google research paper publication

• {{link.text}}

Publications

Google publishes hundreds of research papers each year. Publishing is important to us; it enables us to collaborate and share ideas with, as well as learn from, the broader scientific community. Submissions are often made stronger by the fact that ideas have been tested through real product implementation by the time of publication.

We believe the formal structures of publishing today are changing - in computer science especially, there are multiple ways of disseminating information.  We encourage publication both in conventional scientific venues, and through other venues such as industry forums, standards bodies, and open source software and product feature releases.

Open Source

We understand the value of a collaborative ecosystem and love open source software .

Product and Feature Launches

With every launch, we're publishing progress and pushing functionality.

Industry Standards

Our researchers are often helping to define not just today's products but also tomorrow's.

"Resources" doesn't just mean tangible assets but also intellectual. Incredible datasets and a great team of colleagues foster a rich and collaborative research environment.

Couple big challenges with big resources and Google offers unprecedented research opportunities.

22 Research Areas

• Algorithms and Theory 608 Publications
• Data Management 116 Publications
• Data Mining and Modeling 214 Publications
• Distributed Systems and Parallel Computing 208 Publications
• Economics and Electronic Commerce 209 Publications
• Education Innovation 30 Publications
• General Science 158 Publications
• Hardware and Architecture 67 Publications
• Human-Computer Interaction and Visualization 444 Publications
• Information Retrieval and the Web 213 Publications
• Machine Intelligence 1019 Publications
• Machine Perception 454 Publications
• Machine Translation 48 Publications
• Mobile Systems 72 Publications
• Natural Language Processing 395 Publications
• Networking 210 Publications
• Quantum A.I. 30 Publications
• Robotics 37 Publications
• Security, Privacy and Abuse Prevention 289 Publications
• Software Engineering 100 Publications
• Software Systems 250 Publications
• Speech Processing 264 Publications

3 Collections

• Google AI Residency 60 Publications
• Google Brain Team 305 Publications
• Data Infrastructure and Analysis 10 Publications

Stand on the shoulders of giants

Google Scholar provides a simple way to broadly search for scholarly literature. From one place, you can search across many disciplines and sources: articles, theses, books, abstracts and court opinions, from academic publishers, professional societies, online repositories, universities and other web sites. Google Scholar helps you find relevant work across the world of scholarly research.

How are documents ranked?

Google Scholar aims to rank documents the way researchers do, weighing the full text of each document, where it was published, who it was written by, as well as how often and how recently it has been cited in other scholarly literature.

Features of Google Scholar

• Search all scholarly literature from one convenient place
• Explore related works, citations, authors, and publications
• Locate the complete document through your library or on the web
• Keep up with recent developments in any area of research
• Check who's citing your publications, create a public author profile

Disclaimer: Legal opinions in Google Scholar are provided for informational purposes only and should not be relied on as a substitute for legal advice from a licensed lawyer. Google does not warrant that the information is complete or accurate.

• Privacy & Terms

Reference management. Clean and simple.

Google Scholar: the ultimate guide

What is Google Scholar?

Why is google scholar better than google for finding research papers, the google scholar search results page, the first two lines: core bibliographic information, quick full text-access options, "cited by" count and other useful links, tips for searching google scholar, 1. google scholar searches are not case sensitive, 2. use keywords instead of full sentences, 3. use quotes to search for an exact match, 3. add the year to the search phrase to get articles published in a particular year, 4. use the side bar controls to adjust your search result, 5. use boolean operator to better control your searches, google scholar advanced search interface, customizing search preferences and options, using the "my library" feature in google scholar, the scope and limitations of google scholar, alternatives to google scholar, country-specific google scholar sites, frequently asked questions about google scholar, related articles.

Google Scholar (GS) is a free academic search engine that can be thought of as the academic version of Google. Rather than searching all of the indexed information on the web, it searches repositories of:

• universities
• scholarly websites

This is generally a smaller subset of the pool that Google searches. It's all done automatically, but most of the search results tend to be reliable scholarly sources.

However, Google is typically less careful about what it includes in search results than more curated, subscription-based academic databases like Scopus and Web of Science . As a result, it is important to take some time to assess the credibility of the resources linked through Google Scholar.

➡️ Take a look at our guide on the best academic databases .

One advantage of using Google Scholar is that the interface is comforting and familiar to anyone who uses Google. This lowers the learning curve of finding scholarly information .

There are a number of useful differences from a regular Google search. Google Scholar allows you to:

• copy a formatted citation in different styles including MLA and APA
• export bibliographic data (BibTeX, RIS) to use with reference management software
• explore other works have cited the listed work
• easily find full text versions of the article

Although it is free to search in Google Scholar, most of the content is not freely available. Google does its best to find copies of restricted articles in public repositories. If you are at an academic or research institution, you can also set up a library connection that allows you to see items that are available through your institution.

The Google Scholar results page differs from the Google results page in a few key ways. The search result page is, however, different and it is worth being familiar with the different pieces of information that are shown. Let's have a look at the results for the search term "machine learning.”

• The first line of each result provides the title of the document (e.g. of an article, book, chapter, or report).
• The second line provides the bibliographic information about the document, in order: the author(s), the journal or book it appears in, the year of publication, and the publisher.

Clicking on the title link will bring you to the publisher’s page where you may be able to access more information about the document. This includes the abstract and options to download the PDF.

To the far right of the entry are more direct options for obtaining the full text of the document. In this example, Google has also located a publicly available PDF of the document hosted at umich.edu . Note, that it's not guaranteed that it is the version of the article that was finally published in the journal.

Below the text snippet/abstract you can find a number of useful links.

• Cited by : the cited by link will show other articles that have cited this resource. That is a super useful feature that can help you in many ways. First, it is a good way to track the more recent research that has referenced this article, and second the fact that other researches cited this document lends greater credibility to it. But be aware that there is a lag in publication type. Therefore, an article published in 2017 will not have an extensive number of cited by results. It takes a minimum of 6 months for most articles to get published, so even if an article was using the source, the more recent article has not been published yet.
• Versions : this link will display other versions of the article or other databases where the article may be found, some of which may offer free access to the article.
• Quotation mark icon : this will display a popup with commonly used citation formats such as MLA, APA, Chicago, Harvard, and Vancouver that may be copied and pasted. Note, however, that the Google Scholar citation data is sometimes incomplete and so it is often a good idea to check this data at the source. The "cite" popup also includes links for exporting the citation data as BibTeX or RIS files that any major reference manager can import.

Pro tip: Use a reference manager like Paperpile to keep track of all your sources. Paperpile integrates with Google Scholar and many popular academic research engines and databases, so you can save references and PDFs directly to your library using the Paperpile buttons and later cite them in thousands of citation styles:

Although Google Scholar limits each search to a maximum of 1,000 results , it's still too much to explore, and you need an effective way of locating the relevant articles. Here’s a list of pro tips that will help you save time and search more effectively.

You don’t need to worry about case sensitivity when you’re using Google scholar. In other words, a search for "Machine Learning" will produce the same results as a search for "machine learning.”

Let's say your research topic is about self driving cars. For a regular Google search we might enter something like " what is the current state of the technology used for self driving cars ". In Google Scholar, you will see less than ideal results for this query .

The trick is to build a list of keywords and perform searches for them like self-driving cars, autonomous vehicles, or driverless cars. Google Scholar will assist you on that: if you start typing in the search field you will see related queries suggested by Scholar!

If you put your search phrase into quotes you can search for exact matches of that phrase in the title and the body text of the document. Without quotes, Google Scholar will treat each word separately.

This means that if you search national parks , the words will not necessarily appear together. Grouped words and exact phrases should be enclosed in quotation marks.

A search using “self-driving cars 2015,” for example, will return articles or books published in 2015.

Using the options in the left hand panel you can further restrict the search results by limiting the years covered by the search, the inclusion or exclude of patents, and you can sort the results by relevance or by date.

Searches are not case sensitive, however, there are a number of Boolean operators you can use to control the search and these must be capitalized.

• AND requires both of the words or phrases on either side to be somewhere in the record.
• NOT can be placed in front of a word or phrases to exclude results which include them.
• OR will give equal weight to results which match just one of the words or phrases on either side.

➡️ Read more about how to efficiently search online databases for academic research .

In case you got overwhelmed by the above options, here’s some illustrative examples:

Tip: Use the advanced search features in Google Scholar to narrow down your search results.

You can gain even more fine-grained control over your search by using the advanced search feature. This feature is available by clicking on the hamburger menu in the upper left and selecting the "Advanced search" menu item.

Adjusting the Google Scholar settings is not necessary for getting good results, but offers some additional customization, including the ability to enable the above-mentioned library integrations.

The settings menu is found in the hamburger menu located in the top left of the Google Scholar page. The settings are divided into five sections:

• Collections to search: by default Google scholar searches articles and includes patents, but this default can be changed if you are not interested in patents or if you wish to search case law instead.
• Bibliographic manager: you can export relevant citation data via the “Bibliography manager” subsection.
• Languages: if you wish for results to return only articles written in a specific subset of languages, you can define that here.
• Library links: as noted, Google Scholar allows you to get the Full Text of articles through your institution’s subscriptions, where available. Search for, and add, your institution here to have the relevant link included in your search results.
• Button: the Scholar Button is a Chrome extension which adds a dropdown search box to your toolbar. This allows you to search Google Scholar from any website. Moreover, if you have any text selected on the page and then click the button it will display results from a search on those words when clicked.

When signed in, Google Scholar adds some simple tools for keeping track of and organizing the articles you find. These can be useful if you are not using a full academic reference manager.

All the search results include a “save” button at the end of the bottom row of links, clicking this will add it to your "My Library".

To help you provide some structure, you can create and apply labels to the items in your library. Appended labels will appear at the end of the article titles. For example, the following article has been assigned a “RNA” label:

Within your Google Scholar library, you can also edit the metadata associated with titles. This will often be necessary as Google Scholar citation data is often faulty.

There is no official statement about how big the Scholar search index is, but unofficial estimates are in the range of about 160 million , and it is supposed to continue to grow by several million each year.

Yet, Google Scholar does not return all resources that you may get in search at you local library catalog. For example, a library database could return podcasts, videos, articles, statistics, or special collections. For now, Google Scholar has only the following publication types:

• Journal articles : articles published in journals. It's a mixture of articles from peer reviewed journals, predatory journals and pre-print archives.
• Books : links to the Google limited version of the text, when possible.
• Book chapters : chapters within a book, sometimes they are also electronically available.
• Book reviews : reviews of books, but it is not always apparent that it is a review from the search result.
• Conference proceedings : papers written as part of a conference, typically used as part of presentation at the conference.
• Court opinions .
• Patents : Google Scholar only searches patents if the option is selected in the search settings described above.

The information in Google Scholar is not cataloged by professionals. The quality of the metadata will depend heavily on the source that Google Scholar is pulling the information from. This is a much different process to how information is collected and indexed in scholarly databases such as Scopus or Web of Science .

➡️ Visit our list of the best academic databases .

Google Scholar is by far the most frequently used academic search engine , but it is not the only one. Other academic search engines include:

• Science.gov
• Semantic Scholar
• scholar.google.fr : Sur les épaules d'un géant
• scholar.google.es (Google Académico): A hombros de gigantes
• scholar.google.pt (Google Académico): Sobre os ombros de gigantes
• scholar.google.de : Auf den Schultern von Riesen

➡️ Once you’ve found some research, it’s time to read it. Take a look at our guide on how to read a scientific paper .

No. Google Scholar is a bibliographic search engine rather than a bibliographic database. In order to qualify as a database Google Scholar would need to have stable identifiers for its records.

No. Google Scholar is an academic search engine, but the records found in Google Scholar are scholarly sources.

No. Google Scholar collects research papers from all over the web, including grey literature and non-peer reviewed papers and reports.

Google Scholar does not provide any full text content itself, but links to the full text article on the publisher page, which can either be open access or paywalled content. Google Scholar tries to provide links to free versions, when possible.

The easiest way to access Google scholar is by using The Google Scholar Button. This is a browser extension that allows you easily access Google Scholar from any web page. You can install it from the Chrome Webstore .

How to Write and Publish a Research Paper for a Peer-Reviewed Journal

• Open access
• Published: 30 April 2020
• Volume 36 , pages 909–913, ( 2021 )

Cite this article

You have full access to this open access article

• Clara Busse   ORCID: orcid.org/0000-0002-0178-1000 1 &
• Ella August   ORCID: orcid.org/0000-0001-5151-1036 1 , 2  

269k Accesses

15 Citations

719 Altmetric

Explore all metrics

Communicating research findings is an essential step in the research process. Often, peer-reviewed journals are the forum for such communication, yet many researchers are never taught how to write a publishable scientific paper. In this article, we explain the basic structure of a scientific paper and describe the information that should be included in each section. We also identify common pitfalls for each section and recommend strategies to avoid them. Further, we give advice about target journal selection and authorship. In the online resource 1 , we provide an example of a high-quality scientific paper, with annotations identifying the elements we describe in this article.

Similar content being viewed by others

How to design bibliometric research: an overview and a framework proposal

Literature reviews as independent studies: guidelines for academic practice

How metric-based performance evaluation systems fuel the growth of questionable publications?

Avoid common mistakes on your manuscript.

Introduction

Writing a scientific paper is an important component of the research process, yet researchers often receive little formal training in scientific writing. This is especially true in low-resource settings. In this article, we explain why choosing a target journal is important, give advice about authorship, provide a basic structure for writing each section of a scientific paper, and describe common pitfalls and recommendations for each section. In the online resource 1 , we also include an annotated journal article that identifies the key elements and writing approaches that we detail here. Before you begin your research, make sure you have ethical clearance from all relevant ethical review boards.

Select a Target Journal Early in the Writing Process

We recommend that you select a “target journal” early in the writing process; a “target journal” is the journal to which you plan to submit your paper. Each journal has a set of core readers and you should tailor your writing to this readership. For example, if you plan to submit a manuscript about vaping during pregnancy to a pregnancy-focused journal, you will need to explain what vaping is because readers of this journal may not have a background in this topic. However, if you were to submit that same article to a tobacco journal, you would not need to provide as much background information about vaping.

Information about a journal’s core readership can be found on its website, usually in a section called “About this journal” or something similar. For example, the Journal of Cancer Education presents such information on the “Aims and Scope” page of its website, which can be found here: https://www.springer.com/journal/13187/aims-and-scope .

Peer reviewer guidelines from your target journal are an additional resource that can help you tailor your writing to the journal and provide additional advice about crafting an effective article [ 1 ]. These are not always available, but it is worth a quick web search to find out.

Identify Author Roles Early in the Process

Early in the writing process, identify authors, determine the order of authors, and discuss the responsibilities of each author. Standard author responsibilities have been identified by The International Committee of Medical Journal Editors (ICMJE) [ 2 ]. To set clear expectations about each team member’s responsibilities and prevent errors in communication, we also suggest outlining more detailed roles, such as who will draft each section of the manuscript, write the abstract, submit the paper electronically, serve as corresponding author, and write the cover letter. It is best to formalize this agreement in writing after discussing it, circulating the document to the author team for approval. We suggest creating a title page on which all authors are listed in the agreed-upon order. It may be necessary to adjust authorship roles and order during the development of the paper. If a new author order is agreed upon, be sure to update the title page in the manuscript draft.

In the case where multiple papers will result from a single study, authors should discuss who will author each paper. Additionally, authors should agree on a deadline for each paper and the lead author should take responsibility for producing an initial draft by this deadline.

Structure of the Introduction Section

The introduction section should be approximately three to five paragraphs in length. Look at examples from your target journal to decide the appropriate length. This section should include the elements shown in Fig.  1 . Begin with a general context, narrowing to the specific focus of the paper. Include five main elements: why your research is important, what is already known about the topic, the “gap” or what is not yet known about the topic, why it is important to learn the new information that your research adds, and the specific research aim(s) that your paper addresses. Your research aim should address the gap you identified. Be sure to add enough background information to enable readers to understand your study. Table 1 provides common introduction section pitfalls and recommendations for addressing them.

The main elements of the introduction section of an original research article. Often, the elements overlap

Methods Section

The purpose of the methods section is twofold: to explain how the study was done in enough detail to enable its replication and to provide enough contextual detail to enable readers to understand and interpret the results. In general, the essential elements of a methods section are the following: a description of the setting and participants, the study design and timing, the recruitment and sampling, the data collection process, the dataset, the dependent and independent variables, the covariates, the analytic approach for each research objective, and the ethical approval. The hallmark of an exemplary methods section is the justification of why each method was used. Table 2 provides common methods section pitfalls and recommendations for addressing them.

Results Section

The focus of the results section should be associations, or lack thereof, rather than statistical tests. Two considerations should guide your writing here. First, the results should present answers to each part of the research aim. Second, return to the methods section to ensure that the analysis and variables for each result have been explained.

