From brain drain to brain bridges: How mobile PhD scientists connect home countries to global science

For middle-income countries committed to catching up with the scientific and technological frontier, connecting to global knowledge is fundamental (Barnard et al. 2015, Wagner 2018). A common strategy to build this connection is to fund promising students to pursue PhDs at top universities abroad, with the expectation that they will return with a set of new skills and international networks (Dahlman 2010, Fry and Furman 2023, Kim 1998, Müller et al. 2023).

However, this costly strategy creates a persistent dilemma: when these mobile PhDs choose not to return despite the conditionalities of the funding schemes, are they a national loss – a ‘brain drain’ – or can they still bring benefits to their home countries from abroad? A growing body of research suggests that diaspora networks can facilitate knowledge flows, investment, and innovation, which instead fosters a ‘brain bank’ or ‘brain circulation’ (Agrawal et al. 2011, Saxenian 2005, Batista et al. 2025). This evidence shows the benefits that mobile scientists bring to their own careers and the direct collaborations they foster.

Yet crucial questions remain unanswered. Do these international networks extend to colleagues who never leave the home country? And, if they do, what is the nature of these new connections? Are mobile scientists acting as matchmakers who create self-sustaining links, or are they short-lived bridges whose continued presence is required to keep the international scientific connections alive?

A unique window into Colombia’s scientific networks

In our research (Ito et al. 2025), we investigate these questions by tracing the careers of over 19,000 Colombian scientists. We find that internationally mobile PhDs are indeed vital conduits connecting their non-mobile peers to foreign scientists. But we uncover a crucial catch: these connections are not independent. They are largely sustained by the mobile scientists’ continued role as mediators, which turn them from simple ‘introducers’ into essential, ongoing collaborators.

Colombia provides an ideal setting for this research. The country has invested heavily in internationalising its scientific community, including sponsoring thousands of students for doctoral training abroad. Moreover, the government established a detailed academic CV platform (CvLAC), which offers unusually detailed information on researchers’ careers.

By combining data from this CV platform, PhD scholarship lists from the main funding agencies in Colombia, and a global publication database (OpenAlex), we reconstruct the mobility patterns of scientists from 1990 to 2021. This allows us to distinguish between non-mobile scientists – researchers who completed all their studies in Colombia – and mobile scientists – Colombian researchers who obtained a PhD abroad. We further divide the second group into returnees (who came back to Colombia permanently), diaspora (who remained abroad), and intermittent (who move between Colombia and other countries).

Mobile scientists as international connectors 

Our analysis shows a clear and robust pattern: collaborating with mobile scientists significantly increases a non-mobile scientist’s connections to foreign scientists. As a non-mobile scientist, having a mobile collaborator is associated with a 12 percentage point increase in the likelihood of co-authoring with a foreign scientist.

This effect is not driven primarily by returnees. Instead, our findings indicate that diaspora and intermittent scientists are the more powerful mediators than returnees. Collaborating with a member of the diaspora or an intermittent scientist is associated with a larger increase in a non-mobile scientist’s proportion and number of publications with foreign co-authors. This is likely because those (mobile scientists) who remain connected to global research hubs maintain larger, more active international networks.

A crucial catch: These ‘bridges’ must be maintained

This discovery that mobile scientists build bridges is encouraging, but are these bridges self-supporting once built? To find out, we tracked non-mobile scientists over time, comparing those who started collaborating with at least one mobile scientist to those who did not.

As shown in Figure 1, there is a clear jump in foreign collaborations right after a non-mobile researcher begins working with a mobile one. The more mobile co-authors a non-mobile scientist works with (from Panels 1 to 4), the stronger this effect becomes, suggesting a dose-response relationship.

However, the story changes dramatically when we dig deeper. We repeated the analysis, this time excluding any three-way publications where the authorial team included the non-mobile scientist, the mobile bridge, and a foreign scientist. In other words, we looked for evidence of new, direct, two-way relationships forming between the non-mobile and foreign scientists.

The results were stark: the effect vanished (Figure 2). The positive effects of collaborating with a mobile scientist largely disappear, and new international collaborations are almost entirely dependent on the continued presence of the mobile scientist in the research team. They are not making introductions that develop into independent partnerships; they are acting as crucial, ongoing nodes in a three-way collaboration.

