Knowledge moves with people. What role does international mobility play in research and innovation systems? This chapter examines the evidence conveyed through ReICO data and indicators on the role of international talent in shaping national research and innovation capacities. It explores mobility during doctoral and master’s degree programmes, the contribution of foreign-born individuals to the highly qualified workforce and mobility patterns revealed through data on scientific publications.
Research and Innovation Careers Observatory 2026
Key Indicators on Talent, Mobility and Working Conditions
Abstract
International mobility expands talent pools and strengthens knowledge exchange
International mobility for the purpose of enrolling in advanced degree programmes has become an important source of talent for research and innovation systems. Across the OECD, 19% of master's graduates and 29% of doctoral graduates in 2023 were internationally mobile, meaning that they had crossed an international border for the purpose of study. The share of internationally mobile doctoral graduates has increased over time, rising from around 24% in 2015.
However, international mobility remains uneven across countries, fields of study and demographic groups. There are also considerable differences across countries in the share of foreign-born individuals holding advanced qualifications, particularly at doctoral level, where shares range from below 5% to more than 70%. Although countries with higher shares of internationally mobile graduates generally also have larger shares of foreign-born advanced degree holders, the relationship is far from uniform. This suggests that countries rely on different combinations of student mobility, skilled migration and graduate retention to build their talent base.
International mobility also shapes the global networks through which knowledge is created and exchanged. Bibliometric evidence shows that the mobility of scientific authors is closely associated with cross-border scientific collaboration. Around 10% of scientific authors with affiliations in OECD countries in 2024 showed evidence of previous international mobility. Mobile researchers are also disproportionately concentrated in internationally connected research systems.
Together, these patterns suggest that mobility contributes not only to expanding the supply of highly qualified people, but also to strengthening collaborations, knowledge flows and international networks that underpin the performance of research and innovation systems.
Much of what makes research and innovation productive lies within individuals, through their knowledge, their practical skills, and their engagement with others working on similar tasks and challenges (OECD, 2013[1]). The movement of people is therefore one of the principal channels through which knowledge diffuses across research and innovation ecosystems. The R&I workforce is characterised by high levels of mobility, within and between countries, employing or affiliated organisations and often across different economic sectors.
Mobility in R&I careers takes many more forms than moving geographies. These forms can be distinguished by the degree of physical relocation involved, as well as the nature and duration of movement (Figure 4.1). At one end of the spectrum are forms of engagement that do not involve physical relocation, such as cross-border collaboration and co-authorship of scientific publications, participation in international projects and networks, and remote work with organisations in other countries. At the other end are long-term moves involving relocation, including permanent migration, intersectoral mobility and changes of employer. Between these extremes are temporary forms of mobility, such as secondments, short research visits, and degree mobility during higher education and doctoral training. These time-bound moves may occur within or between countries, sectors or employers.
Within and across R&I sectors, by degree of relocation
Source: Authors’ elaboration
A considerable amount of connectedness and reinforcement can take place across different forms of mobility within the same career trajectory. For instance, collaboration without relocation at one point in a career might later lead to physical mobility on a short- or long-term basis or may influence the destination of choice for subsequent physical mobility (Rottenkolber et al., 2026[2]). Conversely, diaspora communities and historical institutional ties to a location of origin may promote subsequent collaboration or knowledge transfer between mobile R&I personnel and non-mobile counterparts within their origin countries (Ito et al., 2025[3]; Batista et al., 2025[4]). As studies have highlighted these beneficial connections, the framing of policy debates on mobility has shifted from solely treating the outflow of highly skilled people as an irrevocable national loss (“brain drain”) towards a broader view of mobility as "talent circulation", which under specific conditions can benefit all parties involved.
The ReICO project’s work on mobility is informed by several operational definitions used across international data collections and research projects. These definitions vary from narrow measures based on physical relocation to broader concepts that include virtual mobility and international engagement (Box 4.1). Choices of definition and associated data collection instruments determine which forms of mobility can be observed and monitored by policymakers and other interested parties.
ReICO directly uses, or takes into account, several mobility concepts used across international data projects.
International student mobility in the UNESCO/OECD/EUROSTAT (UOE) education data collection
The UOE data collection, a joint initiative across international organisations to collect data on formal education systems, defined international (or mobile) students as those who have crossed a border for the purpose of study, identified by prior education or residence abroad. International mobility is further classified into degree mobility (enrolment abroad leading to a foreign qualification) and credit mobility (shorter exchange or study-abroad periods undertaken while remaining enrolled at home). The UOE data collection captures degree mobility only (UNESCO-UIS / OECD / EUROSTAT, 2026[5]).
OECD migration data collections
The OECD's Database on Immigrants in OECD Countries (DIOC) identifies migrant stocks on the basis of country of birth. These statistics (applicable to wider workforce migration, not just R&I) distinguish between permanent-type migration (inflows granting, or capable of leading to, indefinite residence) from temporary migration, which covers time-limited categories such as work permits, seasonal and working-holiday schemes, and international students (OECD, 2026[6]).
Classifications of mobility of R&I workforce – the MORE studies
There are no internationally agreed definitions of mobility for R&I personnel. Instead, mobility concepts are commonly developed at project level. For example, the European Commission's series of studies on Mobility and career paths Of Researchers in Europe (MORE) studies distinguish between long-term international mobility (stays abroad of more than three months), short-term international mobility (less than three months), PhD mobility (obtaining a doctorate abroad), intersectoral mobility (movement between higher education, government, private non-profit and business sectors), and interdisciplinary mobility (movement between research fields). The most recent study (MORE4) also includes virtual mobility, dual positions and international collaboration and networks - forms of international engagement that do not require relocation (European Commission, n.d.[7]).