Begin the results section by describing the number of participants in the final sample and details such as the number who were approached to participate, the proportion who were eligible and who enrolled, and the number of participants who dropped out. The next part of the results should describe the participant characteristics. After that, you may organize your results by the aim or by putting the most exciting results first. Do not forget to report your non-significant associations. These are still findings.

Tables and figures capture the reader’s attention and efficiently communicate your main findings [ 3 ]. Each table and figure should have a clear message and should complement, rather than repeat, the text. Tables and figures should communicate all salient details necessary for a reader to understand the findings without consulting the text. Include information on comparisons and tests, as well as information about the sample and timing of the study in the title, legend, or in a footnote. Note that figures are often more visually interesting than tables, so if it is feasible to make a figure, make a figure. To avoid confusing the reader, either avoid abbreviations in tables and figures, or define them in a footnote. Note that there should not be citations in the results section and you should not interpret results here. Table 3 provides common results section pitfalls and recommendations for addressing them.

Discussion Section

Opposite the introduction section, the discussion should take the form of a right-side-up triangle beginning with interpretation of your results and moving to general implications (Fig.  2 ). This section typically begins with a restatement of the main findings, which can usually be accomplished with a few carefully-crafted sentences.

Major elements of the discussion section of an original research article. Often, the elements overlap

Next, interpret the meaning or explain the significance of your results, lifting the reader’s gaze from the study’s specific findings to more general applications. Then, compare these study findings with other research. Are these findings in agreement or disagreement with those from other studies? Does this study impart additional nuance to well-accepted theories? Situate your findings within the broader context of scientific literature, then explain the pathways or mechanisms that might give rise to, or explain, the results.

Journals vary in their approach to strengths and limitations sections: some are embedded paragraphs within the discussion section, while some mandate separate section headings. Keep in mind that every study has strengths and limitations. Candidly reporting yours helps readers to correctly interpret your research findings.

The next element of the discussion is a summary of the potential impacts and applications of the research. Should these results be used to optimally design an intervention? Does the work have implications for clinical protocols or public policy? These considerations will help the reader to further grasp the possible impacts of the presented work.

Finally, the discussion should conclude with specific suggestions for future work. Here, you have an opportunity to illuminate specific gaps in the literature that compel further study. Avoid the phrase “future research is necessary” because the recommendation is too general to be helpful to readers. Instead, provide substantive and specific recommendations for future studies. Table 4 provides common discussion section pitfalls and recommendations for addressing them.

Follow the Journal’s Author Guidelines

After you select a target journal, identify the journal’s author guidelines to guide the formatting of your manuscript and references. Author guidelines will often (but not always) include instructions for titles, cover letters, and other components of a manuscript submission. Read the guidelines carefully. If you do not follow the guidelines, your article will be sent back to you.

Finally, do not submit your paper to more than one journal at a time. Even if this is not explicitly stated in the author guidelines of your target journal, it is considered inappropriate and unprofessional.

Your title should invite readers to continue reading beyond the first page [ 4 , 5 ]. It should be informative and interesting. Consider describing the independent and dependent variables, the population and setting, the study design, the timing, and even the main result in your title. Because the focus of the paper can change as you write and revise, we recommend you wait until you have finished writing your paper before composing the title.

Be sure that the title is useful for potential readers searching for your topic. The keywords you select should complement those in your title to maximize the likelihood that a researcher will find your paper through a database search. Avoid using abbreviations in your title unless they are very well known, such as SNP, because it is more likely that someone will use a complete word rather than an abbreviation as a search term to help readers find your paper.

After you have written a complete draft, use the checklist (Fig. 3 ) below to guide your revisions and editing. Additional resources are available on writing the abstract and citing references [ 5 ]. When you feel that your work is ready, ask a trusted colleague or two to read the work and provide informal feedback. The box below provides a checklist that summarizes the key points offered in this article.

Checklist for manuscript quality

Data Availability

Michalek AM (2014) Down the rabbit hole…advice to reviewers. J Cancer Educ 29:4–5

Article   Google Scholar  

International Committee of Medical Journal Editors. Defining the role of authors and contributors: who is an author? http://www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-role-of-authosrs-and-contributors.html . Accessed 15 January, 2020

Vetto JT (2014) Short and sweet: a short course on concise medical writing. J Cancer Educ 29(1):194–195

Brett M, Kording K (2017) Ten simple rules for structuring papers. PLoS ComputBiol. https://doi.org/10.1371/journal.pcbi.1005619

Lang TA (2017) Writing a better research article. J Public Health Emerg. https://doi.org/10.21037/jphe.2017.11.06

Download references

Acknowledgments

Ella August is grateful to the Sustainable Sciences Institute for mentoring her in training researchers on writing and publishing their research.

Code Availability

Not applicable.

Author information

Authors and affiliations.

Department of Maternal and Child Health, University of North Carolina Gillings School of Global Public Health, 135 Dauer Dr, 27599, Chapel Hill, NC, USA

Clara Busse & Ella August

Department of Epidemiology, University of Michigan School of Public Health, 1415 Washington Heights, Ann Arbor, MI, 48109-2029, USA

Ella August

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Ella August .

Ethics declarations

Conflicts of interests.

The authors declare that they have no conflict of interest.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

(PDF 362 kb)

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Busse, C., August, E. How to Write and Publish a Research Paper for a Peer-Reviewed Journal. J Canc Educ 36 , 909–913 (2021). https://doi.org/10.1007/s13187-020-01751-z

Download citation

Published : 30 April 2020

Issue Date : October 2021

DOI : https://doi.org/10.1007/s13187-020-01751-z

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Manuscripts
• Scientific writing
• Find a journal
• Publish with us
• Track your research

• Google Quantum AI
• Research Publications

Research publications

Publishing our work allows us to share ideas and work collaboratively to advance the field of quantum computing.

• Open access
• Published: 18 April 2024

Research ethics and artificial intelligence for global health: perspectives from the global forum on bioethics in research

• James Shaw 1 , 13 ,
• Joseph Ali 2 , 3 ,
• Caesar A. Atuire 4 , 5 ,
• Phaik Yeong Cheah 6 ,
• Armando Guio Español 7 ,
• Judy Wawira Gichoya 8 ,
• Adrienne Hunt 9 ,
• Daudi Jjingo 10 ,
• Katherine Littler 9 ,
• Daniela Paolotti 11 &
• Effy Vayena 12  

BMC Medical Ethics volume  25 , Article number:  46 ( 2024 ) Cite this article

1145 Accesses

6 Altmetric

Metrics details

The ethical governance of Artificial Intelligence (AI) in health care and public health continues to be an urgent issue for attention in policy, research, and practice. In this paper we report on central themes related to challenges and strategies for promoting ethics in research involving AI in global health, arising from the Global Forum on Bioethics in Research (GFBR), held in Cape Town, South Africa in November 2022.

The GFBR is an annual meeting organized by the World Health Organization and supported by the Wellcome Trust, the US National Institutes of Health, the UK Medical Research Council (MRC) and the South African MRC. The forum aims to bring together ethicists, researchers, policymakers, research ethics committee members and other actors to engage with challenges and opportunities specifically related to research ethics. In 2022 the focus of the GFBR was “Ethics of AI in Global Health Research”. The forum consisted of 6 case study presentations, 16 governance presentations, and a series of small group and large group discussions. A total of 87 participants attended the forum from 31 countries around the world, representing disciplines of bioethics, AI, health policy, health professional practice, research funding, and bioinformatics. In this paper, we highlight central insights arising from GFBR 2022.

We describe the significance of four thematic insights arising from the forum: (1) Appropriateness of building AI, (2) Transferability of AI systems, (3) Accountability for AI decision-making and outcomes, and (4) Individual consent. We then describe eight recommendations for governance leaders to enhance the ethical governance of AI in global health research, addressing issues such as AI impact assessments, environmental values, and fair partnerships.

Conclusions

The 2022 Global Forum on Bioethics in Research illustrated several innovations in ethical governance of AI for global health research, as well as several areas in need of urgent attention internationally. This summary is intended to inform international and domestic efforts to strengthen research ethics and support the evolution of governance leadership to meet the demands of AI in global health research.

Peer Review reports

Introduction

The ethical governance of Artificial Intelligence (AI) in health care and public health continues to be an urgent issue for attention in policy, research, and practice [ 1 , 2 , 3 ]. Beyond the growing number of AI applications being implemented in health care, capabilities of AI models such as Large Language Models (LLMs) expand the potential reach and significance of AI technologies across health-related fields [ 4 , 5 ]. Discussion about effective, ethical governance of AI technologies has spanned a range of governance approaches, including government regulation, organizational decision-making, professional self-regulation, and research ethics review [ 6 , 7 , 8 ]. In this paper, we report on central themes related to challenges and strategies for promoting ethics in research involving AI in global health research, arising from the Global Forum on Bioethics in Research (GFBR), held in Cape Town, South Africa in November 2022. Although applications of AI for research, health care, and public health are diverse and advancing rapidly, the insights generated at the forum remain highly relevant from a global health perspective. After summarizing important context for work in this domain, we highlight categories of ethical issues emphasized at the forum for attention from a research ethics perspective internationally. We then outline strategies proposed for research, innovation, and governance to support more ethical AI for global health.

In this paper, we adopt the definition of AI systems provided by the Organization for Economic Cooperation and Development (OECD) as our starting point. Their definition states that an AI system is “a machine-based system that can, for a given set of human-defined objectives, make predictions, recommendations, or decisions influencing real or virtual environments. AI systems are designed to operate with varying levels of autonomy” [ 9 ]. The conceptualization of an algorithm as helping to constitute an AI system, along with hardware, other elements of software, and a particular context of use, illustrates the wide variety of ways in which AI can be applied. We have found it useful to differentiate applications of AI in research as those classified as “AI systems for discovery” and “AI systems for intervention”. An AI system for discovery is one that is intended to generate new knowledge, for example in drug discovery or public health research in which researchers are seeking potential targets for intervention, innovation, or further research. An AI system for intervention is one that directly contributes to enacting an intervention in a particular context, for example informing decision-making at the point of care or assisting with accuracy in a surgical procedure.

The mandate of the GFBR is to take a broad view of what constitutes research and its regulation in global health, with special attention to bioethics in Low- and Middle- Income Countries. AI as a group of technologies demands such a broad view. AI development for health occurs in a variety of environments, including universities and academic health sciences centers where research ethics review remains an important element of the governance of science and innovation internationally [ 10 , 11 ]. In these settings, research ethics committees (RECs; also known by different names such as Institutional Review Boards or IRBs) make decisions about the ethical appropriateness of projects proposed by researchers and other institutional members, ultimately determining whether a given project is allowed to proceed on ethical grounds [ 12 ].

However, research involving AI for health also takes place in large corporations and smaller scale start-ups, which in some jurisdictions fall outside the scope of research ethics regulation. In the domain of AI, the question of what constitutes research also becomes blurred. For example, is the development of an algorithm itself considered a part of the research process? Or only when that algorithm is tested under the formal constraints of a systematic research methodology? In this paper we take an inclusive view, in which AI development is included in the definition of research activity and within scope for our inquiry, regardless of the setting in which it takes place. This broad perspective characterizes the approach to “research ethics” we take in this paper, extending beyond the work of RECs to include the ethical analysis of the wide range of activities that constitute research as the generation of new knowledge and intervention in the world.

Ethical governance of AI in global health

The ethical governance of AI for global health has been widely discussed in recent years. The World Health Organization (WHO) released its guidelines on ethics and governance of AI for health in 2021, endorsing a set of six ethical principles and exploring the relevance of those principles through a variety of use cases. The WHO guidelines also provided an overview of AI governance, defining governance as covering “a range of steering and rule-making functions of governments and other decision-makers, including international health agencies, for the achievement of national health policy objectives conducive to universal health coverage.” (p. 81) The report usefully provided a series of recommendations related to governance of seven domains pertaining to AI for health: data, benefit sharing, the private sector, the public sector, regulation, policy observatories/model legislation, and global governance. The report acknowledges that much work is yet to be done to advance international cooperation on AI governance, especially related to prioritizing voices from Low- and Middle-Income Countries (LMICs) in global dialogue.

One important point emphasized in the WHO report that reinforces the broader literature on global governance of AI is the distribution of responsibility across a wide range of actors in the AI ecosystem. This is especially important to highlight when focused on research for global health, which is specifically about work that transcends national borders. Alami et al. (2020) discussed the unique risks raised by AI research in global health, ranging from the unavailability of data in many LMICs required to train locally relevant AI models to the capacity of health systems to absorb new AI technologies that demand the use of resources from elsewhere in the system. These observations illustrate the need to identify the unique issues posed by AI research for global health specifically, and the strategies that can be employed by all those implicated in AI governance to promote ethically responsible use of AI in global health research.

RECs and the regulation of research involving AI

RECs represent an important element of the governance of AI for global health research, and thus warrant further commentary as background to our paper. Despite the importance of RECs, foundational questions have been raised about their capabilities to accurately understand and address ethical issues raised by studies involving AI. Rahimzadeh et al. (2023) outlined how RECs in the United States are under-prepared to align with recent federal policy requiring that RECs review data sharing and management plans with attention to the unique ethical issues raised in AI research for health [ 13 ]. Similar research in South Africa identified variability in understanding of existing regulations and ethical issues associated with health-related big data sharing and management among research ethics committee members [ 14 , 15 ]. The effort to address harms accruing to groups or communities as opposed to individuals whose data are included in AI research has also been identified as a unique challenge for RECs [ 16 , 17 ]. Doerr and Meeder (2022) suggested that current regulatory frameworks for research ethics might actually prevent RECs from adequately addressing such issues, as they are deemed out of scope of REC review [ 16 ]. Furthermore, research in the United Kingdom and Canada has suggested that researchers using AI methods for health tend to distinguish between ethical issues and social impact of their research, adopting an overly narrow view of what constitutes ethical issues in their work [ 18 ].

The challenges for RECs in adequately addressing ethical issues in AI research for health care and public health exceed a straightforward survey of ethical considerations. As Ferretti et al. (2021) contend, some capabilities of RECs adequately cover certain issues in AI-based health research, such as the common occurrence of conflicts of interest where researchers who accept funds from commercial technology providers are implicitly incentivized to produce results that align with commercial interests [ 12 ]. However, some features of REC review require reform to adequately meet ethical needs. Ferretti et al. outlined weaknesses of RECs that are longstanding and those that are novel to AI-related projects, proposing a series of directions for development that are regulatory, procedural, and complementary to REC functionality. The work required on a global scale to update the REC function in response to the demands of research involving AI is substantial.

These issues take greater urgency in the context of global health [ 19 ]. Teixeira da Silva (2022) described the global practice of “ethics dumping”, where researchers from high income countries bring ethically contentious practices to RECs in low-income countries as a strategy to gain approval and move projects forward [ 20 ]. Although not yet systematically documented in AI research for health, risk of ethics dumping in AI research is high. Evidence is already emerging of practices of “health data colonialism”, in which AI researchers and developers from large organizations in high-income countries acquire data to build algorithms in LMICs to avoid stricter regulations [ 21 ]. This specific practice is part of a larger collection of practices that characterize health data colonialism, involving the broader exploitation of data and the populations they represent primarily for commercial gain [ 21 , 22 ]. As an additional complication, AI algorithms trained on data from high-income contexts are unlikely to apply in straightforward ways to LMIC settings [ 21 , 23 ]. In the context of global health, there is widespread acknowledgement about the need to not only enhance the knowledge base of REC members about AI-based methods internationally, but to acknowledge the broader shifts required to encourage their capabilities to more fully address these and other ethical issues associated with AI research for health [ 8 ].

Although RECs are an important part of the story of the ethical governance of AI for global health research, they are not the only part. The responsibilities of supra-national entities such as the World Health Organization, national governments, organizational leaders, commercial AI technology providers, health care professionals, and other groups continue to be worked out internationally. In this context of ongoing work, examining issues that demand attention and strategies to address them remains an urgent and valuable task.

The GFBR is an annual meeting organized by the World Health Organization and supported by the Wellcome Trust, the US National Institutes of Health, the UK Medical Research Council (MRC) and the South African MRC. The forum aims to bring together ethicists, researchers, policymakers, REC members and other actors to engage with challenges and opportunities specifically related to research ethics. Each year the GFBR meeting includes a series of case studies and keynotes presented in plenary format to an audience of approximately 100 people who have applied and been competitively selected to attend, along with small-group breakout discussions to advance thinking on related issues. The specific topic of the forum changes each year, with past topics including ethical issues in research with people living with mental health conditions (2021), genome editing (2019), and biobanking/data sharing (2018). The forum is intended to remain grounded in the practical challenges of engaging in research ethics, with special interest in low resource settings from a global health perspective. A post-meeting fellowship scheme is open to all LMIC participants, providing a unique opportunity to apply for funding to further explore and address the ethical challenges that are identified during the meeting.

