Scientific research has long been recognised as a central driver of technological progress and economic growth (Romer 1990). It also plays a critical role in shaping industrial innovation and informing policy decisions (Branstetter and Ogura 2005, Hjort 2021). In recent years, the global scientific landscape has become the subject of growing policy debate. Governments in many advanced economies are reconsidering public investment in science amid fiscal pressure: for instance, US federal science funding has recently faced billions of dollars in cancelled grants (Bedekovics and Ragland 2025). On the other hand, concerns about international competition – particularly the rapid rise of research from middle- and low-income countries like China and India – have intensified discussions about the future of scientific leadership. Despite the growing scientific capacity across regions and the rising intensity of global scientific competition, we still lack systematic, large-scale metrics to capture the geography of scientific production over time. It is important to know where science is produced, what geographies it focuses on, how it disseminates across geographies, and how these trends vary over time and over fields.

In a new study (Nagaraj and Yao 2026), we provide such a multifaceted analysis. Specifically, we focus on three dimensions – where science is produced, what science studies, and where science is consumed – using data on 44 million publications from nearly 12,000 journals. To measure the geography of production, we code the institutional locations of authors. To capture the geography of research focus, we characterise the regional orientation of a paper’s topic using natural language processing applied to titles and abstracts (Nagaraj and Yao 2023). This would capture, for example, a development economics article focused on India based on the language used in the abstract, even if the paper was written by US authors. And to quantify the geography of research consumption, we analyse the downstream citation flows across different regions for each publication. Analysing these measures over different fields and the relative quality of science, we offer a detailed portrait of how the geography of science has evolved.

Our paper contributes to the growing literature at the intersection of the science of science and the geography of innovation. Existing research has examined how knowledge production and diffusion shape scientific progress, as well as how innovation clusters and regional spillovers influence inventive activity (Loumeau and Egger 2019, Whalley and Andrew 2021). However, there is limited work on the geography of science and most of it has focused on particular fields, specific countries, or a single dimension of geography – typically the location of authors. As a result, we lack an integrated understanding of how scientific activity is organised globally across multiple dimensions. Here, we provide a comprehensive analysis of the geography of science and a unified framework for understanding the global organization of scientific research.

Shifts in the geography of scientific production

The global distribution of scientific production has changed dramatically over the past four decades, as shown in Figures 1a and 1d. In 1980, the US accounted for over 40% of global scientific publications. High-income EU countries and other high-income countries accounted for over 50%, reflecting the strong concentration of research capacity in a small number of advanced economies. Over time, this dominance has steadily declined as scientific production expanded rapidly across other regions. By 2020, the US share had fallen to about 15% and the EU’s to around 22%, with France dropping from the sixth-largest producer in 1980 to the tenth in 2020. In contrast, China had become the largest producer of scientific publications globally, driven by both an expanding scientific labour force and rising productivity. Although this shift reflects the rapid rise of China, it is also part of a broader expansion of scientific activity across emerging economies. Countries such as India and Russia have increased their share of global publications as investments in higher education, research infrastructure, and scientific training have expanded worldwide.

Figure 1 Trends in the geography of science

Despite these changes in the volume of scientific output, important differences remain across countries. Though with declining dominance, advanced economies – in particular the US and EU – continue to play a central role in frontier science when we focus on top journals and breakthrough publications. While China takes the lead in both quantity and quality, taking over the US to be the biggest producer in top journals, other emerging economies remain struggling to present in the highest-impact venues. These patterns suggest that while scientific production has become more geographically distributed, the structure of the global scientific system remains uneven.

Scientific production is not evenly distributed across research fields, as shown in Figure 2. In the 1980s, the US dominated scientific output across nearly all disciplines, particularly in the social sciences, health sciences, and humanities, while European countries played a secondary but important role in many areas of science. Over time, this landscape has shifted substantially. China has emerged as the largest producer in many fields, especially in engineering, physical sciences, and several areas of the life and environmental sciences. At the same time, high-income European countries have become leading contributors in many areas of the social sciences and humanities. Notably, other middle- and low-income countries play a particularly important role in fields such as agriculture, veterinary, and food sciences.

Focusing on the top 5% journals in 2022 reveals a similar three-way structure. China and Europe remain major contenders across many fields, while the United States continues to lead in several domains – most notably in parts of biomedical and health sciences, psychology, and mathematical sciences. Taken together, these patterns highlight strong cross-country specialisation in the global research system, reflecting differences in national priorities, institutional strengths, and economic development paths.

Figure 2 Dynamics by field of study

The dynamics of research focus and consumption

Besides the geography of production, the geography of research orientation has also shifted markedly. Historically, when scientific work focused on specific geographic regions, it disproportionately examined the West – especially the US – and this pattern remains particularly strong in top science. As shown in Figure 1b, over time, China has become a major subject of study, accounting for over 20% of geographically focused research in 2022.

Other middle- and low-income countries have also become increasingly prominent as objects of scientific study – driven largely by India and Brazil, which together account for nearly 35% of research topics today. This rise reflects both the expansion of local research capacity and growing global interest in issues that disproportionately affect emerging economies. Despite this broader expansion, top journals and breakthrough publications still disproportionately focused on high-income countries, highlighting the continued concentration of high-impact scientific attention.

Finally, we examine the geography of research consumption by analysing where citations to scientific publications originate. Using the Herfindahl–Hirschman Index (HHI) based on the regional distribution of forward citations, we find that the geographic concentration of citations has declined over time, indicating that scientific knowledge is increasingly consumed by a broader set of regions. However, this trend varies substantially depending on where the research is produced. In particular, publications originating from China and other middle- or low-income countries exhibit consistently higher citation concentration than those from the US, Europe, or other high-income regions, suggesting that research from the emerging economies continues to be cited disproportionately within its domestic scientific community rather than globally. These patterns highlight that while scientific production has become more geographically dispersed, the global diffusion of knowledge remains fragmented across regions.

Scientific activity is becoming increasingly global, yet the organisation of the scientific system still reflects longstanding inequalities in attention, recognition, and knowledge diffusion. As more countries invest in research capacity and compete for scientific leadership, understanding how science is produced, studied, and circulated across regions becomes increasingly important. A more geographically inclusive scientific ecosystem will require not only expanding research capacity worldwide but also strengthening the global integration of knowledge, ensuring that ideas generated in different parts of the world can circulate, be recognised, and contribute to scientific progress.

References

Bedekovics, G and W Ragland (2025), “Mapping Federal Funding Cuts to U.S. Colleges and Universities.”

Branstetter, L G and Y Ogura (2005), “Is academic science driving a surge in industrial innovation? Evidence from patent citations”, NBER Working Paper 11561.

Hjort, J., Moreira, D., Rao, G., & Santini, J. F. (2021), “How research affects policy: Experimental evidence from 2,150 Brazilian municipalities”, American Economic Review 111(5): 1442-1480.

Loumeau, N and P Egge (2019), “The economic geography of innovation”, VoxEU.org, 16 January.

Romer, P M (1990), “Endogenous Technological Change”, Journal of Political Economy 98(5, Part 2): S71–S102.

Nagaraj, A and H Yao (2023), “Geographic Bias in Management Scholarship: Data-driven Estimates and Trends”, in E Fohim (ed.), Decolonizing Management and Organization Studies–Why, How, and What, Research in the Sociology of Organizations.

Nagaraj, A and R H Yao (2026), “The Geography of Science”, NBER Working Paper No. 34694).

Whalley, A and M Andrew (2021), “The geography of innovation between 1866 and today”, VoxEU.org, 20 February.