Understanding these mobility patterns, and what drives them, is of interest to policymakers working to expand and diversify the available talent within their R&I systems. Data availability varies across different forms of mobility. Some are well captured through official statistics or established proxies, such as degree mobility, permanent migration and international co-authorship, while others remain poorly measured, including short-term mobility in the business sector and cross-border work arrangements. Most available evidence relates to longer-term international relocation, drawing on administrative records, surveys of science professionals and, increasingly, digital traces generated through R&I activities.
Public policies relating to mobility often focus on attracting and retaining skilled R&I talent from abroad. Attracting international talent can be a fast and efficient means of boosting the domestic workforce, compared with the timeframes and investment required to train home-grown talent, but it also has implications for other countries’ policies and R&I systems. R&I systems and employing organisations like firms or universities compete for a relatively small global pool of highly skilled R&I talent, and thus national policy design in this area is often informed and shaped by which opportunities are on offer elsewhere. Specific policy instruments chosen nonetheless vary considerably across countries, reflecting differences in the structure of their R&I systems and the levers available to governments, as illustrated in Box 4.2.
Governments need evidence to judge whether these measures work, to understand not only who arrives as a result of these policies, but whether they stay, and the underlying drivers of their subsequent career decisions. This calls for drawing together the evidence that already exists, establishing what other data is needed to evaluate policy, and improving its availability over time.
This chapter examines three questions that can be addressed using ReICO and related data sources. A first question concerns advanced education and training: what share of master's and doctoral qualifications are earned by internationally mobile graduates, how that share has changed over the past decade, and how it varies across countries, fields and sex. A second question addresses the extent to which national R&I systems draw on highly qualified people from abroad. Finally, the chapter examines international mobility patterns of scientific authors, as revealed through bibliometric data.
Attracting and retaining highly skilled talent from abroad is an explicit policy objective in many countries, with governments targeting occupations and sectors of strategic economic or social importance, where domestic supply is constrained by demography or otherwise not sufficient to meet needs (OECD, 2025[8]). Within R&I systems, for example, acute skills shortages in critical sectors such as semiconductors and quantum computing risk constraining the rate at which technology can progress and diffuse in these areas (OECD, 2025[9]).
In response, governments deploy a broad mix of policy instruments, spanning immigration legislation and quotas, financial incentives such as salary subsidies for employers or tax breaks, and information and support services for new arrivals. Research-specific instruments include reforms to researcher recruitment and public sector employment rules, improved pension portability, and fellowships targeted at researchers based overseas (EC-OECD, 2026[10]) Though policy instruments across countries are often familiar in their objectives, they differ in the levers they use (funding, immigration, or active recruitment), whom they target (researchers abroad, returning nationals, or graduates already in the country), and how support is structured and financed, as illustrated in the examples below.
Selected recent initiatives for attracting international talent in OECD countries and the European Union
France – Choose France for Science (2025): A platform operated by the National Research Agency (ANR) under France 2030, through which universities and research organisations apply for state co-funding (up to 50% of a project’s total cost) to host international researchers in priority fields such as health, climate, artificial intelligence and low-carbon energy.
Germany – Global Minds Initiative Germany (2025): A federal initiative offering international researchers career prospects at all post-doctoral stages, channelling additional funding through established Humboldt Foundation and German Research Foundation (DFG) programmes, with German Academic Exchange Service (DAAD) measures for international students and doctoral candidates and integration support including for family members.
Japan – J-RISE Initiative (2025): A cross-ministerial package of around JPY 100 billion (USD 620 million), drawing on proceeds of the Japan University Fund, to promote international "brain circulation", with recruitment support programmes enabling top universities to host outstanding early-career researchers from overseas, including Japanese nationals abroad, under globally competitive conditions. In addition, a Startup Visa programme (2025) allows foreign entrepreneurs to live in Japan for up to two years while they establish their business before they must meet the stricter requirements of the standard "Business Manager" visa requirements.
Korea – K-STAR Visa Track (2025): A Ministry of Justice scheme expanding a fast-track residency pathway from a few science and technology institutes to international graduates of 32 designated universities (from January 2026). Graduates recommended by their university president obtain resident status immediately upon graduation without an employment requirement, with permanent residency possible in three years instead of six and special naturalisation open to exceptional researchers regardless of length of stay.
Netherlands – Fund to Accommodate Excellent International Scientists (2025): The Dutch Research Council (NWO) runs a EUR 50 million (USD 57 million) fund (the "Tulip Fund") through which research organisations nominate leading scientists working outside the EU/EEA and Switzerland, receiving up to EUR 1 million (USD 1.1 million) per recruited researcher, across all disciplines.
United Kingdom – Global Talent Network (2024): A Department for Business and Trade network building a pipeline of international professionals for priority science and technology roles (artificial intelligence, quantum, life sciences), reinforced in 2025 by a Global Talent Taskforce and a GBP 54 million (USD 72 million) Global Talent Fund covering relocation and research costs of world-class researchers and their teams.
European Union – Choose Europe (2025): A European Commission package of EUR 500 million (USD 573 million) for 2025-27 to make Europe "a magnet for researchers", including European Research Council (ERC) relocation top-ups of up to EUR 2 million (USD 2.3 million), a planned seven-year ERC "super grant", extended Marie Skłodowska-Curie Actions postdoctoral fellowships, streamlined entry procedures, and a forthcoming European Research Area Act enshrining freedom of scientific research in EU law.
Note: Information for France, Germany and the United Kingdom is drawn from the EC-OECD STIP Compass database, 2025 edition.
Source: European Commission/OECD (2026[10]) European Commission (2025[11]), Cabinet Office, Government of Japan (2025[12]) Ministry of Economy, Trade and Industry, Japan (2026[13]), Dutch Research Council (NWO) (2025[14]), Ministry of Justice, Republic of Korea (2025[15])
Shares of internationally mobile graduates in doctoral education in the OECD area are increasing
R&I careers often begin with acquiring an advanced degree, and research careers often commence with doctoral training. These stages of education also provide a first international mobility experience for many R&I personnel. Internationally mobile graduates accounted for a considerable proportion of graduates from master’s and doctoral programmes across the OECD in 2023, accounting for 19.4% of master’s graduates and 29.2% of doctoral graduates on average (Figure 4.2). The shares of mobile doctoral graduates were particularly high in Belgium, Luxembourg, New Zealand and Switzerland, each reaching or exceeding 50% of all graduates, while Australia, Austria, Iceland and the United Kingdom had shares greater than 40%.