In 2022, the focus of the GFBR was “Ethics of AI in Global Health Research”. The forum consisted of 6 case study presentations (both short and long form) reporting on specific initiatives related to research ethics and AI for health, and 16 governance presentations (both short and long form) reporting on actual approaches to governing AI in different country settings. A keynote presentation from Professor Effy Vayena addressed the topic of the broader context for AI ethics in a rapidly evolving field. A total of 87 participants attended the forum from 31 countries around the world, representing disciplines of bioethics, AI, health policy, health professional practice, research funding, and bioinformatics. The 2-day forum addressed a wide range of themes. The conference report provides a detailed overview of each of the specific topics addressed while a policy paper outlines the cross-cutting themes (both documents are available at the GFBR website: https://www.gfbr.global/past-meetings/16th-forum-cape-town-south-africa-29-30-november-2022/ ). As opposed to providing a detailed summary in this paper, we aim to briefly highlight central issues raised, solutions proposed, and the challenges facing the research ethics community in the years to come.

In this way, our primary aim in this paper is to present a synthesis of the challenges and opportunities raised at the GFBR meeting and in the planning process, followed by our reflections as a group of authors on their significance for governance leaders in the coming years. We acknowledge that the views represented at the meeting and in our results are a partial representation of the universe of views on this topic; however, the GFBR leadership invested a great deal of resources in convening a deeply diverse and thoughtful group of researchers and practitioners working on themes of bioethics related to AI for global health including those based in LMICs. We contend that it remains rare to convene such a strong group for an extended time and believe that many of the challenges and opportunities raised demand attention for more ethical futures of AI for health. Nonetheless, our results are primarily descriptive and are thus not explicitly grounded in a normative argument. We make effort in the Discussion section to contextualize our results by describing their significance and connecting them to broader efforts to reform global health research and practice.

Uniquely important ethical issues for AI in global health research

Presentations and group dialogue over the course of the forum raised several issues for consideration, and here we describe four overarching themes for the ethical governance of AI in global health research. Brief descriptions of each issue can be found in Table  1 . Reports referred to throughout the paper are available at the GFBR website provided above.

The first overarching thematic issue relates to the appropriateness of building AI technologies in response to health-related challenges in the first place. Case study presentations referred to initiatives where AI technologies were highly appropriate, such as in ear shape biometric identification to more accurately link electronic health care records to individual patients in Zambia (Alinani Simukanga). Although important ethical issues were raised with respect to privacy, trust, and community engagement in this initiative, the AI-based solution was appropriately matched to the challenge of accurately linking electronic records to specific patient identities. In contrast, forum participants raised questions about the appropriateness of an initiative using AI to improve the quality of handwashing practices in an acute care hospital in India (Niyoshi Shah), which led to gaming the algorithm. Overall, participants acknowledged the dangers of techno-solutionism, in which AI researchers and developers treat AI technologies as the most obvious solutions to problems that in actuality demand much more complex strategies to address [ 24 ]. However, forum participants agreed that RECs in different contexts have differing degrees of power to raise issues of the appropriateness of an AI-based intervention.

The second overarching thematic issue related to whether and how AI-based systems transfer from one national health context to another. One central issue raised by a number of case study presentations related to the challenges of validating an algorithm with data collected in a local environment. For example, one case study presentation described a project that would involve the collection of personally identifiable data for sensitive group identities, such as tribe, clan, or religion, in the jurisdictions involved (South Africa, Nigeria, Tanzania, Uganda and the US; Gakii Masunga). Doing so would enable the team to ensure that those groups were adequately represented in the dataset to ensure the resulting algorithm was not biased against specific community groups when deployed in that context. However, some members of these communities might desire to be represented in the dataset, whereas others might not, illustrating the need to balance autonomy and inclusivity. It was also widely recognized that collecting these data is an immense challenge, particularly when historically oppressive practices have led to a low-trust environment for international organizations and the technologies they produce. It is important to note that in some countries such as South Africa and Rwanda, it is illegal to collect information such as race and tribal identities, re-emphasizing the importance for cultural awareness and avoiding “one size fits all” solutions.

The third overarching thematic issue is related to understanding accountabilities for both the impacts of AI technologies and governance decision-making regarding their use. Where global health research involving AI leads to longer-term harms that might fall outside the usual scope of issues considered by a REC, who is to be held accountable, and how? This question was raised as one that requires much further attention, with law being mixed internationally regarding the mechanisms available to hold researchers, innovators, and their institutions accountable over the longer term. However, it was recognized in breakout group discussion that many jurisdictions are developing strong data protection regimes related specifically to international collaboration for research involving health data. For example, Kenya’s Data Protection Act requires that any internationally funded projects have a local principal investigator who will hold accountability for how data are shared and used [ 25 ]. The issue of research partnerships with commercial entities was raised by many participants in the context of accountability, pointing toward the urgent need for clear principles related to strategies for engagement with commercial technology companies in global health research.

The fourth and final overarching thematic issue raised here is that of consent. The issue of consent was framed by the widely shared recognition that models of individual, explicit consent might not produce a supportive environment for AI innovation that relies on the secondary uses of health-related datasets to build AI algorithms. Given this recognition, approaches such as community oversight of health data uses were suggested as a potential solution. However, the details of implementing such community oversight mechanisms require much further attention, particularly given the unique perspectives on health data in different country settings in global health research. Furthermore, some uses of health data do continue to require consent. One case study of South Africa, Nigeria, Kenya, Ethiopia and Uganda suggested that when health data are shared across borders, individual consent remains necessary when data is transferred from certain countries (Nezerith Cengiz). Broader clarity is necessary to support the ethical governance of health data uses for AI in global health research.

Recommendations for ethical governance of AI in global health research

Dialogue at the forum led to a range of suggestions for promoting ethical conduct of AI research for global health, related to the various roles of actors involved in the governance of AI research broadly defined. The strategies are written for actors we refer to as “governance leaders”, those people distributed throughout the AI for global health research ecosystem who are responsible for ensuring the ethical and socially responsible conduct of global health research involving AI (including researchers themselves). These include RECs, government regulators, health care leaders, health professionals, corporate social accountability officers, and others. Enacting these strategies would bolster the ethical governance of AI for global health more generally, enabling multiple actors to fulfill their roles related to governing research and development activities carried out across multiple organizations, including universities, academic health sciences centers, start-ups, and technology corporations. Specific suggestions are summarized in Table  2 .

First, forum participants suggested that governance leaders including RECs, should remain up to date on recent advances in the regulation of AI for health. Regulation of AI for health advances rapidly and takes on different forms in jurisdictions around the world. RECs play an important role in governance, but only a partial role; it was deemed important for RECs to acknowledge how they fit within a broader governance ecosystem in order to more effectively address the issues within their scope. Not only RECs but organizational leaders responsible for procurement, researchers, and commercial actors should all commit to efforts to remain up to date about the relevant approaches to regulating AI for health care and public health in jurisdictions internationally. In this way, governance can more adequately remain up to date with advances in regulation.

Second, forum participants suggested that governance leaders should focus on ethical governance of health data as a basis for ethical global health AI research. Health data are considered the foundation of AI development, being used to train AI algorithms for various uses [ 26 ]. By focusing on ethical governance of health data generation, sharing, and use, multiple actors will help to build an ethical foundation for AI development among global health researchers.

Third, forum participants believed that governance processes should incorporate AI impact assessments where appropriate. An AI impact assessment is the process of evaluating the potential effects, both positive and negative, of implementing an AI algorithm on individuals, society, and various stakeholders, generally over time frames specified in advance of implementation [ 27 ]. Although not all types of AI research in global health would warrant an AI impact assessment, this is especially relevant for those studies aiming to implement an AI system for intervention into health care or public health. Organizations such as RECs can use AI impact assessments to boost understanding of potential harms at the outset of a research project, encouraging researchers to more deeply consider potential harms in the development of their study.

Fourth, forum participants suggested that governance decisions should incorporate the use of environmental impact assessments, or at least the incorporation of environment values when assessing the potential impact of an AI system. An environmental impact assessment involves evaluating and anticipating the potential environmental effects of a proposed project to inform ethical decision-making that supports sustainability [ 28 ]. Although a relatively new consideration in research ethics conversations [ 29 ], the environmental impact of building technologies is a crucial consideration for the public health commitment to environmental sustainability. Governance leaders can use environmental impact assessments to boost understanding of potential environmental harms linked to AI research projects in global health over both the shorter and longer terms.

Fifth, forum participants suggested that governance leaders should require stronger transparency in the development of AI algorithms in global health research. Transparency was considered essential in the design and development of AI algorithms for global health to ensure ethical and accountable decision-making throughout the process. Furthermore, whether and how researchers have considered the unique contexts into which such algorithms may be deployed can be surfaced through stronger transparency, for example in describing what primary considerations were made at the outset of the project and which stakeholders were consulted along the way. Sharing information about data provenance and methods used in AI development will also enhance the trustworthiness of the AI-based research process.

Sixth, forum participants suggested that governance leaders can encourage or require community engagement at various points throughout an AI project. It was considered that engaging patients and communities is crucial in AI algorithm development to ensure that the technology aligns with community needs and values. However, participants acknowledged that this is not a straightforward process. Effective community engagement requires lengthy commitments to meeting with and hearing from diverse communities in a given setting, and demands a particular set of skills in communication and dialogue that are not possessed by all researchers. Encouraging AI researchers to begin this process early and build long-term partnerships with community members is a promising strategy to deepen community engagement in AI research for global health. One notable recommendation was that research funders have an opportunity to incentivize and enable community engagement with funds dedicated to these activities in AI research in global health.

Seventh, forum participants suggested that governance leaders can encourage researchers to build strong, fair partnerships between institutions and individuals across country settings. In a context of longstanding imbalances in geopolitical and economic power, fair partnerships in global health demand a priori commitments to share benefits related to advances in medical technologies, knowledge, and financial gains. Although enforcement of this point might be beyond the remit of RECs, commentary will encourage researchers to consider stronger, fairer partnerships in global health in the longer term.

Eighth, it became evident that it is necessary to explore new forms of regulatory experimentation given the complexity of regulating a technology of this nature. In addition, the health sector has a series of particularities that make it especially complicated to generate rules that have not been previously tested. Several participants highlighted the desire to promote spaces for experimentation such as regulatory sandboxes or innovation hubs in health. These spaces can have several benefits for addressing issues surrounding the regulation of AI in the health sector, such as: (i) increasing the capacities and knowledge of health authorities about this technology; (ii) identifying the major problems surrounding AI regulation in the health sector; (iii) establishing possibilities for exchange and learning with other authorities; (iv) promoting innovation and entrepreneurship in AI in health; and (vi) identifying the need to regulate AI in this sector and update other existing regulations.

Ninth and finally, forum participants believed that the capabilities of governance leaders need to evolve to better incorporate expertise related to AI in ways that make sense within a given jurisdiction. With respect to RECs, for example, it might not make sense for every REC to recruit a member with expertise in AI methods. Rather, it will make more sense in some jurisdictions to consult with members of the scientific community with expertise in AI when research protocols are submitted that demand such expertise. Furthermore, RECs and other approaches to research governance in jurisdictions around the world will need to evolve in order to adopt the suggestions outlined above, developing processes that apply specifically to the ethical governance of research using AI methods in global health.

Research involving the development and implementation of AI technologies continues to grow in global health, posing important challenges for ethical governance of AI in global health research around the world. In this paper we have summarized insights from the 2022 GFBR, focused specifically on issues in research ethics related to AI for global health research. We summarized four thematic challenges for governance related to AI in global health research and nine suggestions arising from presentations and dialogue at the forum. In this brief discussion section, we present an overarching observation about power imbalances that frames efforts to evolve the role of governance in global health research, and then outline two important opportunity areas as the field develops to meet the challenges of AI in global health research.

Dialogue about power is not unfamiliar in global health, especially given recent contributions exploring what it would mean to de-colonize global health research, funding, and practice [ 30 , 31 ]. Discussions of research ethics applied to AI research in global health contexts are deeply infused with power imbalances. The existing context of global health is one in which high-income countries primarily located in the “Global North” charitably invest in projects taking place primarily in the “Global South” while recouping knowledge, financial, and reputational benefits [ 32 ]. With respect to AI development in particular, recent examples of digital colonialism frame dialogue about global partnerships, raising attention to the role of large commercial entities and global financial capitalism in global health research [ 21 , 22 ]. Furthermore, the power of governance organizations such as RECs to intervene in the process of AI research in global health varies widely around the world, depending on the authorities assigned to them by domestic research governance policies. These observations frame the challenges outlined in our paper, highlighting the difficulties associated with making meaningful change in this field.

Despite these overarching challenges of the global health research context, there are clear strategies for progress in this domain. Firstly, AI innovation is rapidly evolving, which means approaches to the governance of AI for health are rapidly evolving too. Such rapid evolution presents an important opportunity for governance leaders to clarify their vision and influence over AI innovation in global health research, boosting the expertise, structure, and functionality required to meet the demands of research involving AI. Secondly, the research ethics community has strong international ties, linked to a global scholarly community that is committed to sharing insights and best practices around the world. This global community can be leveraged to coordinate efforts to produce advances in the capabilities and authorities of governance leaders to meaningfully govern AI research for global health given the challenges summarized in our paper.

Limitations

Our paper includes two specific limitations that we address explicitly here. First, it is still early in the lifetime of the development of applications of AI for use in global health, and as such, the global community has had limited opportunity to learn from experience. For example, there were many fewer case studies, which detail experiences with the actual implementation of an AI technology, submitted to GFBR 2022 for consideration than was expected. In contrast, there were many more governance reports submitted, which detail the processes and outputs of governance processes that anticipate the development and dissemination of AI technologies. This observation represents both a success and a challenge. It is a success that so many groups are engaging in anticipatory governance of AI technologies, exploring evidence of their likely impacts and governing technologies in novel and well-designed ways. It is a challenge that there is little experience to build upon of the successful implementation of AI technologies in ways that have limited harms while promoting innovation. Further experience with AI technologies in global health will contribute to revising and enhancing the challenges and recommendations we have outlined in our paper.

Second, global trends in the politics and economics of AI technologies are evolving rapidly. Although some nations are advancing detailed policy approaches to regulating AI more generally, including for uses in health care and public health, the impacts of corporate investments in AI and political responses related to governance remain to be seen. The excitement around large language models (LLMs) and large multimodal models (LMMs) has drawn deeper attention to the challenges of regulating AI in any general sense, opening dialogue about health sector-specific regulations. The direction of this global dialogue, strongly linked to high-profile corporate actors and multi-national governance institutions, will strongly influence the development of boundaries around what is possible for the ethical governance of AI for global health. We have written this paper at a point when these developments are proceeding rapidly, and as such, we acknowledge that our recommendations will need updating as the broader field evolves.

Ultimately, coordination and collaboration between many stakeholders in the research ethics ecosystem will be necessary to strengthen the ethical governance of AI in global health research. The 2022 GFBR illustrated several innovations in ethical governance of AI for global health research, as well as several areas in need of urgent attention internationally. This summary is intended to inform international and domestic efforts to strengthen research ethics and support the evolution of governance leadership to meet the demands of AI in global health research.

Data availability

All data and materials analyzed to produce this paper are available on the GFBR website: https://www.gfbr.global/past-meetings/16th-forum-cape-town-south-africa-29-30-november-2022/ .

Clark P, Kim J, Aphinyanaphongs Y, Marketing, Food US. Drug Administration Clearance of Artificial Intelligence and Machine Learning Enabled Software in and as Medical devices: a systematic review. JAMA Netw Open. 2023;6(7):e2321792–2321792.

Article   Google Scholar  

Potnis KC, Ross JS, Aneja S, Gross CP, Richman IB. Artificial intelligence in breast cancer screening: evaluation of FDA device regulation and future recommendations. JAMA Intern Med. 2022;182(12):1306–12.

Siala H, Wang Y. SHIFTing artificial intelligence to be responsible in healthcare: a systematic review. Soc Sci Med. 2022;296:114782.

Yang X, Chen A, PourNejatian N, Shin HC, Smith KE, Parisien C, et al. A large language model for electronic health records. NPJ Digit Med. 2022;5(1):194.

Meskó B, Topol EJ. The imperative for regulatory oversight of large language models (or generative AI) in healthcare. NPJ Digit Med. 2023;6(1):120.

Jobin A, Ienca M, Vayena E. The global landscape of AI ethics guidelines. Nat Mach Intell. 2019;1(9):389–99.

Minssen T, Vayena E, Cohen IG. The challenges for Regulating Medical Use of ChatGPT and other large Language models. JAMA. 2023.

Ho CWL, Malpani R. Scaling up the research ethics framework for healthcare machine learning as global health ethics and governance. Am J Bioeth. 2022;22(5):36–8.

