furman, M. Porter, S. Stern, The determinants of national innovative capacity.Research Policy31(6), 899–933 (2002)

    2002

  41. [41]

    Azoulay, J

    P. Azoulay, J. Krieger, A. Nagaraj, Old Moats for New Models: Openness, Control, and Competition in Generative Artificial Intelligence.Entrepreneurship and Innovation Policy and the Economy5(2025)

    2025

  42. [42]

    Bikard, M

    M. Bikard, M. Marx, Bridging Academia and Industry: How Geographic Hubs Connect Uni- versity Science and Corporate Technology.Management Science66(8), 3295–3798 (2020)

    2020

  43. [43]

    Storper, T

    M. Storper, T. Kemeny, N. Makarem, T. Osman,The Rise and Fall of Urban Economies: Lessons from San Francisco and Los Angeles(Stanford University Press) (2015)

    2015

  44. [44]

    Miller,Chip War: The Fight for the World’s Most Critical Technology(Simon and Schuster) (2022)

    C. Miller,Chip War: The Fight for the World’s Most Critical Technology(Simon and Schuster) (2022). 16

    2022

  45. [45]

    Storper, A

    M. Storper, A. Venables, Buzz: Face-To-Face Contact and the Urban Economy.Journal of Economic Geography4(4), 351–370 (2004)

    2004

  46. [46]

    Moser, A

    P. Moser, A. Voena, F. Waldinger, German-Jewish Emigres and U.S. Invention.American Economic Review104(10), 3222–3255 (2014)

    2014

  47. [47]

    Teodoridis, M

    F. Teodoridis, M. Bikard, K. Vakili, Creativity at the Knowledge Frontier: The Impact of Specialization in Fast- and Slow-paced Domains*.Administrative Science Quarterly64(4), 894–927 (2019), publisher: SAGE Publications Ltd, doi:10.1177/0001839218793384

    work page doi:10.1177/0001839218793384 2019

  48. [48]

    Teplitskiy, S

    M. Teplitskiy, S. Park, N. Thompson, D. Karger, Intentional and Serendipitous Diffusion of Ideas: Evidence from Academic Conferences (arXiv:2209.01175) (2024)

    work page arXiv 2024

  49. [49]

    Naddaf, Scientific Conferences are Leaving the U.S

    M. Naddaf, Scientific Conferences are Leaving the U.S. amid Border Fears.Nature(2025), https://www.nature.com/articles/d41586-025-01636-5

    2025

  50. [50]

    D. Shi, W. Liu, Y. Wang, Has China’s Young Thousand Talents Program been Successful in Recruiting and Nurturing Top-Caliber Scientists?Science379(6627), 62–65 (2023)

    2023

  51. [51]

    Owens, China’s research clout leads to growth in homegrown science publishing.Nature (2024),https://www.nature.com/articles/d41586-024-01596-2

    B. Owens, China’s research clout leads to growth in homegrown science publishing.Nature (2024),https://www.nature.com/articles/d41586-024-01596-2

    2024

  52. [52]

    start points

    M. Djuric, Canada’s tariff list targets Republican states, Trump allies.Politico(2025),https: //www.politico.com/news/2025/02/03/canada-tariff-targets-00202005. Acknowledgments I thank Olav Sorenson, James Evans, Alan Marco, Megan MacGarvie, and Alexander Furnas for helpful comments, and participants at the Columbia Management, Analytics, and Data confere...

    2025