Percentage of doctoral or master’s graduates
Notes: 1. Latest reference year differs: 2021 for the Netherlands (doctorate). 2. Earliest reference year differs: 2016 for Colombia and Italy, 2017 for Japan (master’s), 2018 for Israel and Poland (doctorate), and 2019 for Spain (doctorate).
Source: OECD, Research and Innovation Career Observatory (ReICO) Database based on OECD Education database, June 2026.
Across countries, shares of internationally mobile students are in general higher at master’s and doctoral level than at lower levels of tertiary education, underlining that mobility becomes more common as training becomes more specialised (OECD, 2023[16]). At the same time, international mobility was typically more prevalent among doctoral graduates than master's graduates in 2023, although the reverse was true in Australia, Latvia, Lithuania, Poland, Romania and the United Kingdom. The underlying reasons for these exceptions may vary across specific countries. For example, in some countries it is common for universities to actively offer English-language training to students from abroad for specialised “long first degrees” of five-year programmes leading to master’s, particularly in health fields. In others, the prestige of the higher education system or strong post-graduation career prospects may make master's-level study especially attractive to students from abroad.
While not all mobile graduates remain in their host countries, evidence from some countries shows reasonably high retention rates (for example, a recent study in Finland found that 63% of doctoral graduates remained resident in Finland three years after graduation), especially when countries offer post-study schemes to graduates (OECD, 2026[17]).
Overall student mobility at advanced tertiary levels has grown substantially between 2015 and 2023, despite the challenging period of the Covid-19 pandemic which severely curtailed international travel from 2020-2022 (OECD, 2025[18]; OECD, 2026[17]). Across the OECD, the share of internationally mobile doctorate graduates rose from around 24% in 2015 to 29% in 2023. The corresponding EU27 share increased from 19% to 25%. At master's level, the OECD average share rose from 15% to 19% between 2015 and 2023, while the EU27 share increased from around 11% to just under 18%.
These patterns highlight that international mobility to engage in advanced degree programmes has overall continued to expand, and that it is most pronounced at the level closest to entry into research careers. Several implications can be drawn from these patterns. Rising shares of mobile graduates, alongside overall growth in graduate numbers (see Chapter 2), expand the supply of highly qualified talent available to R&I labour markets. At the same time, they indicate that international mobility plays an increasingly important role in supporting national R&I systems. This can create opportunities through access to a broader talent pool may also make some systems more sensitive to changes in geopolitical conditions or policies affecting international mobility, reducing access to skilled R&I personnel on which some systems have come to rely (OECD, 2025[9]).
Most internationally mobile doctoral graduates are in STEM fields
STEM accounts for the majority of mobile doctoral graduates in almost every country shown in Figure 4.3. The OECD average is just under 56% and the share exceeds 60% in a large group of countries. In contrast, arts, humanities and education account for a much smaller share of mobile doctoral graduates in almost all countries, typically amounting to between 5% and 25% of all mobile doctoral graduates.
Percentage of internationally mobile doctoral graduates in all fields of study
Note: 1. Reference year 2022 for Latvia, New Zealand and Slovenia.
Source: OECD, Research and Innovation Career Observatory (ReICO) Database based on OECD Education database, June 2026.
There are only two countries (Bulgaria and Korea) where the arts, humanities and education field grouping accounts for a larger share of mobile doctoral graduates than from STEM fields.
The STEM share among mobile graduates reflects overall field composition within doctoral training as well as the specific choice patterns of internationally mobile students. As discussed in Chapter 2, STEM is the main combined field of award of doctorates in a majority of OECD countries. Nevertheless, Figure 4.3 shows that mobile doctoral graduates are often more concentrated in STEM fields than doctoral graduates overall. In countries such as Canada, Iceland, Japan, Norway, Sweden and the United States, STEM fields account for a substantially larger share of mobile doctoral graduates than of doctoral graduates overall, suggesting that international mobility plays a particularly important role in the supply of doctoral talent in STEM fields in many countries compared with other fields of study.
Aggregate STEM field shares mask concentrations of mobile graduates within individual component fields. Available national data shows that graduates from abroad account for a majority of doctorates in high-demand fields in some countries. The United States provides an illustrative example - in 2024, the majority of doctorates in computer and information sciences (61%), engineering (54%) and mathematics and statistics (51%) were awarded to temporary visa holders1 (National Center for Science and Engineering Statistics (NCSES), 2026[19]).
In all countries, women have lower mobility rates at doctoral level compared to master’s
Women make up a smaller share of internationally mobile graduates at doctoral level than at master's level in every country, as Figure 4.4 shows. At doctoral level, women make up a minority of mobile graduates in most countries: their share exceeds 50% only in Chile, Greece, Israel, New Zealand and Slovenia, and falls below 40% in Estonia, Hungary, Latvia, Norway, the Slovak Republic, Türkiye and the United States. At master's level, however, women account for more than half of mobile graduates in most countries and more than 60% in Iceland, Italy and Slovenia.
In the large majority of countries, at both doctorate and master’s level, the share of women among internationally mobile graduates falls below their share among graduates overall (Figure 4.4). This pattern suggests that, overall, women are less likely than men to undertake international mobility for advanced studies. This may reflect the design of talent attraction policies, which are often directed to STEM fields where women are less likely to enrol in advanced degree programmes, or other constraints and incentives bearing on women’s decision to move abroad rather than on entry into graduate study. Future analysis drawing on sex- and field-disaggregated data could help establish how much of the gap is accounted for by different factors.