Yeung K. Recommendation of the council on artificial intelligence (OECD). Int Leg Mater. 2020;59(1):27–34.

Maddox TM, Rumsfeld JS, Payne PR. Questions for artificial intelligence in health care. JAMA. 2019;321(1):31–2.

Dzau VJ, Balatbat CA, Ellaissi WF. Revisiting academic health sciences systems a decade later: discovery to health to population to society. Lancet. 2021;398(10318):2300–4.

Ferretti A, Ienca M, Sheehan M, Blasimme A, Dove ES, Farsides B, et al. Ethics review of big data research: what should stay and what should be reformed? BMC Med Ethics. 2021;22(1):1–13.

Rahimzadeh V, Serpico K, Gelinas L. Institutional review boards need new skills to review data sharing and management plans. Nat Med. 2023;1–3.

Kling S, Singh S, Burgess TL, Nair G. The role of an ethics advisory committee in data science research in sub-saharan Africa. South Afr J Sci. 2023;119(5–6):1–3.

Google Scholar  

Cengiz N, Kabanda SM, Esterhuizen TM, Moodley K. Exploring perspectives of research ethics committee members on the governance of big data in sub-saharan Africa. South Afr J Sci. 2023;119(5–6):1–9.

Doerr M, Meeder S. Big health data research and group harm: the scope of IRB review. Ethics Hum Res. 2022;44(4):34–8.

Ballantyne A, Stewart C. Big data and public-private partnerships in healthcare and research: the application of an ethics framework for big data in health and research. Asian Bioeth Rev. 2019;11(3):315–26.

Samuel G, Chubb J, Derrick G. Boundaries between research ethics and ethical research use in artificial intelligence health research. J Empir Res Hum Res Ethics. 2021;16(3):325–37.

Murphy K, Di Ruggiero E, Upshur R, Willison DJ, Malhotra N, Cai JC, et al. Artificial intelligence for good health: a scoping review of the ethics literature. BMC Med Ethics. 2021;22(1):1–17.

Teixeira da Silva JA. Handling ethics dumping and neo-colonial research: from the laboratory to the academic literature. J Bioethical Inq. 2022;19(3):433–43.

Ferryman K. The dangers of data colonialism in precision public health. Glob Policy. 2021;12:90–2.

Couldry N, Mejias UA. Data colonialism: rethinking big data’s relation to the contemporary subject. Telev New Media. 2019;20(4):336–49.

Organization WH. Ethics and governance of artificial intelligence for health: WHO guidance. 2021.

Metcalf J, Moss E. Owning ethics: corporate logics, silicon valley, and the institutionalization of ethics. Soc Res Int Q. 2019;86(2):449–76.

Data Protection Act - OFFICE OF THE DATA PROTECTION COMMISSIONER KENYA [Internet]. 2021 [cited 2023 Sep 30]. https://www.odpc.go.ke/dpa-act/ .

Sharon T, Lucivero F. Introduction to the special theme: the expansion of the health data ecosystem–rethinking data ethics and governance. Big Data & Society. Volume 6. London, England: SAGE Publications Sage UK; 2019. p. 2053951719852969.

Reisman D, Schultz J, Crawford K, Whittaker M. Algorithmic impact assessments: a practical Framework for Public Agency. AI Now. 2018.

Morgan RK. Environmental impact assessment: the state of the art. Impact Assess Proj Apprais. 2012;30(1):5–14.

Samuel G, Richie C. Reimagining research ethics to include environmental sustainability: a principled approach, including a case study of data-driven health research. J Med Ethics. 2023;49(6):428–33.

Kwete X, Tang K, Chen L, Ren R, Chen Q, Wu Z, et al. Decolonizing global health: what should be the target of this movement and where does it lead us? Glob Health Res Policy. 2022;7(1):3.

Abimbola S, Asthana S, Montenegro C, Guinto RR, Jumbam DT, Louskieter L, et al. Addressing power asymmetries in global health: imperatives in the wake of the COVID-19 pandemic. PLoS Med. 2021;18(4):e1003604.

Benatar S. Politics, power, poverty and global health: systems and frames. Int J Health Policy Manag. 2016;5(10):599.

Download references

Acknowledgements

We would like to acknowledge the outstanding contributions of the attendees of GFBR 2022 in Cape Town, South Africa. This paper is authored by members of the GFBR 2022 Planning Committee. We would like to acknowledge additional members Tamra Lysaght, National University of Singapore, and Niresh Bhagwandin, South African Medical Research Council, for their input during the planning stages and as reviewers of the applications to attend the Forum.

This work was supported by Wellcome [222525/Z/21/Z], the US National Institutes of Health, the UK Medical Research Council (part of UK Research and Innovation), and the South African Medical Research Council through funding to the Global Forum on Bioethics in Research.

Author information

Authors and affiliations.

Department of Physical Therapy, Temerty Faculty of Medicine, University of Toronto, Toronto, Canada

Berman Institute of Bioethics, Johns Hopkins University, Baltimore, MD, USA

Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD, USA

Department of Philosophy and Classics, University of Ghana, Legon-Accra, Ghana

Caesar A. Atuire

Centre for Tropical Medicine and Global Health, Nuffield Department of Medicine, University of Oxford, Oxford, UK

Mahidol Oxford Tropical Medicine Research Unit, Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand

Phaik Yeong Cheah

Berkman Klein Center, Harvard University, Bogotá, Colombia

Armando Guio Español

Department of Radiology and Informatics, Emory University School of Medicine, Atlanta, GA, USA

Judy Wawira Gichoya

Health Ethics & Governance Unit, Research for Health Department, Science Division, World Health Organization, Geneva, Switzerland

Adrienne Hunt & Katherine Littler

African Center of Excellence in Bioinformatics and Data Intensive Science, Infectious Diseases Institute, Makerere University, Kampala, Uganda

Daudi Jjingo

ISI Foundation, Turin, Italy

Daniela Paolotti

Department of Health Sciences and Technology, ETH Zurich, Zürich, Switzerland

Effy Vayena

Joint Centre for Bioethics, Dalla Lana School of Public Health, University of Toronto, Toronto, Canada

You can also search for this author in PubMed   Google Scholar

Contributions

JS led the writing, contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. JA contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. CA contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. PYC contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. AE contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. JWG contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. AH contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. DJ contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. KL contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. DP contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper. EV contributed to conceptualization and analysis, critically reviewed and provided feedback on drafts of this paper, and provided final approval of the paper.

Corresponding author

Correspondence to James Shaw .

Ethics declarations

Ethics approval and consent to participate.

Not applicable.

Consent for publication

Competing interests.

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Shaw, J., Ali, J., Atuire, C.A. et al. Research ethics and artificial intelligence for global health: perspectives from the global forum on bioethics in research. BMC Med Ethics 25 , 46 (2024). https://doi.org/10.1186/s12910-024-01044-w

Download citation

Received : 31 October 2023

Accepted : 01 April 2024

Published : 18 April 2024

DOI : https://doi.org/10.1186/s12910-024-01044-w

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Artificial intelligence
• Machine learning
• Research ethics
• Global health

BMC Medical Ethics

ISSN: 1472-6939

• General enquiries: [email protected]

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 17 April 2024

The economic commitment of climate change

• Maximilian Kotz   ORCID: orcid.org/0000-0003-2564-5043 1 , 2 ,
• Anders Levermann   ORCID: orcid.org/0000-0003-4432-4704 1 , 2 &
• Leonie Wenz   ORCID: orcid.org/0000-0002-8500-1568 1 , 3  

Nature volume  628 ,  pages 551–557 ( 2024 ) Cite this article

87k Accesses

3461 Altmetric

Metrics details

• Environmental economics
• Environmental health
• Interdisciplinary studies
• Projection and prediction

Global projections of macroeconomic climate-change damages typically consider impacts from average annual and national temperatures over long time horizons 1 , 2 , 3 , 4 , 5 , 6 . Here we use recent empirical findings from more than 1,600 regions worldwide over the past 40 years to project sub-national damages from temperature and precipitation, including daily variability and extremes 7 , 8 . Using an empirical approach that provides a robust lower bound on the persistence of impacts on economic growth, we find that the world economy is committed to an income reduction of 19% within the next 26 years independent of future emission choices (relative to a baseline without climate impacts, likely range of 11–29% accounting for physical climate and empirical uncertainty). These damages already outweigh the mitigation costs required to limit global warming to 2 °C by sixfold over this near-term time frame and thereafter diverge strongly dependent on emission choices. Committed damages arise predominantly through changes in average temperature, but accounting for further climatic components raises estimates by approximately 50% and leads to stronger regional heterogeneity. Committed losses are projected for all regions except those at very high latitudes, at which reductions in temperature variability bring benefits. The largest losses are committed at lower latitudes in regions with lower cumulative historical emissions and lower present-day income.

Similar content being viewed by others

Climate damage projections beyond annual temperature

Investment incentive reduced by climate damages can be restored by optimal policy

Climate economics support for the UN climate targets

Projections of the macroeconomic damage caused by future climate change are crucial to informing public and policy debates about adaptation, mitigation and climate justice. On the one hand, adaptation against climate impacts must be justified and planned on the basis of an understanding of their future magnitude and spatial distribution 9 . This is also of importance in the context of climate justice 10 , as well as to key societal actors, including governments, central banks and private businesses, which increasingly require the inclusion of climate risks in their macroeconomic forecasts to aid adaptive decision-making 11 , 12 . On the other hand, climate mitigation policy such as the Paris Climate Agreement is often evaluated by balancing the costs of its implementation against the benefits of avoiding projected physical damages. This evaluation occurs both formally through cost–benefit analyses 1 , 4 , 5 , 6 , as well as informally through public perception of mitigation and damage costs 13 .

Projections of future damages meet challenges when informing these debates, in particular the human biases relating to uncertainty and remoteness that are raised by long-term perspectives 14 . Here we aim to overcome such challenges by assessing the extent of economic damages from climate change to which the world is already committed by historical emissions and socio-economic inertia (the range of future emission scenarios that are considered socio-economically plausible 15 ). Such a focus on the near term limits the large uncertainties about diverging future emission trajectories, the resulting long-term climate response and the validity of applying historically observed climate–economic relations over long timescales during which socio-technical conditions may change considerably. As such, this focus aims to simplify the communication and maximize the credibility of projected economic damages from future climate change.

In projecting the future economic damages from climate change, we make use of recent advances in climate econometrics that provide evidence for impacts on sub-national economic growth from numerous components of the distribution of daily temperature and precipitation 3 , 7 , 8 . Using fixed-effects panel regression models to control for potential confounders, these studies exploit within-region variation in local temperature and precipitation in a panel of more than 1,600 regions worldwide, comprising climate and income data over the past 40 years, to identify the plausibly causal effects of changes in several climate variables on economic productivity 16 , 17 . Specifically, macroeconomic impacts have been identified from changing daily temperature variability, total annual precipitation, the annual number of wet days and extreme daily rainfall that occur in addition to those already identified from changing average temperature 2 , 3 , 18 . Moreover, regional heterogeneity in these effects based on the prevailing local climatic conditions has been found using interactions terms. The selection of these climate variables follows micro-level evidence for mechanisms related to the impacts of average temperatures on labour and agricultural productivity 2 , of temperature variability on agricultural productivity and health 7 , as well as of precipitation on agricultural productivity, labour outcomes and flood damages 8 (see Extended Data Table 1 for an overview, including more detailed references). References  7 , 8 contain a more detailed motivation for the use of these particular climate variables and provide extensive empirical tests about the robustness and nature of their effects on economic output, which are summarized in Methods . By accounting for these extra climatic variables at the sub-national level, we aim for a more comprehensive description of climate impacts with greater detail across both time and space.

Constraining the persistence of impacts

A key determinant and source of discrepancy in estimates of the magnitude of future climate damages is the extent to which the impact of a climate variable on economic growth rates persists. The two extreme cases in which these impacts persist indefinitely or only instantaneously are commonly referred to as growth or level effects 19 , 20 (see Methods section ‘Empirical model specification: fixed-effects distributed lag models’ for mathematical definitions). Recent work shows that future damages from climate change depend strongly on whether growth or level effects are assumed 20 . Following refs.  2 , 18 , we provide constraints on this persistence by using distributed lag models to test the significance of delayed effects separately for each climate variable. Notably, and in contrast to refs.  2 , 18 , we use climate variables in their first-differenced form following ref.  3 , implying a dependence of the growth rate on a change in climate variables. This choice means that a baseline specification without any lags constitutes a model prior of purely level effects, in which a permanent change in the climate has only an instantaneous effect on the growth rate 3 , 19 , 21 . By including lags, one can then test whether any effects may persist further. This is in contrast to the specification used by refs.  2 , 18 , in which climate variables are used without taking the first difference, implying a dependence of the growth rate on the level of climate variables. In this alternative case, the baseline specification without any lags constitutes a model prior of pure growth effects, in which a change in climate has an infinitely persistent effect on the growth rate. Consequently, including further lags in this alternative case tests whether the initial growth impact is recovered 18 , 19 , 21 . Both of these specifications suffer from the limiting possibility that, if too few lags are included, one might falsely accept the model prior. The limitations of including a very large number of lags, including loss of data and increasing statistical uncertainty with an increasing number of parameters, mean that such a possibility is likely. By choosing a specification in which the model prior is one of level effects, our approach is therefore conservative by design, avoiding assumptions of infinite persistence of climate impacts on growth and instead providing a lower bound on this persistence based on what is observable empirically (see Methods section ‘Empirical model specification: fixed-effects distributed lag models’ for further exposition of this framework). The conservative nature of such a choice is probably the reason that ref.  19 finds much greater consistency between the impacts projected by models that use the first difference of climate variables, as opposed to their levels.

We begin our empirical analysis of the persistence of climate impacts on growth using ten lags of the first-differenced climate variables in fixed-effects distributed lag models. We detect substantial effects on economic growth at time lags of up to approximately 8–10 years for the temperature terms and up to approximately 4 years for the precipitation terms (Extended Data Fig. 1 and Extended Data Table 2 ). Furthermore, evaluation by means of information criteria indicates that the inclusion of all five climate variables and the use of these numbers of lags provide a preferable trade-off between best-fitting the data and including further terms that could cause overfitting, in comparison with model specifications excluding climate variables or including more or fewer lags (Extended Data Fig. 3 , Supplementary Methods Section  1 and Supplementary Table 1 ). We therefore remove statistically insignificant terms at later lags (Supplementary Figs. 1 – 3 and Supplementary Tables 2 – 4 ). Further tests using Monte Carlo simulations demonstrate that the empirical models are robust to autocorrelation in the lagged climate variables (Supplementary Methods Section  2 and Supplementary Figs. 4 and 5 ), that information criteria provide an effective indicator for lag selection (Supplementary Methods Section  2 and Supplementary Fig. 6 ), that the results are robust to concerns of imperfect multicollinearity between climate variables and that including several climate variables is actually necessary to isolate their separate effects (Supplementary Methods Section  3 and Supplementary Fig. 7 ). We provide a further robustness check using a restricted distributed lag model to limit oscillations in the lagged parameter estimates that may result from autocorrelation, finding that it provides similar estimates of cumulative marginal effects to the unrestricted model (Supplementary Methods Section 4 and Supplementary Figs. 8 and 9 ). Finally, to explicitly account for any outstanding uncertainty arising from the precise choice of the number of lags, we include empirical models with marginally different numbers of lags in the error-sampling procedure of our projection of future damages. On the basis of the lag-selection procedure (the significance of lagged terms in Extended Data Fig. 1 and Extended Data Table 2 , as well as information criteria in Extended Data Fig. 3 ), we sample from models with eight to ten lags for temperature and four for precipitation (models shown in Supplementary Figs. 1 – 3 and Supplementary Tables 2 – 4 ). In summary, this empirical approach to constrain the persistence of climate impacts on economic growth rates is conservative by design in avoiding assumptions of infinite persistence, but nevertheless provides a lower bound on the extent of impact persistence that is robust to the numerous tests outlined above.

Committed damages until mid-century

We combine these empirical economic response functions (Supplementary Figs. 1 – 3 and Supplementary Tables 2 – 4 ) with an ensemble of 21 climate models (see Supplementary Table 5 ) from the Coupled Model Intercomparison Project Phase 6 (CMIP-6) 22 to project the macroeconomic damages from these components of physical climate change (see Methods for further details). Bias-adjusted climate models that provide a highly accurate reproduction of observed climatological patterns with limited uncertainty (Supplementary Table 6 ) are used to avoid introducing biases in the projections. Following a well-developed literature 2 , 3 , 19 , these projections do not aim to provide a prediction of future economic growth. Instead, they are a projection of the exogenous impact of future climate conditions on the economy relative to the baselines specified by socio-economic projections, based on the plausibly causal relationships inferred by the empirical models and assuming ceteris paribus. Other exogenous factors relevant for the prediction of economic output are purposefully assumed constant.

