Investment in S&T by low- and middle-income countries has changed the geography of science over the past 30 years

An analysis was conducted by Carlos Henrique de Brito Cruz, former scientific director of FAPESP, using the Elsevier Scopus database, which covers 100 million publications and over 7,000 scientific publishers worldwide.

By Karina Toledo  |  Agência FAPESP – An analysis based on the Elsevier Scopus database points to a profound transformation in the geography of science over the last 30 years. In 1994, the vast majority (87%) of published articles had authors based in high-income countries (HICs). By 2024, however, the majority (60%) of articles were authored by scientists in low- and middle-income countries (LMICs), including Brazil, India, Russia, and China, among many others. The database covers over 100 million publications from more than 7,000 scientific publishers worldwide.

The growth in the number of Chinese authors – which surpassed that of the United States in 2015 – contributed significantly to this movement. The study also found that several other low- and middle-income nations played significant roles. “If we exclude China from this analysis, as well as the United States, the 28 low- and middle-income countries with the most authors are ahead [in number of authors] of the 27 countries of the European Union plus the United Kingdom,” comments Carlos Henrique de Brito Cruz, former scientific director of FAPESP, current senior vice president for Research Networks at Elsevier, and author of the study.

The data are included in a science-policy brief presented by Brito Cruz at the 10th Multistakeholder Forum on Science, Technology, and Innovation for the Sustainable Development Goals (SDGs). The forum was held in May at the United Nations (UN) headquarters in New York, United States.

“To identify the location of each author, we used the institutional address that the researcher provides when signing a publication [one of the fields in the Scopus database, which also includes the country where the institution is based]. This address is independent of the person’s nationality; it indicates where they work,” Brito Cruz explains to Agência FAPESP. “We measured the number of scientific articles with authors in each country and also the number of authors in each country.”

According to World Bank criteria, countries are categorized as high-income if they have a gross national income per capita of USD 13,845 or more. There are 79 such nations. Below this value are low- and middle-income countries, of which there are 134. “There are many more inhabitants in low- and middle-income countries, yet 30 years ago, only 9% of authors lived in these nations. In 2024, that number jumped to 60%,” says Brito Cruz.

According to the survey, the number of articles published globally has grown at a rate of 5% per year, largely due to an increase in authors from emerging countries. This growth can be attributed to the fact that, over the past 50 years, most low- and middle-income countries have been busy building scientific and technological (S&T) capacity. This includes creating universities, research institutes, development agencies, industrial research and development (R&D) facilities, and other types of scientific infrastructure.

In Brito Cruz’s assessment, many emerging economies realized that to achieve economic and social development, they needed S&T capacity in strategic sectors. “Some countries did this with more commitment, determination, and effectiveness, others with less. In 1980, China had 1.4 authors per million inhabitants, Iran had 10, and Brazil had 31. By 2024, China had reached 1,710 authors per million inhabitants, Iran had reached 1,231, and Brazil had reached 932,” he says.

According to Brito Cruz, Brazil was surpassed by Iran in 2011 and by China in 2019. From 1993 to 2003, the number of authors per million inhabitants grew by 12% per year in Brazil and China and by 22% in Iran. From 2014 to 2024, Brazil’s growth was 4% per year, China’s was 12%, and Iran’s was 5%.

“The growth rate of high-income countries is 3% per year. It isn’t good for countries that are still developing their S&T systems [low- and middle-income countries] to have similar numbers to those of already established countries. Brazil has been losing growth momentum since 2012, and we should be concerned about this, which I’ve called ‘King Lear syndrome.’ In Shakespeare’s play, the Fool says to the king: ‘You shouldn't have grown old before you grew wise.’ Brazil needs more researchers to avoid growing too old before it grows wise,” he points out.

Agenda 2030

Scientific publications related to the UN’s Agenda 2030, which was launched in 2015, also reflect this new geography of science. By 2024, 59% of publications on the SDGs were authored by individuals from low- and middle-income countries. Between 2015 and 2024, the growth rate of articles on the SDGs was higher in these countries (12.7% per year) than in high-income countries (5.4% per year). For ten of the 17 SDGs, the share of articles with authors from low- and middle-income countries was already higher than that of high-income countries in 2024: 59% versus 54%. As the text explains, the sum differs from 100% because some publications have authors in both groups.

“The efforts of emerging countries to build capacity in science and technology help advance the 2030 Agenda more quickly. In other words, without these efforts, we’d be much farther behind in achieving the SDGs,” says Brito Cruz.

Of the 38 leading development agencies mentioned in over 10,000 articles on the SDGs published between 2020 and 2024, 20 are from low- and middle-income countries. Notably, nine are from China, four are from India, and three are from Brazil, including the National Council for Scientific and Technological Development (CNPq), the Coordination for the Improvement of Higher Education Personnel (CAPES), and FAPESP. In Brazil, approximately one-third of the studies supported by these agencies are related to one of the 17 SDGs. CNPq and CAPES are among the ten national funding agencies that have supported the most articles related to the SDGs.

Main gaps

Despite the growth in the number of authors and the volume of articles, scientific production in low- and middle-income countries continues to face challenges. For example, bibliometric indicators for the period between 2021 and 2024 show a lower percentage of international co-authorship: 25.2%, compared to 37.4% in high-income countries. “One difficulty is the cost of collaborating – travel, accommodation. Another is the instability that exists in research funding in many low- and middle-income countries, such as Brazil. Collaborators expect predictability. There are also barriers related to the language of each country and the lack of proficiency in English, which has been the preferred language of scientific research for decades,” says Brito Cruz.

The study also revealed a disparity in the number of articles resulting from collaborations between academia and the business sector. In low- and middle-income countries, this figure stands at 1.9%, compared to 3.6% in high-income countries. According to Brito Cruz, one explanation is that there are generally fewer advanced companies with intensive R&D activities in low- and middle-income countries.

“In order for low- and middle-income countries to more effectively transform knowledge into economic development, in addition to having more universities and research institutes, it’s essential that they have more companies active in R&D. To understand the determinants and limitations of this university-business interaction, it’s necessary to look at the problem from two sides: one is the academic side – what percentage of publications with authors in universities have co-authors in companies; the other is the business side – what percentage of articles with authors in companies have co-authors in universities. In Brazil, for example, 3% of publications with authors in universities have co-authors in companies, and 87% of publications with authors in companies have co-authors in universities. In the United States, 6% of publications with authors in universities have co-authors in companies, and 65% of publications with authors in companies have co-authors in universities. It’s clear that, in Brazil, authors of ideas at universities can’t and wouldn’t be able to find more co-authors to collaborate with in companies because almost all of those available are already collaborating [87%],” he explains.
In the final part of the article, Brito Cruz argues that this new geography of science requires adjustments to global science and technology policies. His recommendations include supporting low- and middle-income countries in creating and strengthening their own research funding agencies, promoting more collaboration between these countries (South-South cooperation), and adjusting international research governance (committees, organizations) to reflect the growing participation and contributions of emerging countries, which are currently underrepresented.

The article “How a changing geography of research can accelerate the advancement of the U.N. SDGs in low- and middle-income countries” can be accessed at sdgs.un.org/documents/de-brito-cruz-ch-2025-how-changing-geography-research-can-accelerate-advancement-un-sdgs.