Governments often hope to retain highly qualified graduates in their labour markets following their graduation, including within research and innovation systems, often putting in place specific schemes to retain such graduates, although the specifics and conditions can vary considerably across countries (OECD, 2026[17]). Many of these specific post-graduation schemes are designed to be temporary in nature, making it incumbent on the recent graduates to find more permanent opportunities that provide a long-term right to remain in the country or a path towards permanent residency. International graduates are often at an initial disadvantage in the job market compared to native-born graduates, which creates barriers to long-term retention of talent, although evidence from some countries suggests that this disadvantage lessens over time (OECD, 2026[17]).
Percentage of women among internationally mobile or all graduates
Note: 1. Reference year 2022 for Bulgaria.
Source: OECD, Research and Innovation Career Observatory (ReICO) Database based on OECD Education database, June 2026.
Foreign-born advanced degree holders are a major component of the R&I workforce in several systems
In the absence of data linking individuals from study to employment, the resident foreign-born population with advanced degrees offers the most widely available approximation of the extent to which international talent is integrated into R&I workforces. Place of birth is the most widely collected marker of international origin in official statistics (Box 4.1). Foreign-born status is, however, an imperfect proxy for mobility. Unlike internationally mobile graduates, it includes not only those who moved abroad to study and stayed, but also people who migrated after completing their education or who were educated in the destination country. As a result, foreign-born indicators are best interpreted as a measure of countries’ long-term accumulation of human capital from abroad rather than recent mobility or graduate retention.
Foreign-born advanced degree holders make up an important share of the highly qualified working-age population in many countries (Figure 4.5). Foreign-born account for close to three-quarters of all doctorate holders in Luxembourg and around 60% in New Zealand and Switzerland but make up less than 5% of all doctorate holders in Korea, Lithuania, Romania and the Slovak Republic. Many of the countries with higher shares have strong traditions of inward migration and settlement, such as Canada and New Zealand, or economies that invest heavily in research and development, such as Denmark, Sweden and Switzerland.
In most countries the foreign-born share is higher at doctoral than at master's level, and in several the gap between degree levels is substantial: in Denmark, Norway and Sweden, for example, the foreign-born share of doctorate holders exceeds that of master's degree holders by 15 percentage points or more, and sizeable gaps are also evident in Finland, France, Spain and the United States. This is reflective of the strongly international character of doctoral training and of research careers more generally, where post-doctoral mobility is embedded as a norm in career paths. A minority of countries, Australia, Chile, Germany, and the United Kingdom among them, show the opposite pattern, with the foreign-born more prevalent among master's degree holders than doctorate holders. Several of these countries also attract large numbers of international students at master's level (as shown in Figure 4.2). High shares of foreign-born advanced degree holders may reflect the retention of these graduates following their studies or indicate countries with migration channels that select on advanced qualifications obtained abroad. Future ReICO work will examine the alignment of policy contexts with international mobility indicators in more detail.
Percentage of doctorate or master’s degree holders aged 25-64
Note: Data for more countries are available on the database. 1. Reference year differs: 2020 for Korea, Slovenia (doctorate) and the Slovak Republic (doctorate), 2021 for Australia, Belgium, Canada, Chile (doctorate), Croatia, Czechia, Lithuania, Malta, Romania and Slovenia (master’s degree), 2022 for Ireland, 2023 for Finland, Iceland, Italy, New Zealand, Portugal (doctorate), Spain, Türkiye and the United States. 2. Definition of foreign-born differs: refers to foreign citizenship.
Source: OECD, Research and Innovation Career Observatory (ReICO) Database based on multiple data sources, June 2026. Further information on the specific sources used to compute this indicator can be found in the metadata of the ReICO database.
Shares of internationally mobile graduates and shares of foreign-born advanced degree holders are positively related
The two lenses used so far in this chapter - where students move to complete their studies, and where advanced degree holders were born - describe different populations. However, as Figure 4.6 shows, the share of graduates who are internationally mobile and the share of advanced degree holders who are foreign-born are positively related across countries, at doctorate level (Panel A) as well as master's level (Panel B). The association is of similar strength in both cases, with variation in one measure accounting for more than half the cross-country variation in the other (R² = 0.58 and 0.59 for doctorate and master’s respectively). Nonetheless, dispersion around the line is considerable for both doctorate and master’s degree holders. Variations of this size indicate that stocks of foreign-born degree holders are built by quite different routes across countries. Countries above the line may be building stocks through retention of mobile graduates trained in the domestic education system, while those below the line more strongly cultivating arrivals of people who qualified in another country.
Share of foreign-born degree holders aged 25-64 and share of internationally mobile graduates among all graduates at the level
Note: Reference year may differ for some countries.
Source: OECD, Research and Innovation Career Observatory (ReICO) Database based on multiple data sources, June 2026. Further information on the specific sources used to compute this indicator can be found in the metadata of the ReICO database.
Overall, Figure 4.6 shows that foreign-born advanced degree holders and internationally mobile graduates together make a sizeable contribution to the talent base of many R&I systems. Mobility provides considerable strength to these countries’ R&I systems, raising the size of the potential R&I workforce far above what native-born supply alone would allow, and supporting the building of beneficial collaborative ties and international networks. At the same time, it is a strength that partly rests on continued attractiveness of R&I careers within countries and may be vulnerable to changes in immigration policy, visa and settlement conditions, research funding availability, or competing destinations increasing in relative attractiveness, all of which can rapidly affect the human resource base for R&I in countries with large shares of mobile or foreign talent. The countries at the other end of the distribution, however, face an opposing challenge, with low foreign-born shares and low inward mobility leaving R&I systems more reliant on maintaining a domestic talent pipeline and/or increasing efforts to attract talent from abroad.