A Monte Carlo procedure that samples from climate model projections, empirical models with different numbers of lags and model parameter estimates (obtained by 1,000 block-bootstrap resamples of each of the regressions in Supplementary Figs. 1 – 3 and Supplementary Tables 2 – 4 ) is used to estimate the combined uncertainty from these sources. Given these uncertainty distributions, we find that projected global damages are statistically indistinguishable across the two most extreme emission scenarios until 2049 (at the 5% significance level; Fig. 1 ). As such, the climate damages occurring before this time constitute those to which the world is already committed owing to the combination of past emissions and the range of future emission scenarios that are considered socio-economically plausible 15 . These committed damages comprise a permanent income reduction of 19% on average globally (population-weighted average) in comparison with a baseline without climate-change impacts (with a likely range of 11–29%, following the likelihood classification adopted by the Intergovernmental Panel on Climate Change (IPCC); see caption of Fig. 1 ). Even though levels of income per capita generally still increase relative to those of today, this constitutes a permanent income reduction for most regions, including North America and Europe (each with median income reductions of approximately 11%) and with South Asia and Africa being the most strongly affected (each with median income reductions of approximately 22%; Fig. 1 ). Under a middle-of-the road scenario of future income development (SSP2, in which SSP stands for Shared Socio-economic Pathway), this corresponds to global annual damages in 2049 of 38 trillion in 2005 international dollars (likely range of 19–59 trillion 2005 international dollars). Compared with empirical specifications that assume pure growth or pure level effects, our preferred specification that provides a robust lower bound on the extent of climate impact persistence produces damages between these two extreme assumptions (Extended Data Fig. 3 ).

Estimates of the projected reduction in income per capita from changes in all climate variables based on empirical models of climate impacts on economic output with a robust lower bound on their persistence (Extended Data Fig. 1 ) under a low-emission scenario compatible with the 2 °C warming target and a high-emission scenario (SSP2-RCP2.6 and SSP5-RCP8.5, respectively) are shown in purple and orange, respectively. Shading represents the 34% and 10% confidence intervals reflecting the likely and very likely ranges, respectively (following the likelihood classification adopted by the IPCC), having estimated uncertainty from a Monte Carlo procedure, which samples the uncertainty from the choice of physical climate models, empirical models with different numbers of lags and bootstrapped estimates of the regression parameters shown in Supplementary Figs. 1 – 3 . Vertical dashed lines show the time at which the climate damages of the two emission scenarios diverge at the 5% and 1% significance levels based on the distribution of differences between emission scenarios arising from the uncertainty sampling discussed above. Note that uncertainty in the difference of the two scenarios is smaller than the combined uncertainty of the two respective scenarios because samples of the uncertainty (climate model and empirical model choice, as well as model parameter bootstrap) are consistent across the two emission scenarios, hence the divergence of damages occurs while the uncertainty bounds of the two separate damage scenarios still overlap. Estimates of global mitigation costs from the three IAMs that provide results for the SSP2 baseline and SSP2-RCP2.6 scenario are shown in light green in the top panel, with the median of these estimates shown in bold.

Damages already outweigh mitigation costs

We compare the damages to which the world is committed over the next 25 years to estimates of the mitigation costs required to achieve the Paris Climate Agreement. Taking estimates of mitigation costs from the three integrated assessment models (IAMs) in the IPCC AR6 database 23 that provide results under comparable scenarios (SSP2 baseline and SSP2-RCP2.6, in which RCP stands for Representative Concentration Pathway), we find that the median committed climate damages are larger than the median mitigation costs in 2050 (six trillion in 2005 international dollars) by a factor of approximately six (note that estimates of mitigation costs are only provided every 10 years by the IAMs and so a comparison in 2049 is not possible). This comparison simply aims to compare the magnitude of future damages against mitigation costs, rather than to conduct a formal cost–benefit analysis of transitioning from one emission path to another. Formal cost–benefit analyses typically find that the net benefits of mitigation only emerge after 2050 (ref.  5 ), which may lead some to conclude that physical damages from climate change are simply not large enough to outweigh mitigation costs until the second half of the century. Our simple comparison of their magnitudes makes clear that damages are actually already considerably larger than mitigation costs and the delayed emergence of net mitigation benefits results primarily from the fact that damages across different emission paths are indistinguishable until mid-century (Fig. 1 ).

Although these near-term damages constitute those to which the world is already committed, we note that damage estimates diverge strongly across emission scenarios after 2049, conveying the clear benefits of mitigation from a purely economic point of view that have been emphasized in previous studies 4 , 24 . As well as the uncertainties assessed in Fig. 1 , these conclusions are robust to structural choices, such as the timescale with which changes in the moderating variables of the empirical models are estimated (Supplementary Figs. 10 and 11 ), as well as the order in which one accounts for the intertemporal and international components of currency comparison (Supplementary Fig. 12 ; see Methods for further details).

Damages from variability and extremes

Committed damages primarily arise through changes in average temperature (Fig. 2 ). This reflects the fact that projected changes in average temperature are larger than those in other climate variables when expressed as a function of their historical interannual variability (Extended Data Fig. 4 ). Because the historical variability is that on which the empirical models are estimated, larger projected changes in comparison with this variability probably lead to larger future impacts in a purely statistical sense. From a mechanistic perspective, one may plausibly interpret this result as implying that future changes in average temperature are the most unprecedented from the perspective of the historical fluctuations to which the economy is accustomed and therefore will cause the most damage. This insight may prove useful in terms of guiding adaptation measures to the sources of greatest damage.

Estimates of the median projected reduction in sub-national income per capita across emission scenarios (SSP2-RCP2.6 and SSP2-RCP8.5) as well as climate model, empirical model and model parameter uncertainty in the year in which climate damages diverge at the 5% level (2049, as identified in Fig. 1 ). a , Impacts arising from all climate variables. b – f , Impacts arising separately from changes in annual mean temperature ( b ), daily temperature variability ( c ), total annual precipitation ( d ), the annual number of wet days (>1 mm) ( e ) and extreme daily rainfall ( f ) (see Methods for further definitions). Data on national administrative boundaries are obtained from the GADM database version 3.6 and are freely available for academic use ( https://gadm.org/ ).

Nevertheless, future damages based on empirical models that consider changes in annual average temperature only and exclude the other climate variables constitute income reductions of only 13% in 2049 (Extended Data Fig. 5a , likely range 5–21%). This suggests that accounting for the other components of the distribution of temperature and precipitation raises net damages by nearly 50%. This increase arises through the further damages that these climatic components cause, but also because their inclusion reveals a stronger negative economic response to average temperatures (Extended Data Fig. 5b ). The latter finding is consistent with our Monte Carlo simulations, which suggest that the magnitude of the effect of average temperature on economic growth is underestimated unless accounting for the impacts of other correlated climate variables (Supplementary Fig. 7 ).

In terms of the relative contributions of the different climatic components to overall damages, we find that accounting for daily temperature variability causes the largest increase in overall damages relative to empirical frameworks that only consider changes in annual average temperature (4.9 percentage points, likely range 2.4–8.7 percentage points, equivalent to approximately 10 trillion international dollars). Accounting for precipitation causes smaller increases in overall damages, which are—nevertheless—equivalent to approximately 1.2 trillion international dollars: 0.01 percentage points (−0.37–0.33 percentage points), 0.34 percentage points (0.07–0.90 percentage points) and 0.36 percentage points (0.13–0.65 percentage points) from total annual precipitation, the number of wet days and extreme daily precipitation, respectively. Moreover, climate models seem to underestimate future changes in temperature variability 25 and extreme precipitation 26 , 27 in response to anthropogenic forcing as compared with that observed historically, suggesting that the true impacts from these variables may be larger.

The distribution of committed damages

The spatial distribution of committed damages (Fig. 2a ) reflects a complex interplay between the patterns of future change in several climatic components and those of historical economic vulnerability to changes in those variables. Damages resulting from increasing annual mean temperature (Fig. 2b ) are negative almost everywhere globally, and larger at lower latitudes in regions in which temperatures are already higher and economic vulnerability to temperature increases is greatest (see the response heterogeneity to mean temperature embodied in Extended Data Fig. 1a ). This occurs despite the amplified warming projected at higher latitudes 28 , suggesting that regional heterogeneity in economic vulnerability to temperature changes outweighs heterogeneity in the magnitude of future warming (Supplementary Fig. 13a ). Economic damages owing to daily temperature variability (Fig. 2c ) exhibit a strong latitudinal polarisation, primarily reflecting the physical response of daily variability to greenhouse forcing in which increases in variability across lower latitudes (and Europe) contrast decreases at high latitudes 25 (Supplementary Fig. 13b ). These two temperature terms are the dominant determinants of the pattern of overall damages (Fig. 2a ), which exhibits a strong polarity with damages across most of the globe except at the highest northern latitudes. Future changes in total annual precipitation mainly bring economic benefits except in regions of drying, such as the Mediterranean and central South America (Fig. 2d and Supplementary Fig. 13c ), but these benefits are opposed by changes in the number of wet days, which produce damages with a similar pattern of opposite sign (Fig. 2e and Supplementary Fig. 13d ). By contrast, changes in extreme daily rainfall produce damages in all regions, reflecting the intensification of daily rainfall extremes over global land areas 29 , 30 (Fig. 2f and Supplementary Fig. 13e ).

The spatial distribution of committed damages implies considerable injustice along two dimensions: culpability for the historical emissions that have caused climate change and pre-existing levels of socio-economic welfare. Spearman’s rank correlations indicate that committed damages are significantly larger in countries with smaller historical cumulative emissions, as well as in regions with lower current income per capita (Fig. 3 ). This implies that those countries that will suffer the most from the damages already committed are those that are least responsible for climate change and which also have the least resources to adapt to it.

Estimates of the median projected change in national income per capita across emission scenarios (RCP2.6 and RCP8.5) as well as climate model, empirical model and model parameter uncertainty in the year in which climate damages diverge at the 5% level (2049, as identified in Fig. 1 ) are plotted against cumulative national emissions per capita in 2020 (from the Global Carbon Project) and coloured by national income per capita in 2020 (from the World Bank) in a and vice versa in b . In each panel, the size of each scatter point is weighted by the national population in 2020 (from the World Bank). Inset numbers indicate the Spearman’s rank correlation ρ and P -values for a hypothesis test whose null hypothesis is of no correlation, as well as the Spearman’s rank correlation weighted by national population.

To further quantify this heterogeneity, we assess the difference in committed damages between the upper and lower quartiles of regions when ranked by present income levels and historical cumulative emissions (using a population weighting to both define the quartiles and estimate the group averages). On average, the quartile of countries with lower income are committed to an income loss that is 8.9 percentage points (or 61%) greater than the upper quartile (Extended Data Fig. 6 ), with a likely range of 3.8–14.7 percentage points across the uncertainty sampling of our damage projections (following the likelihood classification adopted by the IPCC). Similarly, the quartile of countries with lower historical cumulative emissions are committed to an income loss that is 6.9 percentage points (or 40%) greater than the upper quartile, with a likely range of 0.27–12 percentage points. These patterns reemphasize the prevalence of injustice in climate impacts 31 , 32 , 33 in the context of the damages to which the world is already committed by historical emissions and socio-economic inertia.

Contextualizing the magnitude of damages

The magnitude of projected economic damages exceeds previous literature estimates 2 , 3 , arising from several developments made on previous approaches. Our estimates are larger than those of ref.  2 (see first row of Extended Data Table 3 ), primarily because of the facts that sub-national estimates typically show a steeper temperature response (see also refs.  3 , 34 ) and that accounting for other climatic components raises damage estimates (Extended Data Fig. 5 ). However, we note that our empirical approach using first-differenced climate variables is conservative compared with that of ref.  2 in regard to the persistence of climate impacts on growth (see introduction and Methods section ‘Empirical model specification: fixed-effects distributed lag models’), an important determinant of the magnitude of long-term damages 19 , 21 . Using a similar empirical specification to ref.  2 , which assumes infinite persistence while maintaining the rest of our approach (sub-national data and further climate variables), produces considerably larger damages (purple curve of Extended Data Fig. 3 ). Compared with studies that do take the first difference of climate variables 3 , 35 , our estimates are also larger (see second and third rows of Extended Data Table 3 ). The inclusion of further climate variables (Extended Data Fig. 5 ) and a sufficient number of lags to more adequately capture the extent of impact persistence (Extended Data Figs. 1 and 2 ) are the main sources of this difference, as is the use of specifications that capture nonlinearities in the temperature response when compared with ref.  35 . In summary, our estimates develop on previous studies by incorporating the latest data and empirical insights 7 , 8 , as well as in providing a robust empirical lower bound on the persistence of impacts on economic growth, which constitutes a middle ground between the extremes of the growth-versus-levels debate 19 , 21 (Extended Data Fig. 3 ).

Compared with the fraction of variance explained by the empirical models historically (<5%), the projection of reductions in income of 19% may seem large. This arises owing to the fact that projected changes in climatic conditions are much larger than those that were experienced historically, particularly for changes in average temperature (Extended Data Fig. 4 ). As such, any assessment of future climate-change impacts necessarily requires an extrapolation outside the range of the historical data on which the empirical impact models were evaluated. Nevertheless, these models constitute the most state-of-the-art methods for inference of plausibly causal climate impacts based on observed data. Moreover, we take explicit steps to limit out-of-sample extrapolation by capping the moderating variables of the interaction terms at the 95th percentile of the historical distribution (see Methods ). This avoids extrapolating the marginal effects outside what was observed historically. Given the nonlinear response of economic output to annual mean temperature (Extended Data Fig. 1 and Extended Data Table 2 ), this is a conservative choice that limits the magnitude of damages that we project. Furthermore, back-of-the-envelope calculations indicate that the projected damages are consistent with the magnitude and patterns of historical economic development (see Supplementary Discussion Section  5 ).

Missing impacts and spatial spillovers

Despite assessing several climatic components from which economic impacts have recently been identified 3 , 7 , 8 , this assessment of aggregate climate damages should not be considered comprehensive. Important channels such as impacts from heatwaves 31 , sea-level rise 36 , tropical cyclones 37 and tipping points 38 , 39 , as well as non-market damages such as those to ecosystems 40 and human health 41 , are not considered in these estimates. Sea-level rise is unlikely to be feasibly incorporated into empirical assessments such as this because historical sea-level variability is mostly small. Non-market damages are inherently intractable within our estimates of impacts on aggregate monetary output and estimates of these impacts could arguably be considered as extra to those identified here. Recent empirical work suggests that accounting for these channels would probably raise estimates of these committed damages, with larger damages continuing to arise in the global south 31 , 36 , 37 , 38 , 39 , 40 , 41 , 42 .

Moreover, our main empirical analysis does not explicitly evaluate the potential for impacts in local regions to produce effects that ‘spill over’ into other regions. Such effects may further mitigate or amplify the impacts we estimate, for example, if companies relocate production from one affected region to another or if impacts propagate along supply chains. The current literature indicates that trade plays a substantial role in propagating spillover effects 43 , 44 , making their assessment at the sub-national level challenging without available data on sub-national trade dependencies. Studies accounting for only spatially adjacent neighbours indicate that negative impacts in one region induce further negative impacts in neighbouring regions 45 , 46 , 47 , 48 , suggesting that our projected damages are probably conservative by excluding these effects. In Supplementary Fig. 14 , we assess spillovers from neighbouring regions using a spatial-lag model. For simplicity, this analysis excludes temporal lags, focusing only on contemporaneous effects. The results show that accounting for spatial spillovers can amplify the overall magnitude, and also the heterogeneity, of impacts. Consistent with previous literature, this indicates that the overall magnitude (Fig. 1 ) and heterogeneity (Fig. 3 ) of damages that we project in our main specification may be conservative without explicitly accounting for spillovers. We note that further analysis that addresses both spatially and trade-connected spillovers, while also accounting for delayed impacts using temporal lags, would be necessary to adequately address this question fully. These approaches offer fruitful avenues for further research but are beyond the scope of this manuscript, which primarily aims to explore the impacts of different climate conditions and their persistence.

Policy implications

We find that the economic damages resulting from climate change until 2049 are those to which the world economy is already committed and that these greatly outweigh the costs required to mitigate emissions in line with the 2 °C target of the Paris Climate Agreement (Fig. 1 ). This assessment is complementary to formal analyses of the net costs and benefits associated with moving from one emission path to another, which typically find that net benefits of mitigation only emerge in the second half of the century 5 . Our simple comparison of the magnitude of damages and mitigation costs makes clear that this is primarily because damages are indistinguishable across emissions scenarios—that is, committed—until mid-century (Fig. 1 ) and that they are actually already much larger than mitigation costs. For simplicity, and owing to the availability of data, we compare damages to mitigation costs at the global level. Regional estimates of mitigation costs may shed further light on the national incentives for mitigation to which our results already hint, of relevance for international climate policy. Although these damages are committed from a mitigation perspective, adaptation may provide an opportunity to reduce them. Moreover, the strong divergence of damages after mid-century reemphasizes the clear benefits of mitigation from a purely economic perspective, as highlighted in previous studies 1 , 4 , 6 , 24 .

