Technological breakthroughs rarely stay where they start. From AI and cloud computing to 3D printing and autonomous vehicles, new technologies diffuse rapidly, but unevenly, across firms, regions and countries. Some quickly adopt the latest tools; others lag for years. What explains this pattern? Prior literature on technology diffusion points to geography and income per capita as key barriers (Comin and Hobijn 2004, 2010, Comin and Mestieri 2018). Multinational firms and firm-to-firm relationships through global value chains are believed to play an important role in facilitating international technology diffusion (World Bank 2020). However, understanding the role of firm-to-firm relationships in the cross-country spread of new technologies has proved difficult due to the lack of detailed panel data on both the adoption of new technologies across countries and firms, and on different types of international firm-to-firm linkages.

Using online job postings to measure the international diffusion of new technologies

In a recent paper (Bastos et al. 2026), we provide new evidence on how technologies spread globally by tracing the diffusion of 29 frontier technologies across 17 advanced economies between 2014 and 2022. We draw on more than 500 million online job postings linked to detailed firm-to-firm networks data. Following the approach pioneered by Bloom et al. (2021), we identify technology-related hiring through granular keywords in job adverts, using this as a proxy for technology diffusion (similar to Acemoglu et al. 2022 and Goldfarb et al. 2023). We then link these postings to global business relationships – buyers, suppliers, affiliates, and innovation partners – to observe whether and how these networks help transmitting technologies across borders. Across countries, technology-related job postings rise systematically with income levels (Figure 1). Across firms, they are heavily concentrated within Fortune 500 firms and their supply chains (Figure 2).

Figure 1 Income and technology job posting share in 2022

Figure 2 Technology job postings in multinationals and their supply chains

Several patterns emerge from our data. First, we find that as technologies mature, technology-related hiring becomes steadily more dispersed across countries and regions, extending the US-only evidence in Bloom et al. (2021) to a global scale. Second, diffusion is unequal and runs through a few large firms and their business networks. Roughly one-third of all postings for these frontier technologies originate from Fortune 500 companies and their immediate networks (affiliates, buyers, suppliers, and innovation partners). Within this group, pioneer firms maintain a lasting advantage: although their hiring premium declines over time, it remains more than double that of other firms even after seven years.

The role of firm-to-firm linkages in technology diffusion

A central contribution of our study is to examine the role played by firm-to-firm linkages in the diffusion of new technologies. We measure ex ante relationships – links in place before a technology’s emergence – and ask whether connections with the technology pioneers predict ex post diffusion, captured by subsequent growth in technology-related hiring. This approach allows us to see which types of ties actually transmit new technologies across firms and borders.

We find firms that purchased from pioneers before a technology’s take-off expand technology-related hiring about one and a half times faster than otherwise similar firms. Buyers often face stringent standards, data exchange, and technical integration with pioneers – mechanisms long associated with faster learning and adoption in the learning-by-importing and reverse-engineering traditions. Co-development or R&D collaborations with pioneer firms show a similar pattern: firms engaged in innovation partnerships adopt new technologies more quickly. Distance and income still play a role – diffusion is faster to richer and geographically closer places, but the buyer and innovation-partner effects remain robust even after accounting for these fundamentals. By contrast, prior upstream supplier links and common ownership show little independent role once buyer and partnership ties are taken into account. Together, this evidence suggests that the world’s largest firms and their networks are key conduits for technology diffusion. Yet some linkages matter more than others. Relationships rooted in buyer interactions and innovation partnerships appear to move technologies faster.

Conclusion

This column documents how new technologies diffuse internationally through firm-to-firm networks that connect technology pioneers to other producers. Using a rich combination of online job postings and global firm relationship data, we show that the diffusion of new technologies is highly uneven and disproportionately mediated by a relatively small set of large firms and their networks. Geography and income still matter for technology diffusion, but they do not tell the whole story of how new technologies travel.

An important finding is that not all firm links are equally important for the transmission of new technologies. Buyer relationships and innovation partnerships with technology pioneers play a particularly important role, accelerating technology adoption even after accounting for distance and income differences. These links likely embody intensive interactions – data sharing, co-development, technical integration – that facilitate learning in ways that other relationships do not. By contrast, upstream supplier ties and common ownership structures appear to be less relevant channels of technology diffusion.

These findings help reconcile two facts about global technology diffusion: its remarkable reach and its persistent inequality. While frontier technologies eventually spread widely, the path they follow is shaped by pre-existing business relationships that favour firms already embedded in global production and innovation networks. As a result, firms and countries that are weakly connected to technology pioneers risk falling behind. For policymakers, the results suggest that fostering technology adoption requires more than improving domestic capabilities or reducing trade costs. Policies that lower barriers to participation in global value chains, encourage collaboration with leading firms, and support cross-border innovation partnerships can strengthen the channels through which new technologies spread. In a world where growth increasingly depends on the rapid absorption of new technologies, understanding – and shaping – the networks that carry them is crucial.

Authors’ note: The views expressed are entirely those of the authors and not necessarily those of the institutions with which the authors are affiliated.

References

Acemoglu, D, D Autor, J Hazell and P Restrepo (2022), “Artificial Intelligence and Jobs: Evidence from Online Vacancies”, Journal of Labor Economics 40(S1): S293–S340.

Bastos, P, K Stapleton, D Taglioni and H Y Wei (2026), “Firm Networks and Global Technology Diffusion”, Journal of International Economics.

Bloom, N, T A Hassan, A Kalyani, J Lerner and A Tahoun (2021), “The Diffusion of New Technologies”, NBER Working Paper No. 28999.

Comin, D and B Hobijn (2004), “Cross-country Technology Adoption: Making the Theories Face the Facts”, Journal of Monetary Economics 51: 39–83.

Comin, D and B Hobijn (2010), “An Exploration of Technology Diffusion”, American Economic Review 100: 2031–2059.

Comin, D and M Mestieri (2018), “If Technology Has Arrived Everywhere, Why Has Income Diverged?”, American Economic Journal: Macroeconomics 10(3): 137–178.

Goldfarb, A, B Taska and F Teodoridis (2023), “Could Machine Learning Be a General Purpose Technology? A Comparison of Emerging Technologies Using Data from Online Job Postings”, Research Policy 52(1): 104653.

World Bank (2020), Trading for development in the age of global value chains, World Development Report 2020.