Age composition of foreign-born doctorate holders varies across countries, reflecting historical mobility patterns
As noted, the stock of foreign-born doctorate holders is the accumulation of decades of arrivals. Countries may have reached their current point through different routes - current stocks could be based on sustained policy efforts to recruit early-career workforce, for example or a legacy wave of migration from earlier years that has since diminished. The age distribution of foreign-born doctorate holders can help to indicate different cases across countries. For example, large shares of younger foreign-born doctorate holders is a reasonable indication that countries have successfully attracted or retained recent graduates or cultivated this talent among foreign-born already in the country. Concentration in older groups, on the other hand, might indicate a stock based on earlier migration waves migration not subsequently maintained.
As Figure 4.7 shows, countries exhibit very varied age profiles among doctorate holders. The age profile skews relatively young in Australia, Belgium, Czechia, Denmark, Spain and the United Kingdom. In these countries close to one-quarter or more of foreign-born doctorate holders are under 35, with the highest share (29%) in Australia. On the other hand, this younger group makes up less than 10% of foreign-born doctorate holders in Croatia, Greece and Latvia. These patterns also reflect to some extent the age of graduation of doctorate holders in different countries (see Chapter 2). Still, some countries exhibit notable younger age profiles. Overall, across the younger age bands, more than 70% of foreign-born doctorate holders are under 45 in Australia and Denmark, compared with less than 30% in Latvia.
Percentage of foreign-born doctorate holders aged 25-64
Notes: Data for more countries are available on the database. 1. Reference year differs: 2020 for Greece, Hungary, Korea, Mexico, and the Slovak Republic, 2021 for Australia, Belgium, Canada, Chile, Croatia, Czechia, Lithuania, Malta and Romania, 2022 for Ireland and Luxembourg, and 2023 for Finland, Italy, New Zealand, Portugal, Spain, Sweden, Türkiye, the United Kingdom and the United States. 2. Definition of foreign-born differs: refers to foreign citizenship.
Source: OECD, Research and Innovation Career Observatory (ReICO) Database based on multiple data sources, June 2026. Further information on the specific sources used to compute this indicator can be found in the metadata of the ReICO database.
Conversely, 55-64 year-olds make up around a quarter or more of the foreign-born doctorate population in Croatia, Greece, Luxembourg, Romania and the Slovak Republic, and exceed 40% in Latvia and Lithuania. These age patterns, in which foreign-born doctorate holders are concentrated in the older age bands largely echo the patterns discussed in Chapter 2, with similar age profiles in foreign-born doctorates in countries in which the doctoral population in general is ageing. This can indicate that migration is not currently offsetting ageing domestic highly skilled population in many countries and greater efforts are likely to be needed in the coming years to replenish and build workforces. Many countries are already working on such measures. In Latvia, for example, one of the countries most adversely affected by these trends, comprehensive reforms of doctoral education and the wider academic career framework are underway in efforts to increase supply of research and innovation workforce (OECD, 2026[20]).
A final element of interest when considering renewal of the foreign-born R&I workforce is the extent to which countries can recruit or retain foreign-born doctorate holders equally from across different groups of the available talent pool. Figure 4.8 shows that women consistently make up a larger share of foreign-born master’s degree holders than doctorate holders. The share of women exceeds 50% in most countries and reaches 60% or more in Greece, Korea, Lithuania, Portugal and the Slovak Republic. At doctoral level, the picture differs. Women make up half of foreign-born doctorate holders in just six countries: Chile, Italy, Korea, Latvia, Lithuania and Spain, while the share falls below 40% in Austria, Finland, Luxembourg, Türkiye and the United States. While these patterns partly reflect differences in the composition of fields and qualifications, evidence suggests that women researchers have historically been less likely than men to engage in international mobility, although more recent studies indicate that this gap is narrowing (Zhao et al., 2023[21]).
Percentage of foreign-born doctorate or master’s degree holders aged 25-64
Notes: Data for more countries are available on the database. 1. Reference year differs: 2020 for Greece (doctorate), Hungary (doctorate), Korea and the Slovak Republic (doctorate), 2021 for Australia, Austria (doctorate), Belgium, Canada, Chile (doctorate), Croatia, Czechia, Lithuania, Malta and Romania, 2022 for Ireland and Luxembourg (doctorate), and 2023 for Finland, France (doctorate), Greece (master’s), Italy, New Zealand, Norway (master's), Portugal (doctorate), Spain, Türkiye, the United Kingdom (doctorate) and the United States. 2. Definition of foreign-born differs: refers to foreign citizenship.
Source: OECD, Research and Innovation Career Observatory (ReICO) Database based on multiple data sources, June 2026. Further information on the specific sources used to compute this indicator can be found in the metadata of the ReICO database.
Digital footprints of R&I activities can support monitoring of mobility
The indicators examined so far in this chapter describe mobility among graduates and the prevalence of foreign-born individuals in the wider population of advanced degree holders. As discussed in Chapter 1, advanced degree holders are an especially important group for R&I systems, since they supply vital knowledge and skills and given the importance of the doctorate as a principal formal entry requirement for many research career pathways. Data on the mobility of holders of advanced qualifications therefore provides important signals for policymakers. However, as highlighted, this group does not overlap exactly with the R&I workforce. As shown in Chapter 3, many doctorate holders work outside research, in public administration, management or professional services, while R&D is more often than not carried out by people without a doctorate. The extent to which the sector and industry of employment is associated with mobility patterns cannot be easily ascertained from existing official data, calling for consideration of a wider range of data sources and development of novel methodologies for working with them.
Digital traces generated through the professional activities of those working in R&I offer new opportunities to measure and analyse geographic mobility. Researchers, inventors, entrepreneurs and other R&I professionals leave digital traces across online records, including scientific publications, patent applications, professional networking platforms, conference participation records, and organisation websites. Despite their limitations, these data sources often contain information on institutional affiliations and locations that can be linked over time to reconstruct individual mobility trajectories (Box 4.3).