Historical climate data

Historical daily 2-m temperature and precipitation totals (in mm) are obtained for the period 1979–2019 from the W5E5 database. The W5E5 dataset comes from ERA-5, a state-of-the-art reanalysis of historical observations, but has been bias-adjusted by applying version 2.0 of the WATCH Forcing Data to ERA-5 reanalysis data and precipitation data from version 2.3 of the Global Precipitation Climatology Project to better reflect ground-based measurements 49 , 50 , 51 . We obtain these data on a 0.5° × 0.5° grid from the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP) database. Notably, these historical data have been used to bias-adjust future climate projections from CMIP-6 (see the following section), ensuring consistency between the distribution of historical daily weather on which our empirical models were estimated and the climate projections used to estimate future damages. These data are publicly available from the ISIMIP database. See refs.  7 , 8 for robustness tests of the empirical models to the choice of climate data reanalysis products.

Future climate data

Daily 2-m temperature and precipitation totals (in mm) are taken from 21 climate models participating in CMIP-6 under a high (RCP8.5) and a low (RCP2.6) greenhouse gas emission scenario from 2015 to 2100. The data have been bias-adjusted and statistically downscaled to a common half-degree grid to reflect the historical distribution of daily temperature and precipitation of the W5E5 dataset using the trend-preserving method developed by the ISIMIP 50 , 52 . As such, the climate model data reproduce observed climatological patterns exceptionally well (Supplementary Table 5 ). Gridded data are publicly available from the ISIMIP database.

Historical economic data

Historical economic data come from the DOSE database of sub-national economic output 53 . We use a recent revision to the DOSE dataset that provides data across 83 countries, 1,660 sub-national regions with varying temporal coverage from 1960 to 2019. Sub-national units constitute the first administrative division below national, for example, states for the USA and provinces for China. Data come from measures of gross regional product per capita (GRPpc) or income per capita in local currencies, reflecting the values reported in national statistical agencies, yearbooks and, in some cases, academic literature. We follow previous literature 3 , 7 , 8 , 54 and assess real sub-national output per capita by first converting values from local currencies to US dollars to account for diverging national inflationary tendencies and then account for US inflation using a US deflator. Alternatively, one might first account for national inflation and then convert between currencies. Supplementary Fig. 12 demonstrates that our conclusions are consistent when accounting for price changes in the reversed order, although the magnitude of estimated damages varies. See the documentation of the DOSE dataset for further discussion of these choices. Conversions between currencies are conducted using exchange rates from the FRED database of the Federal Reserve Bank of St. Louis 55 and the national deflators from the World Bank 56 .

Future socio-economic data

Baseline gridded gross domestic product (GDP) and population data for the period 2015–2100 are taken from the middle-of-the-road scenario SSP2 (ref.  15 ). Population data have been downscaled to a half-degree grid by the ISIMIP following the methodologies of refs.  57 , 58 , which we then aggregate to the sub-national level of our economic data using the spatial aggregation procedure described below. Because current methodologies for downscaling the GDP of the SSPs use downscaled population to do so, per-capita estimates of GDP with a realistic distribution at the sub-national level are not readily available for the SSPs. We therefore use national-level GDP per capita (GDPpc) projections for all sub-national regions of a given country, assuming homogeneity within countries in terms of baseline GDPpc. Here we use projections that have been updated to account for the impact of the COVID-19 pandemic on the trajectory of future income, while remaining consistent with the long-term development of the SSPs 59 . The choice of baseline SSP alters the magnitude of projected climate damages in monetary terms, but when assessed in terms of percentage change from the baseline, the choice of socio-economic scenario is inconsequential. Gridded SSP population data and national-level GDPpc data are publicly available from the ISIMIP database. Sub-national estimates as used in this study are available in the code and data replication files.

Climate variables

Following recent literature 3 , 7 , 8 , we calculate an array of climate variables for which substantial impacts on macroeconomic output have been identified empirically, supported by further evidence at the micro level for plausible underlying mechanisms. See refs.  7 , 8 for an extensive motivation for the use of these particular climate variables and for detailed empirical tests on the nature and robustness of their effects on economic output. To summarize, these studies have found evidence for independent impacts on economic growth rates from annual average temperature, daily temperature variability, total annual precipitation, the annual number of wet days and extreme daily rainfall. Assessments of daily temperature variability were motivated by evidence of impacts on agricultural output and human health, as well as macroeconomic literature on the impacts of volatility on growth when manifest in different dimensions, such as government spending, exchange rates and even output itself 7 . Assessments of precipitation impacts were motivated by evidence of impacts on agricultural productivity, metropolitan labour outcomes and conflict, as well as damages caused by flash flooding 8 . See Extended Data Table 1 for detailed references to empirical studies of these physical mechanisms. Marked impacts of daily temperature variability, total annual precipitation, the number of wet days and extreme daily rainfall on macroeconomic output were identified robustly across different climate datasets, spatial aggregation schemes, specifications of regional time trends and error-clustering approaches. They were also found to be robust to the consideration of temperature extremes 7 , 8 . Furthermore, these climate variables were identified as having independent effects on economic output 7 , 8 , which we further explain here using Monte Carlo simulations to demonstrate the robustness of the results to concerns of imperfect multicollinearity between climate variables (Supplementary Methods Section  2 ), as well as by using information criteria (Supplementary Table 1 ) to demonstrate that including several lagged climate variables provides a preferable trade-off between optimally describing the data and limiting the possibility of overfitting.

We calculate these variables from the distribution of daily, d , temperature, T x , d , and precipitation, P x , d , at the grid-cell, x , level for both the historical and future climate data. As well as annual mean temperature, \({\bar{T}}_{x,y}\) , and annual total precipitation, P x , y , we calculate annual, y , measures of daily temperature variability, \({\widetilde{T}}_{x,y}\) :

the number of wet days, Pwd x , y :

and extreme daily rainfall:

in which T x , d , m , y is the grid-cell-specific daily temperature in month m and year y , \({\bar{T}}_{x,m,{y}}\) is the year and grid-cell-specific monthly, m , mean temperature, D m and D y the number of days in a given month m or year y , respectively, H the Heaviside step function, 1 mm the threshold used to define wet days and P 99.9 x is the 99.9th percentile of historical (1979–2019) daily precipitation at the grid-cell level. Units of the climate measures are degrees Celsius for annual mean temperature and daily temperature variability, millimetres for total annual precipitation and extreme daily precipitation, and simply the number of days for the annual number of wet days.

We also calculated weighted standard deviations of monthly rainfall totals as also used in ref.  8 but do not include them in our projections as we find that, when accounting for delayed effects, their effect becomes statistically indistinct and is better captured by changes in total annual rainfall.

Spatial aggregation

We aggregate grid-cell-level historical and future climate measures, as well as grid-cell-level future GDPpc and population, to the level of the first administrative unit below national level of the GADM database, using an area-weighting algorithm that estimates the portion of each grid cell falling within an administrative boundary. We use this as our baseline specification following previous findings that the effect of area or population weighting at the sub-national level is negligible 7 , 8 .

Empirical model specification: fixed-effects distributed lag models

Following a wide range of climate econometric literature 16 , 60 , we use panel regression models with a selection of fixed effects and time trends to isolate plausibly exogenous variation with which to maximize confidence in a causal interpretation of the effects of climate on economic growth rates. The use of region fixed effects, μ r , accounts for unobserved time-invariant differences between regions, such as prevailing climatic norms and growth rates owing to historical and geopolitical factors. The use of yearly fixed effects, η y , accounts for regionally invariant annual shocks to the global climate or economy such as the El Niño–Southern Oscillation or global recessions. In our baseline specification, we also include region-specific linear time trends, k r y , to exclude the possibility of spurious correlations resulting from common slow-moving trends in climate and growth.

The persistence of climate impacts on economic growth rates is a key determinant of the long-term magnitude of damages. Methods for inferring the extent of persistence in impacts on growth rates have typically used lagged climate variables to evaluate the presence of delayed effects or catch-up dynamics 2 , 18 . For example, consider starting from a model in which a climate condition, C r , y , (for example, annual mean temperature) affects the growth rate, Δlgrp r , y (the first difference of the logarithm of gross regional product) of region r in year y :

which we refer to as a ‘pure growth effects’ model in the main text. Typically, further lags are included,

and the cumulative effect of all lagged terms is evaluated to assess the extent to which climate impacts on growth rates persist. Following ref.  18 , in the case that,

the implication is that impacts on the growth rate persist up to NL years after the initial shock (possibly to a weaker or a stronger extent), whereas if

then the initial impact on the growth rate is recovered after NL years and the effect is only one on the level of output. However, we note that such approaches are limited by the fact that, when including an insufficient number of lags to detect a recovery of the growth rates, one may find equation ( 6 ) to be satisfied and incorrectly assume that a change in climatic conditions affects the growth rate indefinitely. In practice, given a limited record of historical data, including too few lags to confidently conclude in an infinitely persistent impact on the growth rate is likely, particularly over the long timescales over which future climate damages are often projected 2 , 24 . To avoid this issue, we instead begin our analysis with a model for which the level of output, lgrp r , y , depends on the level of a climate variable, C r , y :

Given the non-stationarity of the level of output, we follow the literature 19 and estimate such an equation in first-differenced form as,

which we refer to as a model of ‘pure level effects’ in the main text. This model constitutes a baseline specification in which a permanent change in the climate variable produces an instantaneous impact on the growth rate and a permanent effect only on the level of output. By including lagged variables in this specification,

we are able to test whether the impacts on the growth rate persist any further than instantaneously by evaluating whether α L  > 0 are statistically significantly different from zero. Even though this framework is also limited by the possibility of including too few lags, the choice of a baseline model specification in which impacts on the growth rate do not persist means that, in the case of including too few lags, the framework reverts to the baseline specification of level effects. As such, this framework is conservative with respect to the persistence of impacts and the magnitude of future damages. It naturally avoids assumptions of infinite persistence and we are able to interpret any persistence that we identify with equation ( 9 ) as a lower bound on the extent of climate impact persistence on growth rates. See the main text for further discussion of this specification choice, in particular about its conservative nature compared with previous literature estimates, such as refs.  2 , 18 .

We allow the response to climatic changes to vary across regions, using interactions of the climate variables with historical average (1979–2019) climatic conditions reflecting heterogenous effects identified in previous work 7 , 8 . Following this previous work, the moderating variables of these interaction terms constitute the historical average of either the variable itself or of the seasonal temperature difference, \({\hat{T}}_{r}\) , or annual mean temperature, \({\bar{T}}_{r}\) , in the case of daily temperature variability 7 and extreme daily rainfall, respectively 8 .

The resulting regression equation with N and M lagged variables, respectively, reads:

in which Δlgrp r , y is the annual, regional GRPpc growth rate, measured as the first difference of the logarithm of real GRPpc, following previous work 2 , 3 , 7 , 8 , 18 , 19 . Fixed-effects regressions were run using the fixest package in R (ref.  61 ).

Estimates of the coefficients of interest α i , L are shown in Extended Data Fig. 1 for N  =  M  = 10 lags and for our preferred choice of the number of lags in Supplementary Figs. 1 – 3 . In Extended Data Fig. 1 , errors are shown clustered at the regional level, but for the construction of damage projections, we block-bootstrap the regressions by region 1,000 times to provide a range of parameter estimates with which to sample the projection uncertainty (following refs.  2 , 31 ).

Spatial-lag model

In Supplementary Fig. 14 , we present the results from a spatial-lag model that explores the potential for climate impacts to ‘spill over’ into spatially neighbouring regions. We measure the distance between centroids of each pair of sub-national regions and construct spatial lags that take the average of the first-differenced climate variables and their interaction terms over neighbouring regions that are at distances of 0–500, 500–1,000, 1,000–1,500 and 1,500–2000 km (spatial lags, ‘SL’, 1 to 4). For simplicity, we then assess a spatial-lag model without temporal lags to assess spatial spillovers of contemporaneous climate impacts. This model takes the form:

in which SL indicates the spatial lag of each climate variable and interaction term. In Supplementary Fig. 14 , we plot the cumulative marginal effect of each climate variable at different baseline climate conditions by summing the coefficients for each climate variable and interaction term, for example, for average temperature impacts as:

These cumulative marginal effects can be regarded as the overall spatially dependent impact to an individual region given a one-unit shock to a climate variable in that region and all neighbouring regions at a given value of the moderating variable of the interaction term.

Constructing projections of economic damage from future climate change

We construct projections of future climate damages by applying the coefficients estimated in equation ( 10 ) and shown in Supplementary Tables 2 – 4 (when including only lags with statistically significant effects in specifications that limit overfitting; see Supplementary Methods Section  1 ) to projections of future climate change from the CMIP-6 models. Year-on-year changes in each primary climate variable of interest are calculated to reflect the year-to-year variations used in the empirical models. 30-year moving averages of the moderating variables of the interaction terms are calculated to reflect the long-term average of climatic conditions that were used for the moderating variables in the empirical models. By using moving averages in the projections, we account for the changing vulnerability to climate shocks based on the evolving long-term conditions (Supplementary Figs. 10 and 11 show that the results are robust to the precise choice of the window of this moving average). Although these climate variables are not differenced, the fact that the bias-adjusted climate models reproduce observed climatological patterns across regions for these moderating variables very accurately (Supplementary Table 6 ) with limited spread across models (<3%) precludes the possibility that any considerable bias or uncertainty is introduced by this methodological choice. However, we impose caps on these moderating variables at the 95th percentile at which they were observed in the historical data to prevent extrapolation of the marginal effects outside the range in which the regressions were estimated. This is a conservative choice that limits the magnitude of our damage projections.

Time series of primary climate variables and moderating climate variables are then combined with estimates of the empirical model parameters to evaluate the regression coefficients in equation ( 10 ), producing a time series of annual GRPpc growth-rate reductions for a given emission scenario, climate model and set of empirical model parameters. The resulting time series of growth-rate impacts reflects those occurring owing to future climate change. By contrast, a future scenario with no climate change would be one in which climate variables do not change (other than with random year-to-year fluctuations) and hence the time-averaged evaluation of equation ( 10 ) would be zero. Our approach therefore implicitly compares the future climate-change scenario to this no-climate-change baseline scenario.

The time series of growth-rate impacts owing to future climate change in region r and year y , δ r , y , are then added to the future baseline growth rates, π r , y (in log-diff form), obtained from the SSP2 scenario to yield trajectories of damaged GRPpc growth rates, ρ r , y . These trajectories are aggregated over time to estimate the future trajectory of GRPpc with future climate impacts:

in which GRPpc r , y =2020 is the initial log level of GRPpc. We begin damage estimates in 2020 to reflect the damages occurring since the end of the period for which we estimate the empirical models (1979–2019) and to match the timing of mitigation-cost estimates from most IAMs (see below).

For each emission scenario, this procedure is repeated 1,000 times while randomly sampling from the selection of climate models, the selection of empirical models with different numbers of lags (shown in Supplementary Figs. 1 – 3 and Supplementary Tables 2 – 4 ) and bootstrapped estimates of the regression parameters. The result is an ensemble of future GRPpc trajectories that reflect uncertainty from both physical climate change and the structural and sampling uncertainty of the empirical models.

Estimates of mitigation costs

We obtain IPCC estimates of the aggregate costs of emission mitigation from the AR6 Scenario Explorer and Database hosted by IIASA 23 . Specifically, we search the AR6 Scenarios Database World v1.1 for IAMs that provided estimates of global GDP and population under both a SSP2 baseline and a SSP2-RCP2.6 scenario to maintain consistency with the socio-economic and emission scenarios of the climate damage projections. We find five IAMs that provide data for these scenarios, namely, MESSAGE-GLOBIOM 1.0, REMIND-MAgPIE 1.5, AIM/GCE 2.0, GCAM 4.2 and WITCH-GLOBIOM 3.1. Of these five IAMs, we use the results only from the first three that passed the IPCC vetting procedure for reproducing historical emission and climate trajectories. We then estimate global mitigation costs as the percentage difference in global per capita GDP between the SSP2 baseline and the SSP2-RCP2.6 emission scenario. In the case of one of these IAMs, estimates of mitigation costs begin in 2020, whereas in the case of two others, mitigation costs begin in 2010. The mitigation cost estimates before 2020 in these two IAMs are mostly negligible, and our choice to begin comparison with damage estimates in 2020 is conservative with respect to the relative weight of climate damages compared with mitigation costs for these two IAMs.