Figure 1: Number of publications with foreign co-authors per year (in logs)

Note: This figure summarises the event study results from Ito et al. (2025). The vertical axis shows the change in the number of foreign co-authored papers for a non-mobile scientist after their first collaboration with a mobile scientist at year 0 (Panel 1). The effect grows and is sustained as they collaborate with more mobile scientists (moving from Panel 1 to 4).

Figure 2: Number of publications with foreign co-authors per year, excluding publications with mobile scientists (in logs)

Note: This figure summarises the results when three-way co-authorships (non-mobile, mobile, and foreign) are excluded. The vertical axis shows the change in the number of foreign co-authored papers for a non-mobile scientist after their first collaboration with a mobile scientist at year 0 (Panel 1). The effect seen in Figure 1 disappears, indicating that the mobile scientist is an essential, ongoing mediator, not just an introducer.

Rethinking ‘brain drain’: Policy implications

Our findings illustrate how mobile scientists contribute to non-mobile scientists acquiring foreign connections and, more importantly, the nature and durability of these connections. These results have important policy implications for countries looking to leverage international mobility for scientific development:

• Sponsoring mobility generates benefits for non-mobile scientists. Policies that support students and scientists studying abroad are not just an investment in individuals – they are an investment in the entire national scientific system, creating tangible benefits for researchers who remain at home.
• Rethink strict return requirements. The conventional wisdom that a scientist abroad is a resource lost may be flawed. Our findings show that diaspora and intermittent scientists are at least as effective bridges to global knowledge. Policies that impose strict conditions for returning may be counterproductive as they can sever valuable networks. A more flexible approach that encourages continued engagement from abroad could yield greater returns.
• Actively maintain the bridges. Since foreign connections are not self-sustaining, policy should aim to foster and maintain the mediating role of mobile scientists. This could include funding for short-term return visits, grants for joint projects that explicitly link diaspora members with home-based labs, and virtual platforms to facilitate ongoing collaborations. Simply making an introduction is not enough; the bridge itself needs to be supported. 
  
  Ultimately, our research suggests moving beyond the simple ‘drain’ versus ‘gain’ dichotomy. A country’s internationally mobile talent, whether at home or abroad, constitutes a dynamic network from which the country can gain. The key for policymakers is not just to build this network, but to actively sustain the connections that make it so valuable.

References
Agrawal, A, D Kapur, J McHale, and A Oettl (2011), “Brain drain or brain bank? The impact of skilled emigration on poor-country innovation,” Journal of Urban Economics 69: 43–55.

Barnard, H, R Cowan, M F de Arroyabe Arranz, and M Müller (2015), “The role of global connectedness in the development of indigenous science in middle-income countries,” in The Handbook of Global Science, Technology, and Innovation, Wiley, 382–406.

Batista, C, D Han, J Haushofer, G Khanna, D McKenzie, A M Mobarak, C Theoharides, and D Yang (2025), “Brain drain or brain gain? Effects of high-skilled international emigration on origin countries,” Science 388(6749): eadr8861.

Dahlman, C (2010), “Innovation strategies in Brazil, China, and India: From imitation to deepening technological capability in the South,” in X Fu and L Soete (eds.), The Rise of Technological Power in the South, Palgrave Macmillan, London, 15–48.

Fry, C, and J L Furman (2023), “Migration and global network formation: Evidence from female scientists in developing countries,” Organization Science.

Ito, R, D Chavarro, T Ciarli, R Cowan, and F Visentin (2025), “From drains to bridges: The role of internationally mobile PhD students in linking non-mobile with foreign scientists,” Journal of Development Economics 177: 103577.

Kim, J (1998), “Economic analysis of foreign education and students abroad,” Journal of Development Economics 56: 337–365.

Müller, M, R Cowan, and H Barnard (2023), “The role of local colleagues in establishing international scientific collaboration: Social capital in emerging science systems,” Industrial and Corporate Change 32: 1077–1108.

Saxenian, A (2005), “From brain drain to brain circulation: Transnational communities and regional upgrading in India and China,” Studies in Comparative International Development 40: 35–61.

Wagner, C S (2018), “The collaborative era in science,” Springer International Publishing.

Reprinted by permission from VoxDev. Authors: Rodrigo Ito (UNU-MERIT), Diego Chavarro, Tommaso Ciarli (UNU-MERIT), Robin Cowan (University of Strasbourg & Maastricht University), and Fabiana Visentin (UNU-MERIT & Maastricht University).