Digital technologies generate flows of novel and timely information about people through the “footprints” they leave when engaging in online actions and activities (OECD, 2026[22]). Compared with traditional surveys and population censuses, digital footprint data offer several advantages. They can provide near real-time information, cover large populations of highly skilled workers, and enable the study of mobility patterns at a fine geographic and occupational level. For example, changes in author affiliations recorded in online repositories of scientific publications can reveal the mobility of scientific researchers between universities and research institutes, while patent records can be used to identify the relocation of inventors across firms, regions and national innovation systems. Professional networking platforms and online CV repositories also permit recording of occupations at a more granular level and can in some cases help identify transitions that are not visible in traditional administrative data, including temporary assignments, international migration, remote cross-border work, and movements between sectors (as further discussed in Chapter 5).
At the same time, the use of digital footprint data to track mobility and other elements of career trajectories of the R&I workforce presents important challenges. Not all R&I professionals have an equivalent online presence, and coverage may vary across countries, occupations and demographic groups. For example, inferences about careers and mobility of scientific authors are far less reliable for authors who publish infrequently and for those moving into or out of roles where publishing is not customary, as is typically the case in industry (Appelt et al., 2015[23]). Furthermore, changes in affiliation may not necessarily correspond to physical relocation, particularly in an era of hybrid and remote work. Furthermore, the use of personal-level digital traces requires careful attention to disambiguation of individuals with identical names and those with multiple online personas for the same individual, privacy, transparency and ethical standards. As a result, information gleaned from digital footprint measures are often most valuable when combined with established statistical and administrative sources.
Despite these limitations, digital footprint data present a promising complement to traditional approaches for understanding flows of talent across borders. Their increasing availability is expanding the capacity of researchers and policymakers to monitor the geographic mobility of some groups of R&I personnel and to assess its implications for innovation performance, human capital development and international scientific collaboration.
Indicators based on the analysis of scientific publication records (i.e. bibliometrics) can illuminate mobility patterns specifically for those working as researchers, particularly researchers in sectors where scientific publications are a primary output of research, such as higher education and government or other public research institutes. This section sets out the main recent trends and patterns across countries relating to mobility patterns, as indicated by available bibliometric data.
Bibliometric data implies considerable differences in mobility flows across countries
There are several possible approaches to measuring implied international mobility flows based on bibliometrics. One of the simplest approaches used by OECD categorises authors with two or more publications in relation to their past affiliation history. With respect to a reference country (or other geographical area): new inflows are authors first observed abroad who subsequently appear in the country, returnees are authors previously in the country who appear abroad before returning, and outflows are authors previously in the country who subsequently appear with a foreign affiliation (OECD, 2026[24]). Authors whose entire record lies within the reference country are classified as stayers. Figure 4.9 shows gross implied mobility flows for scientific authors as measured with reference to 2024, aggregating leavers, returnees and new inflows with respect to each country, relative to all authors found with an affiliation in the country in 2024.
Composition of gross flows, relative to all scientific authors with an affiliation recorded in 2024 with at least two publications
Note: OECD and EU27 totals refer to the aggregate value obtained by combining all Member countries into a single entity.
Source: OECD, Research and Innovation Career Observatory (ReICO) Database based on the OECD Bibliometrics database, June 2026.
The measure sums inflows and outflows and therefore indicates the total implied volume of movement of scientific authors through a country. In total, in the OECD, this calculated rate stood at 10.2% (with the EU27 at 10.7%) indicating that (and taking into account the caveats highlighted above) roughly one scientific author in ten recorded with an affiliation in 2024 had a mobility experience linked to an OECD or EU27 reference country. The spread of this measure across countries is very wide: the rate exceeds 30% in Cyprus and approaches 28% in Luxembourg, but stands at less than 5% in Brazil, Bulgaria, China, India, Indonesia, Japan, Poland and Romania. The lowest share of mobile authors found across the countries with available data is 2.4%, in China.
Figure 4.9 indicates that overall mobility flows tend to be largest, in percentage terms, in a selection of small- or medium-sized systems, such as Cyprus, Ireland, Luxembourg and Switzerland, and closer to- or below average in larger systems such as France, Germany and the United States. This might indicate that smaller economies may not maintain critical mass of research activity across the full range of scientific fields, so a larger proportion of its researchers must engage in mobility to develop their careers. It may also indicate the strength of international networks and connectedness to international collaboration within these larger countries, which can provide the benefits associated with internationalisation of R&I without the need for physical movement. On the other hand, it must be noted that lower gross mobility rates in the largest systems are still consistent with very substantial flows in and out, as detailed in the next section.
In most countries implied net flows broadly balance out over time, with some notable exceptions
While gross flows give an indication of the total extent of churn of scientific authors through a country’s research system, net flows, as shown in Figure 4.10, indicate the overall direction of the movement. Net flows are calculated over a long timeframe to account for high year-on-year volatility in inflows and outflows, particularly in smaller systems.
Inflows minus outflows of scientific authors who changed affiliation to or from a country in the reference period
Note: OECD and EU27 totals refer to the aggregate value obtained by combining all Member countries into a single entity.
Source: OECD, Research and Innovation Career Observatory (ReICO) Database based on the OECD Bibliometrics database, June 2026.
Implied net flows result from subtracting outflows from inflows, so a system with roughly equal inflows and outflows numbers will appear to have close to zero net flow regardless of the volume of flows. Conversely, a system with modest overall mobility, as measured above by gross flows, can register a large net position if most of that mobility is one-directional. The United States illustrates this point: it recorded a relatively low gross mobility rate yet registers by far the largest net gain in scientific authors over time, as shown in Figure 4.10. The EU27 also shows an overall net gain over time, although the patterns vary across individual member countries. In absolute terms, most countries record net flows of less than 500 authors over the period 2010-2024, and only nine individual countries show net flows of more than 1 000 authors in either direction over the period.
Flows of scientific authors can be notably volatile from one year to the next, and a figure aggregated over a long period may obscure whether a country's net position was established early on in a long reference period or acquired only recently. Some countries and regions nonetheless appear to emerge consistently as target destinations for geographically mobile scientific authors. Net flows to the United States were positive in almost every year (except 2011) between 2010 and 2024, while those to the EU27 turned positive in 2020 and have remained so since (OECD, 2026[24]).