Data availability

Data on economic production and ERA-5 climate data are publicly available at https://doi.org/10.5281/zenodo.4681306 (ref. 62 ) and https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5 , respectively. Data on mitigation costs are publicly available at https://data.ene.iiasa.ac.at/ar6/#/downloads . Processed climate and economic data, as well as all other necessary data for reproduction of the results, are available at the public repository https://doi.org/10.5281/zenodo.10562951  (ref. 63 ).

Code availability

All code necessary for reproduction of the results is available at the public repository https://doi.org/10.5281/zenodo.10562951  (ref. 63 ).

Glanemann, N., Willner, S. N. & Levermann, A. Paris Climate Agreement passes the cost-benefit test. Nat. Commun. 11 , 110 (2020).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Burke, M., Hsiang, S. M. & Miguel, E. Global non-linear effect of temperature on economic production. Nature 527 , 235–239 (2015).

Article   ADS   CAS   PubMed   Google Scholar  

Kalkuhl, M. & Wenz, L. The impact of climate conditions on economic production. Evidence from a global panel of regions. J. Environ. Econ. Manag. 103 , 102360 (2020).

Article   Google Scholar  

Moore, F. C. & Diaz, D. B. Temperature impacts on economic growth warrant stringent mitigation policy. Nat. Clim. Change 5 , 127–131 (2015).

Article   ADS   Google Scholar  

Drouet, L., Bosetti, V. & Tavoni, M. Net economic benefits of well-below 2°C scenarios and associated uncertainties. Oxf. Open Clim. Change 2 , kgac003 (2022).

Ueckerdt, F. et al. The economically optimal warming limit of the planet. Earth Syst. Dyn. 10 , 741–763 (2019).

Kotz, M., Wenz, L., Stechemesser, A., Kalkuhl, M. & Levermann, A. Day-to-day temperature variability reduces economic growth. Nat. Clim. Change 11 , 319–325 (2021).

Kotz, M., Levermann, A. & Wenz, L. The effect of rainfall changes on economic production. Nature 601 , 223–227 (2022).

Kousky, C. Informing climate adaptation: a review of the economic costs of natural disasters. Energy Econ. 46 , 576–592 (2014).

Harlan, S. L. et al. in Climate Change and Society: Sociological Perspectives (eds Dunlap, R. E. & Brulle, R. J.) 127–163 (Oxford Univ. Press, 2015).

Bolton, P. et al. The Green Swan (BIS Books, 2020).

Alogoskoufis, S. et al. ECB Economy-wide Climate Stress Test: Methodology and Results European Central Bank, 2021).

Weber, E. U. What shapes perceptions of climate change? Wiley Interdiscip. Rev. Clim. Change 1 , 332–342 (2010).

Markowitz, E. M. & Shariff, A. F. Climate change and moral judgement. Nat. Clim. Change 2 , 243–247 (2012).

Riahi, K. et al. The shared socioeconomic pathways and their energy, land use, and greenhouse gas emissions implications: an overview. Glob. Environ. Change 42 , 153–168 (2017).

Auffhammer, M., Hsiang, S. M., Schlenker, W. & Sobel, A. Using weather data and climate model output in economic analyses of climate change. Rev. Environ. Econ. Policy 7 , 181–198 (2013).

Kolstad, C. D. & Moore, F. C. Estimating the economic impacts of climate change using weather observations. Rev. Environ. Econ. Policy 14 , 1–24 (2020).

Dell, M., Jones, B. F. & Olken, B. A. Temperature shocks and economic growth: evidence from the last half century. Am. Econ. J. Macroecon. 4 , 66–95 (2012).

Newell, R. G., Prest, B. C. & Sexton, S. E. The GDP-temperature relationship: implications for climate change damages. J. Environ. Econ. Manag. 108 , 102445 (2021).

Kikstra, J. S. et al. The social cost of carbon dioxide under climate-economy feedbacks and temperature variability. Environ. Res. Lett. 16 , 094037 (2021).

Article   ADS   CAS   Google Scholar  

Bastien-Olvera, B. & Moore, F. Persistent effect of temperature on GDP identified from lower frequency temperature variability. Environ. Res. Lett. 17 , 084038 (2022).

Eyring, V. et al. Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev. 9 , 1937–1958 (2016).

Byers, E. et al. AR6 scenarios database. Zenodo https://zenodo.org/records/7197970 (2022).

Burke, M., Davis, W. M. & Diffenbaugh, N. S. Large potential reduction in economic damages under UN mitigation targets. Nature 557 , 549–553 (2018).

Kotz, M., Wenz, L. & Levermann, A. Footprint of greenhouse forcing in daily temperature variability. Proc. Natl Acad. Sci. 118 , e2103294118 (2021).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Myhre, G. et al. Frequency of extreme precipitation increases extensively with event rareness under global warming. Sci. Rep. 9 , 16063 (2019).

Min, S.-K., Zhang, X., Zwiers, F. W. & Hegerl, G. C. Human contribution to more-intense precipitation extremes. Nature 470 , 378–381 (2011).

England, M. R., Eisenman, I., Lutsko, N. J. & Wagner, T. J. The recent emergence of Arctic Amplification. Geophys. Res. Lett. 48 , e2021GL094086 (2021).

Fischer, E. M. & Knutti, R. Anthropogenic contribution to global occurrence of heavy-precipitation and high-temperature extremes. Nat. Clim. Change 5 , 560–564 (2015).

Pfahl, S., O’Gorman, P. A. & Fischer, E. M. Understanding the regional pattern of projected future changes in extreme precipitation. Nat. Clim. Change 7 , 423–427 (2017).

Callahan, C. W. & Mankin, J. S. Globally unequal effect of extreme heat on economic growth. Sci. Adv. 8 , eadd3726 (2022).

Diffenbaugh, N. S. & Burke, M. Global warming has increased global economic inequality. Proc. Natl Acad. Sci. 116 , 9808–9813 (2019).

Callahan, C. W. & Mankin, J. S. National attribution of historical climate damages. Clim. Change 172 , 40 (2022).

Burke, M. & Tanutama, V. Climatic constraints on aggregate economic output. National Bureau of Economic Research, Working Paper 25779. https://doi.org/10.3386/w25779 (2019).

Kahn, M. E. et al. Long-term macroeconomic effects of climate change: a cross-country analysis. Energy Econ. 104 , 105624 (2021).

Desmet, K. et al. Evaluating the economic cost of coastal flooding. National Bureau of Economic Research, Working Paper 24918. https://doi.org/10.3386/w24918 (2018).

Hsiang, S. M. & Jina, A. S. The causal effect of environmental catastrophe on long-run economic growth: evidence from 6,700 cyclones. National Bureau of Economic Research, Working Paper 20352. https://doi.org/10.3386/w2035 (2014).

Ritchie, P. D. et al. Shifts in national land use and food production in Great Britain after a climate tipping point. Nat. Food 1 , 76–83 (2020).

Dietz, S., Rising, J., Stoerk, T. & Wagner, G. Economic impacts of tipping points in the climate system. Proc. Natl Acad. Sci. 118 , e2103081118 (2021).

Bastien-Olvera, B. A. & Moore, F. C. Use and non-use value of nature and the social cost of carbon. Nat. Sustain. 4 , 101–108 (2021).

Carleton, T. et al. Valuing the global mortality consequences of climate change accounting for adaptation costs and benefits. Q. J. Econ. 137 , 2037–2105 (2022).

Bastien-Olvera, B. A. et al. Unequal climate impacts on global values of natural capital. Nature 625 , 722–727 (2024).

Malik, A. et al. Impacts of climate change and extreme weather on food supply chains cascade across sectors and regions in Australia. Nat. Food 3 , 631–643 (2022).

Article   ADS   PubMed   Google Scholar  

Kuhla, K., Willner, S. N., Otto, C., Geiger, T. & Levermann, A. Ripple resonance amplifies economic welfare loss from weather extremes. Environ. Res. Lett. 16 , 114010 (2021).

Schleypen, J. R., Mistry, M. N., Saeed, F. & Dasgupta, S. Sharing the burden: quantifying climate change spillovers in the European Union under the Paris Agreement. Spat. Econ. Anal. 17 , 67–82 (2022).

Dasgupta, S., Bosello, F., De Cian, E. & Mistry, M. Global temperature effects on economic activity and equity: a spatial analysis. European Institute on Economics and the Environment, Working Paper 22-1 (2022).

Neal, T. The importance of external weather effects in projecting the macroeconomic impacts of climate change. UNSW Economics Working Paper 2023-09 (2023).

Deryugina, T. & Hsiang, S. M. Does the environment still matter? Daily temperature and income in the United States. National Bureau of Economic Research, Working Paper 20750. https://doi.org/10.3386/w20750 (2014).

Hersbach, H. et al. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc. 146 , 1999–2049 (2020).

Cucchi, M. et al. WFDE5: bias-adjusted ERA5 reanalysis data for impact studies. Earth Syst. Sci. Data 12 , 2097–2120 (2020).

Adler, R. et al. The New Version 2.3 of the Global Precipitation Climatology Project (GPCP) Monthly Analysis Product 1072–1084 (University of Maryland, 2016).

Lange, S. Trend-preserving bias adjustment and statistical downscaling with ISIMIP3BASD (v1.0). Geosci. Model Dev. 12 , 3055–3070 (2019).

Wenz, L., Carr, R. D., Kögel, N., Kotz, M. & Kalkuhl, M. DOSE – global data set of reported sub-national economic output. Sci. Data 10 , 425 (2023).

Article   PubMed   PubMed Central   Google Scholar  

Gennaioli, N., La Porta, R., Lopez De Silanes, F. & Shleifer, A. Growth in regions. J. Econ. Growth 19 , 259–309 (2014).

Board of Governors of the Federal Reserve System (US). U.S. dollars to euro spot exchange rate. https://fred.stlouisfed.org/series/AEXUSEU (2022).

World Bank. GDP deflator. https://data.worldbank.org/indicator/NY.GDP.DEFL.ZS (2022).

Jones, B. & O’Neill, B. C. Spatially explicit global population scenarios consistent with the Shared Socioeconomic Pathways. Environ. Res. Lett. 11 , 084003 (2016).

Murakami, D. & Yamagata, Y. Estimation of gridded population and GDP scenarios with spatially explicit statistical downscaling. Sustainability 11 , 2106 (2019).

Koch, J. & Leimbach, M. Update of SSP GDP projections: capturing recent changes in national accounting, PPP conversion and Covid 19 impacts. Ecol. Econ. 206 (2023).

Carleton, T. A. & Hsiang, S. M. Social and economic impacts of climate. Science 353 , aad9837 (2016).

Article   PubMed   Google Scholar  

Bergé, L. Efficient estimation of maximum likelihood models with multiple fixed-effects: the R package FENmlm. DEM Discussion Paper Series 18-13 (2018).

Kalkuhl, M., Kotz, M. & Wenz, L. DOSE - The MCC-PIK Database Of Subnational Economic output. Zenodo https://zenodo.org/doi/10.5281/zenodo.4681305 (2021).

Kotz, M., Wenz, L. & Levermann, A. Data and code for “The economic commitment of climate change”. Zenodo https://zenodo.org/doi/10.5281/zenodo.10562951 (2024).

Dasgupta, S. et al. Effects of climate change on combined labour productivity and supply: an empirical, multi-model study. Lancet Planet. Health 5 , e455–e465 (2021).

Lobell, D. B. et al. The critical role of extreme heat for maize production in the United States. Nat. Clim. Change 3 , 497–501 (2013).

Zhao, C. et al. Temperature increase reduces global yields of major crops in four independent estimates. Proc. Natl Acad. Sci. 114 , 9326–9331 (2017).

Wheeler, T. R., Craufurd, P. Q., Ellis, R. H., Porter, J. R. & Prasad, P. V. Temperature variability and the yield of annual crops. Agric. Ecosyst. Environ. 82 , 159–167 (2000).

Rowhani, P., Lobell, D. B., Linderman, M. & Ramankutty, N. Climate variability and crop production in Tanzania. Agric. For. Meteorol. 151 , 449–460 (2011).

Ceglar, A., Toreti, A., Lecerf, R., Van der Velde, M. & Dentener, F. Impact of meteorological drivers on regional inter-annual crop yield variability in France. Agric. For. Meteorol. 216 , 58–67 (2016).

Shi, L., Kloog, I., Zanobetti, A., Liu, P. & Schwartz, J. D. Impacts of temperature and its variability on mortality in New England. Nat. Clim. Change 5 , 988–991 (2015).

Xue, T., Zhu, T., Zheng, Y. & Zhang, Q. Declines in mental health associated with air pollution and temperature variability in China. Nat. Commun. 10 , 2165 (2019).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Liang, X.-Z. et al. Determining climate effects on US total agricultural productivity. Proc. Natl Acad. Sci. 114 , E2285–E2292 (2017).

Desbureaux, S. & Rodella, A.-S. Drought in the city: the economic impact of water scarcity in Latin American metropolitan areas. World Dev. 114 , 13–27 (2019).

Damania, R. The economics of water scarcity and variability. Oxf. Rev. Econ. Policy 36 , 24–44 (2020).

Davenport, F. V., Burke, M. & Diffenbaugh, N. S. Contribution of historical precipitation change to US flood damages. Proc. Natl Acad. Sci. 118 , e2017524118 (2021).

Dave, R., Subramanian, S. S. & Bhatia, U. Extreme precipitation induced concurrent events trigger prolonged disruptions in regional road networks. Environ. Res. Lett. 16 , 104050 (2021).

Download references

Acknowledgements

We gratefully acknowledge financing from the Volkswagen Foundation and the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH on behalf of the Government of the Federal Republic of Germany and Federal Ministry for Economic Cooperation and Development (BMZ).

Open access funding provided by Potsdam-Institut für Klimafolgenforschung (PIK) e.V.

Author information

Authors and affiliations.

Research Domain IV, Research Domain IV, Potsdam Institute for Climate Impact Research, Potsdam, Germany

Maximilian Kotz, Anders Levermann & Leonie Wenz

Institute of Physics, Potsdam University, Potsdam, Germany

Maximilian Kotz & Anders Levermann

Mercator Research Institute on Global Commons and Climate Change, Berlin, Germany

Leonie Wenz

You can also search for this author in PubMed   Google Scholar

Contributions

All authors contributed to the design of the analysis. M.K. conducted the analysis and produced the figures. All authors contributed to the interpretation and presentation of the results. M.K. and L.W. wrote the manuscript.

Corresponding author

Correspondence to Leonie Wenz .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

Nature thanks Xin-Zhong Liang, Chad Thackeray and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended data fig. 1 constraining the persistence of historical climate impacts on economic growth rates..

The results of a panel-based fixed-effects distributed lag model for the effects of annual mean temperature ( a ), daily temperature variability ( b ), total annual precipitation ( c ), the number of wet days ( d ) and extreme daily precipitation ( e ) on sub-national economic growth rates. Point estimates show the effects of a 1 °C or one standard deviation increase (for temperature and precipitation variables, respectively) at the lower quartile, median and upper quartile of the relevant moderating variable (green, orange and purple, respectively) at different lagged periods after the initial shock (note that these are not cumulative effects). Climate variables are used in their first-differenced form (see main text for discussion) and the moderating climate variables are the annual mean temperature, seasonal temperature difference, total annual precipitation, number of wet days and annual mean temperature, respectively, in panels a – e (see Methods for further discussion). Error bars show the 95% confidence intervals having clustered standard errors by region. The within-region R 2 , Bayesian and Akaike information criteria for the model are shown at the top of the figure. This figure shows results with ten lags for each variable to demonstrate the observed levels of persistence, but our preferred specifications remove later lags based on the statistical significance of terms shown above and the information criteria shown in Extended Data Fig. 2 . The resulting models without later lags are shown in Supplementary Figs. 1 – 3 .

Extended Data Fig. 2 Incremental lag-selection procedure using information criteria and within-region R 2 .

Starting from a panel-based fixed-effects distributed lag model estimating the effects of climate on economic growth using the real historical data (as in equation ( 4 )) with ten lags for all climate variables (as shown in Extended Data Fig. 1 ), lags are incrementally removed for one climate variable at a time. The resulting Bayesian and Akaike information criteria are shown in a – e and f – j , respectively, and the within-region R 2 and number of observations in k – o and p – t , respectively. Different rows show the results when removing lags from different climate variables, ordered from top to bottom as annual mean temperature, daily temperature variability, total annual precipitation, the number of wet days and extreme annual precipitation. Information criteria show minima at approximately four lags for precipitation variables and ten to eight for temperature variables, indicating that including these numbers of lags does not lead to overfitting. See Supplementary Table 1 for an assessment using information criteria to determine whether including further climate variables causes overfitting.

Extended Data Fig. 3 Damages in our preferred specification that provides a robust lower bound on the persistence of climate impacts on economic growth versus damages in specifications of pure growth or pure level effects.