As discussed earlier, mobility flows in and out of countries cannot be characterised fully as net gains or losses, in light of the likely beneficial impacts of mobility for both origin and destination countries. A country recording a persistent net outflow may nonetheless be an active participant in international science if its emigrant researchers continue to co-author with colleagues at home. Relationships between affiliation changes and extent of international collaboration can be potentially explored to some extent through analysis of bibliometric data. International co-authorship of publications is a standard bibliometric proxy for such collaborations and can give an indication of how its prevalence is related to mobility of scientific authors (Box 4.4).
Bibliometric data allows for the computation of country comparisons of the extent to which scientific publications involve international collaborations (i.e. papers where there is more than one scientific author and where authors are affiliated to institutions in different countries). Co-authorship may arise in relation to international mobility of the connected authors, or independently of it. Figure 4.11 shows that these two indicators are positively correlated at country level (R2 = 0.79). Countries with high mobility rates are, with few exceptions, also those whose authors collaborate most widely across borders. At one end of the distribution, Bulgaria, China, India, Indonesia, Japan, Romania and Türkiye record low values on both measures. At the other, Austria, Belgium, Cyprus, Iceland, Ireland, Luxembourg and Switzerland combine high mobility with high collaboration. Considering shares of publications with international collaboration together with shares of authors engaging in international mobility can indicate whether countries whose scientific authors move frequently are also those where authors frequently collaborate across borders, or whether the two operate as substitutes, with collaboration serving as an alternative route to international engagement where physical relocation prospects are limited. The pattern shown in Figure 4.11 indicates that some modes of international engagement may reinforce rather than substitute for one another, and that international openness of R&I systems may be best analysed across multiple dimensions rather than assessed based on any one form of mobility.
Source: OECD, author’s analysis using Science and bibliometric indicators database, May 2026.
Most mobile scientific authors stay within their broad geographic region
National policymakers are interested not only in the size of mobility flows in and out of countries, but also in their source and destination. Measures to attract talent, such as bilateral agreements, scholarship schemes or recognition arrangements are typically directed at particular countries and are best designed informed by information on the origin of current inward flows. Conversely, schemes to encourage return from the diaspora depend on knowing the destinations of leavers. Geographic patterns of inflow origins and outflow destinations offer a broad indication of these relationships. Figure 4.12 shows these patterns for scientific authors over the period 2010-2024 based on bibliometric data, by aggregating individual country-to-country bilateral flows, showing that such flows are heavily shaped by regional proximity, but not exclusively so.
The figure shows that European countries exchange flows predominantly within Europe, with inflow and outflow shares from other European countries reaching as high as 86% and 88% respectively in Cyprus and 89% and 87% in the Slovak Republic, although patterns vary considerably within the European Union countries (Box 4.5). France and the United Kingdom are the least regionally concentrated European systems on this measure, each having under half their inflows and outflows from Europe and relatively high shares of flows to and from the United States, Canada and the Asia-Pacific region among European countries. Many other countries show broadly similar regionally based patterns of inflows and outflows of scientific authors. Among Asian countries, for example. Indonesia has the largest share of inflows and outflows within its region, with only 8% of authors engaging in outward mobility towards the United States and Canada. Other Asian countries show markedly different patterns of exchange with the United States and Canada, however. In both China and India, inflows of authors from and outflows to the United States and Canada made up the majority or close majority of all such flows between 2010 and 2024.
Percentage of mobile scientific authors moving from or to each country
Note: A total of 51 countries are included in the analysis, comprising all OECD and European Union member countries, along with OECD key partners and accession candidate countries (OECD, n.d.[25]). Countries are grouped by geographic regions, ordered by the number of countries in the group. For each regional group, countries are in descending order based on the share of inflow from Europe.
Source: OECD, author’s analysis using Science and bibliometric indicators database, May 2026.
Bibliometric data allow mobility to be analysed not only at country level but also from other geographic perspectives, including flows into and out of regional or country groupings with policy relevance. These groupings may be based on areas of free movement, reduced administrative barriers to mobility or initiatives where countries cooperate to encourage mobility or attract talent from further afield, complementing nationally based initiatives. In this context, it is of interest to examine how talent flows in and out of country groupings, as well as individual countries.
The European Union provides an example of such an entity, where member countries, in addition to developing national policies, collectively work to attract talented individuals to the EU27 area. Figure 4.13 shows where scientific authors moving into and out of the EU27 came from and went to over the period 2010-2014, with movement between Member States distinguished from movement involving countries outside the EU.
Note: Non-EU27 countries are selected based on top 10 countries for origin or for destination in flow in and out of EU27 countries, excluding other EU27 countries.
Source: OECD, author’s analysis using Science and bibliometric indicators database, May 2026.
Mobility between EU member states accounts for 40% of inflows and 39% of outflows, with remaining shares involving origins and destinations outside the EU. The United States accounts for the largest single share of inflows (14%) and outflows (15%), followed by neighbouring countries the United Kingdom (11% each of inflows and outflows) and Switzerland (4% and 5% respectively) together account for a combined share that exceeds that of the United States for inflows. The rest of the world accounts for the remainder (21% of inflows and 20% of outflows), with no single other origin or destination accounting for more than 2% of total flows inward or outward. This indicates that, at EU level, mobility flows in and out of the EU from major origins and destinations largely converge over time to similar levels rather than flowing predominantly one way.
At the same time, as Figure 4.14 shows, mobility patterns vary considerably across EU countries. The share of total inflows originating from within the EU varies considerably across member states, ranging from under a quarter to more than three-quarters. Ireland has the lowest intra-EU share, at 23%, with the remaining three-quarters of its inflows originating outside the EU; it is followed by Cyprus (29%), France (31%) and Germany (34%), each drawing close to two-thirds of its inflows from outside the EU. At the other end, Lithuania, Luxembourg and the Slovak Republic draw more than half of their inflows from other EU member states.
Source: OECD, author’s analysis using Science and bibliometric indicators database, May 2026.
The evidence presented in this chapter shows that international mobility contributes not only to the supply of highly qualified talent, but also to the networks and collaborations through which knowledge is created, shared and applied. Countries take diverse approaches to degree mobility, migration and talent retention to build their research and innovation capacity, resulting in markedly different mobility profiles. As mobility patterns continue to evolve in response to demographic, economic and geopolitical change, ensuring that national systems can both attract and effectively harness highly skilled talent will remain a key policy priority. A better understanding of how people move between countries, sectors and career stages will be essential for ensuring that mobility supports both individual career development and the long-term resilience of research and innovation systems. Chapter 5 outlines potential avenues to improve insights into mobility patterns, in the context of the ReICO project.
References
[23] Appelt, S. et al. (2015), “Which factors influence the international mobility of research scientists?”, OECD Science, Technology and Industry Working Papers, No. 2015/2, OECD Publishing, Paris, https://doi.org/10.1787/5js1tmrr2233-en.
[4] Batista, C. et al. (2025), “Brain drain or brain gain? Effects of high-skilled international emigration on origin countries”, Science, Vol. 388/6749, https://doi.org/10.1126/science.adr8861.
[12] Cabinet Office, Government of Japan (2025), Announcement of the J-RISE Initiative ~Japan Research & Innovation for Scientific Excellence~, https://www8.cao.go.jp/cstp/stmain/20250613.html.
[14] Dutch Research Council (NWO) (2025), Tulip Fund, https://www.nwo.nl/en/researchprogrammes/tulip-fund.
[10] EC-OECD (2026), STIP Compass: International Database on Science, Technology and Innovation Policy (STIP), https://stip.oecd.org (accessed on July 2026).
[11] European Commission (2025), Choose Europe for Science, https://research-and-innovation.ec.europa.eu/news/all-research-and-innovation-news/choose-europe-science-eu-comes-together-attract-top-research-talent-2025-05-23_en.
[7] European Commission (n.d.), Mobility Patterns and Career Paths of EU Researchers - Project description, https://www.more-4.eu/project-description (accessed on July 2026).
[3] Ito, R. et al. (2025), “From drains to bridges: The role of internationally mobile PhD students in linking non-mobile with foreign scientists”, Journal of Development Economics, Vol. 177, https://doi.org/10.1016/j.jdeveco.2025.103577.
[13] Ministry of Economy, Trade and Industry, Japan (2026), Startup Visa, https://www.meti.go.jp/english/policy/economy/startup_nbp/startup_visa.html (accessed on 2026).
[15] Ministry of Justice, Republic of Korea (2025), 법무부, 과학기술의 미래를 여는 우수인재 정착 유도를 위해 「K-STAR 비자트랙」 신설“ [MOJ establishes the K-STAR Visa Track], https://www.moj.go.kr/bbs/moj/182/599033/artclView.do.
[19] National Center for Science and Engineering Statistics (NCSES) (2026), Doctorate Recipients from U.S. Universities: 2024, https://ncses.nsf.gov/pubs/nsf26315.
[20] OECD (2026), Academic career reform in Latvia: Taking stock and looking forward, OECD Publishing, https://doi.org/10.1787/cd46e69c-en.
[17] OECD (2026), International Students in Higher Education: A comparative analysis of trends, challenges and policy responses in Australia, Canada, France, Germany, the Netherlands and the United Kingdom, Higher Education, OECD Publishing, Paris, https://doi.org/10.1787/005ff28d-en.
[6] OECD (2026), OECD Databases on Migration, https://www.oecd.org/en/data/datasets/oecd-databases-on-migration.html.
[24] OECD (2026), Science and bibliometric indicators - 2026 edition, https://www.oecd.org/en/data/datasets/science-and-bibliometric-indicators.html.
[22] OECD (2026), The OECD Going Digital Measurement Roadmap 2026, OECD Publishing, https://doi.org/10.1787/b455e132-en.
[8] OECD (2025), International Migration Outlook 2025, OECD Publishing, Paris, https://doi.org/10.1787/ae26c893-en.
[9] OECD (2025), OECD Science, Technology and Innovation Outlook 2025: Driving Change in a Shifting Landscape, OECD Publishing, Paris, https://doi.org/10.1787/5fe57b90-en.
[18] OECD (2025), What are the key trends in international student mobility?, OECD Publishing, https://doi.org/10.1787/2a423a76-en.
[16] OECD (2023), Education at a Glance`, OECD Publishing, https://doi.org/10.1787/e13bef63-en.
[1] OECD (2013), Knowledge Networks and Markets, OECD Publishing, https://doi.org/10.1787/5k44wzw9q5zv-en.
[25] OECD (n.d.), Members and partners, https://www.oecd.org/en/about/members-partners.html.
[2] Rottenkolber, A. et al. (2026), “The role of professional networks and institutional prestige in shaping the first career moves of scholars”, PNAS Nexus,, Vol. 5/6, p. 168, https://doi.org/10.1093/pnasnexus/pgag168.
[5] UNESCO-UIS / OECD / EUROSTAT (2026), UNESCO-UIS / OECD / EUROSTAT Manual on Concepts, Definitions, and Classifications, https://ec.europa.eu/eurostat/documents/d/education-and-training/uoe-manual-2026-final.
[21] Zhao, X. et al. (2023), “A gender perspective on the global migration of scholars”, Proc. Natl. Acad. Sci. U.S.A. 120 (10) e2214664120, https://doi.org/10.1073/pnas.2214664120.
← 1. The United States figures refer to doctorate recipients with temporary visas, as recorded in the Survey of Earned Doctorates. This is defined differently to mobile graduates, as it includes candidates on temporary visas whose previous qualifications were obtained in the United States.