Estimates of future damages as shown in Fig. 1 but under the emission scenario RCP8.5 for three separate empirical specifications: in orange our preferred specification, which provides an empirical lower bound on the persistence of climate impacts on economic growth rates while avoiding assumptions of infinite persistence (see main text for further discussion); in purple a specification of ‘pure growth effects’ in which the first difference of climate variables is not taken and no lagged climate variables are included (the baseline specification of ref.  2 ); and in pink a specification of ‘pure level effects’ in which the first difference of climate variables is taken but no lagged terms are included.

Extended Data Fig. 4 Climate changes in different variables as a function of historical interannual variability.

Changes in each climate variable of interest from 1979–2019 to 2035–2065 under the high-emission scenario SSP5-RCP8.5, expressed as a percentage of the historical variability of each measure. Historical variability is estimated as the standard deviation of each detrended climate variable over the period 1979–2019 during which the empirical models were identified (detrending is appropriate because of the inclusion of region-specific linear time trends in the empirical models). See Supplementary Fig. 13 for changes expressed in standard units. Data on national administrative boundaries are obtained from the GADM database version 3.6 and are freely available for academic use ( https://gadm.org/ ).

Extended Data Fig. 5 Contribution of different climate variables to overall committed damages.

a , Climate damages in 2049 when using empirical models that account for all climate variables, changes in annual mean temperature only or changes in both annual mean temperature and one other climate variable (daily temperature variability, total annual precipitation, the number of wet days and extreme daily precipitation, respectively). b , The cumulative marginal effects of an increase in annual mean temperature of 1 °C, at different baseline temperatures, estimated from empirical models including all climate variables or annual mean temperature only. Estimates and uncertainty bars represent the median and 95% confidence intervals obtained from 1,000 block-bootstrap resamples from each of three different empirical models using eight, nine or ten lags of temperature terms.

Extended Data Fig. 6 The difference in committed damages between the upper and lower quartiles of countries when ranked by GDP and cumulative historical emissions.

Quartiles are defined using a population weighting, as are the average committed damages across each quartile group. The violin plots indicate the distribution of differences between quartiles across the two extreme emission scenarios (RCP2.6 and RCP8.5) and the uncertainty sampling procedure outlined in Methods , which accounts for uncertainty arising from the choice of lags in the empirical models, uncertainty in the empirical model parameter estimates, as well as the climate model projections. Bars indicate the median, as well as the 10th and 90th percentiles and upper and lower sixths of the distribution reflecting the very likely and likely ranges following the likelihood classification adopted by the IPCC.

Supplementary information

Supplementary information, peer review file, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Kotz, M., Levermann, A. & Wenz, L. The economic commitment of climate change. Nature 628 , 551–557 (2024). https://doi.org/10.1038/s41586-024-07219-0

Download citation

Received : 25 January 2023

Accepted : 21 February 2024

Published : 17 April 2024

Issue Date : 18 April 2024

DOI : https://doi.org/10.1038/s41586-024-07219-0

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Harvard Library Is Launching Harvard Open Journals Program

Harvard Library is launching a new initiative called the Harvard Open Journals Program (HOJP), which will help researchers advance scholarly publishing that is open access, sustainable, and equitable. HOJP will provide publishing services, resources, and seed funding to participating Harvard researchers for new academic journals. All journal articles will be entirely free for authors and readers, with no barriers to publish or to access.

Martha Whitehead, Vice President for the Harvard Library and University Librarian, sees the initiative as an important step in championing open access. Whitehead said, “We want to model the original ethos of open access by reducing barriers and enabling the free flow of ideas and knowledge across the research ecosystem and beyond to the public at large.”

The Harvard Open Journals Program will offer publishing and hosting services to help the Harvard community launch new open access journals, or to convert existing journals to open access. The program will offer two support models: an overlay model which takes advantage of open access repositories, such as Harvard’s  DASH , and a brand-new academic press model. 

Yuan Li, University Scholarly Communication Officer and Director of Open Scholarship and Research Data Services at Harvard Library, pointed out the innovative nature of the program, “It is new for an institution to support faculty in seeking out an academic press to publish a no-fee open access journal and to provide assistance in securing its long-term funding. And offering a repository overlay journal model provides an alternative that appeals to some editorial boards and is gaining traction through initiatives such as Episciences. As we implement and refine this program on our campus, we hope it will inspire other universities to adopt such approaches to supporting barrier-free scholarly publishing.”

The program is a direct response to faculty interest in alternatives to the article-processing-charge model, in which journals charge author-side fees to publish papers open access. It also supports federal requirements that publications resulting from publicly-funded research be open access.

The open access movement in scholarly publishing seeks to grant free and public online access to publications and data. In recent decades, many researchers have become increasingly concerned that commercial rather than scholarly interests are driving the publishing ecosystem. With some publishers charging article processing fees of over $10,000 per article, skyrocketing costs inhibit many researchers and institutions from publishing in these journals. At the same time, research institutions continue to pay high subscription costs, even as their faculty provide editorial and peer review services mainly for free to the publishers. These practices have led to widespread outcry in the scholarly community, and tensions between publishers and editorial boards have led to the latter’s  mass resignations .

Scott Edwards, Professor of Organismic and Evolutionary Biology, and a member of the Harvard Library Faculty Advisory Council, applauds the library’s exploration of new models for supporting open access publishing. Edwards said, “In this increasingly challenging publishing ecosystem, the Harvard Open Journals Program is a welcome new approach.” 

“These are sustainable and equitable open access publishing models that allow scholars to take control of scholarly communication,” added Li. “I hope that many research-heavy institutions adopt our approach. The first  Harvard Open Access policy launched in 2008 has been adopted nationally and internationally, and it would be great to see similar reach.”

Under Harvard’s Open Access policies, Harvard faculty and researchers give the University a nonexclusive, irrevocable right to distribute their scholarly articles for any non-commercial purpose. Stored and preserved in  DASH , Harvard Library’s open access repository, these articles are made available to the scholarly community and the public—anyone with an internet connection can read them for free.

Harvard Library is working closely with the Office of the Vice Provost for Research on launching the HOJP program. John Shaw, Vice Provost for Research and Harry C. Dudley Professor of Structural and Economic Geology, is eager to promote the initiative in the suite of programs that support faculty research. Shaw said, “The launch of HOJP provides very encouraging options for removing barriers to making research results open and expanding their reach.”

The Harvard Open Journals Program will be open to all journals with a current Harvard affiliate on the editorial team or editorial board. Student-run journals are also eligible, as long as they are sponsored by a Harvard faculty member or administrator.

In preparing to launch HOJP this summer, Harvard Library is currently seeking input on program details from interested faculty. HOJP will begin accepting applications in the fall from journals and editorial boards. Colleen Cressman, Librarian for Open Publishing, will manage the program and can be reached by email for more information.

SCIENCE & ENGINEERING INDICATORS

Publications output: u.s. trends and international comparisons.

• Report PDF (1.4 MB)
• Report - All Formats .ZIP (8.9 MB)
• Supplemental Materials - All Formats .ZIP (4.1 MB)
• MORE DOWNLOADS OPTIONS
• Share on X/Twitter
• Share on Facebook
• Share on LinkedIn
• Send as Email

Publication Output by Region, Country, or Economy and by Scientific Field

This section of the report outlines trends over time in publication output across regions, countries, or economies and by fields of science. This section also provides insights into the research contributions of different regions, countries, or economies and how the focus of their scientific publications has changed over time. In addition, the section highlights variations in the distribution of publications across scientific fields for different regions, countries, or economies and examines trends over time in closed-access and open-access (OA) publications. This section also summarizes federal funding acknowledgments as a source of data to shed light on published research that received federal funding. (See sidebar Using Funding Acknowledgments to Track Federally Funded Research Over Time .)

Output by Region, Country, or Economy

Total worldwide S&E publication output reached 3.3 million articles in 2022, based on entries in the Scopus database. Indicators 2018 : Bibliometric Data Filters )." data-bs-content="Publication output includes only those indexed in the Scopus database. The publication output discussion uses fractional counting, which credits coauthored publications according to the collaborating institutions or regions, countries, or economies based on the proportion of their participating authors. Country assignments refer to the institutional address of authors, with partial credit given for each international coauthorship. As part of the data analysis, filters were employed on the raw Scopus S&E publication data to remove publications with questionable quality, which appear in what are sometimes called predatory journals (NSB Indicators 2018 : Bibliometric Data Filters )." data-endnote-uuid="a5778137-c92e-4694-991d-a1545bdcca18">​ Publication output includes only those indexed in the Scopus database. The publication output discussion uses fractional counting, which credits coauthored publications according to the collaborating institutions or regions, countries, or economies based on the proportion of their participating authors. Country assignments refer to the institutional address of authors, with partial credit given for each international coauthorship. As part of the data analysis, filters were employed on the raw Scopus S&E publication data to remove publications with questionable quality, which appear in what are sometimes called predatory journals (NSB Indicators 2018 : Bibliometric Data Filters ). Approximately 86% of publications in 2022 came from regions, countries, or economies with high-income and upper-middle-income economies ( Figure PBS-1 ). The number of publications from all income-level groups grew between 2003 and 2022 ( Table SPBS-2 ). Also, the number of publications from upper-middle-income economies grew more quickly than the number from high-income economies during the more recent period between 2010 and 2022.

• For grouped bar charts, Tab to the first data element (bar/line data point) which will bring up a pop-up with the data details
• To read the data in all groups Arrow-Down will go back and forth
• For bar/line chart data points are linear and not grouped, Arrow-Down will read each bar/line data points in order
• For line charts, Arrow-Left and Arrow-Right will move to the next set of data points after Tabbing to the first data point
• For stacked bars use the Arrow-Down key again after Tabbing to the first data bar
• Then use Arrow-Right and Arrow-Left to navigate the stacked bars within that stack
• Arrow-Down to advance to the next stack. Arrow-Up reverses

S&E publications, by income group: 2003–22

Article counts refer to publications from a selection of conference proceedings and peer-reviewed journals in S&E fields from Scopus. Articles are classified by their year of publication and are assigned to a region, country, or economy on the basis of the institutional address(es) of the author(s) listed in the article. Articles are credited on a fractional count basis (i.e., for articles produced by authors from different regions, countries, or economies, each region, country, or economy receives fractional credit on the basis of the proportion of its participating authors). Data are not directly comparable with data from Science and Engineering Indicators 2022 ; see the Technical Appendix for information on data filters. Low-income regions, countries, or economies are not included in this figure because of their low publication output. Data by region, country, or economy and income group are available in Table SPBS-2 . Regions, countries, or economies are allocated to income groups based on World Bank data, using their current designation. For example, all of China’s publications from 2003 to 2022 are counted as part of the upper-middle-income category because that is China's current designation.

National Center for Science and Engineering Statistics; Science-Metrix; Elsevier, Scopus abstract and citation database, accessed April 2023; World Bank Country and Lending Groups, accessed May 2023.

Science and Engineering Indicators

In 2022, six regions, countries, or economies each produced more than 100,000 articles: China, the United States, India, Germany, the United Kingdom, and Japan. Together, these leading regions, countries, or economies accounted for over 50% of the worldwide total in 2022 ( Figure PBS-2 ; Table PBS-1 ). Figure PBS-2 and Table PBS-1 , or whole counting, as in Table SPBS-17 . There is a slight difference between the United States and China when looking at the whole-counting total production numbers. Using whole counting for 2022, the United States had 605,633 articles, whereas China had 976,141. A whole-counting measure allocates one full count to each region, country, or economy with an author contributing to the article; in fractional counting, each region, country, or economy receives a proportion of the count based on the number of authors from that region, country, or economy. For example, if an article had four authors—two from the United States, one from China, and one from Brazil—the fractional scores would be half for the United States, a quarter for China, and a quarter for Brazil. The difference between whole and fractional counting indicates that the United States has more authors working with Chinese authors than China has working with U.S. authors." data-bs-content="The proportion of output attributable to the large producers is consistent whether using fractional counting, as in Figure PBS-2 and Table PBS-1 , or whole counting, as in Table SPBS-17 . There is a slight difference between the United States and China when looking at the whole-counting total production numbers. Using whole counting for 2022, the United States had 605,633 articles, whereas China had 976,141. A whole-counting measure allocates one full count to each region, country, or economy with an author contributing to the article; in fractional counting, each region, country, or economy receives a proportion of the count based on the number of authors from that region, country, or economy. For example, if an article had four authors—two from the United States, one from China, and one from Brazil—the fractional scores would be half for the United States, a quarter for China, and a quarter for Brazil. The difference between whole and fractional counting indicates that the United States has more authors working with Chinese authors than China has working with U.S. authors." data-endnote-uuid="a3ba485a-d862-4069-8166-933aa601ea66">​ The proportion of output attributable to the large producers is consistent whether using fractional counting, as in Figure PBS-2 and Table PBS-1 , or whole counting, as in Table SPBS-17 . There is a slight difference between the United States and China when looking at the whole-counting total production numbers. Using whole counting for 2022, the United States had 605,633 articles, whereas China had 976,141. A whole-counting measure allocates one full count to each region, country, or economy with an author contributing to the article; in fractional counting, each region, country, or economy receives a proportion of the count based on the number of authors from that region, country, or economy. For example, if an article had four authors—two from the United States, one from China, and one from Brazil—the fractional scores would be half for the United States, a quarter for China, and a quarter for Brazil. The difference between whole and fractional counting indicates that the United States has more authors working with Chinese authors than China has working with U.S. authors. In absolute numbers, the growth in worldwide annual publication output (from 2.0 million in 2010 to 3.3 million in 2022) was driven in particular by two countries: China (42% of additional publications during that period) and India (11%) together accounted for more than half of that increase in publications ( Figure PBS-3 ). Russia, South Korea, Iran, and Brazil made notable contributions to the growth in the number of publications from the rest of the world from 2010 to 2022 ( Figure PBS-3 ; Table SPBS-2 ). Generally, the set of the top 15 producers of S&E articles was the same each year between 2010 and 2022, with the exception of Iran replacing Taiwan in the top 15 beginning in 2014 ( Table PBS-1 ; Table SPBS-2 ).

S&E publications for 10 leading regions, countries, or economies: 2022

Article counts refer to publications from a selection of conference proceedings and peer-reviewed journals in S&E fields from Scopus. Articles are classified by their year of publication and are assigned to a region, country, or economy on the basis of the institutional address(es) of the author(s) listed in the article. Articles are credited on a fractional count basis (i.e., for articles produced by authors from different regions, countries, or economies, each region, country, or economy receives fractional credit on the basis of the proportion of its participating authors). Data by all countries, regions, and economies are available in Table SPBS-2 .

National Center for Science and Engineering Statistics; Science-Metrix; Elsevier, Scopus abstract and citation database, accessed April 2023.

S&E publications in all fields for 15 largest producing regions, countries, or economies: 2012 and 2022

na = not applicable.

The regions, countries, or economies are ranked based on the 2022 total. Article counts refer to publications from conference proceedings and peer-reviewed journal articles in S&E and indexed in Scopus (see Technical Appendix for more details). Articles are classified by their year of publication and are assigned to a region, country, or economy on the basis of the institutional address(es) of the author(s) listed in the article. Articles are credited on a fractional count basis (i.e., for articles from multiple regions, countries, or economies, each region, country, or economy receives fractional credit on the basis of the proportion of its participating authors). Detail may not add to total because of regions, countries, or economies that are not shown. Proportions are based on the world total excluding unclassified addresses (data not presented). Details and other regions, countries, or economies are available in Table SPBS-2 .

S&E publications, by selected region, country, or economy and rest of world: 2003–22

Article counts refer to publications from a selection of conference proceedings and peer-reviewed journals in S&E fields from Scopus. Articles are classified by their year of publication and are assigned to a region, country, or economy on the basis of the institutional address(es) of the author(s) listed in the article. Articles are credited on a fractional count basis (i.e., for articles produced by authors from different countries, each country receives fractional credit on the basis of the proportion of its participating authors). Data for all regions, countries, and economies are available in Table SPBS-2 .

The U.S. trend of moderate but increasing publication output varied by state. The National Science Board’s (NSB’s) State Data Tool ( https://ncses.nsf.gov/indicators/states/ ) provides state-level data based on each state’s doctorate population and research and development (R&D) funding. Indicators include academic S&E article output per 1,000 science, engineering, and health doctorate holders in academia (NSB 2021a) and academic S&E article output per $1 million in academic S&E R&D funding (NSB 2021b).

Output by Scientific Field

The distribution of publications by field of science across region, country, or economy may indicate research priorities and capabilities. Health sciences was the field of science in which most articles were published in 2022, representing almost a quarter of all publications ( Table SPBS-2 and Table SPBS-10 ). Other fields with large numbers of publications included engineering (17% of publications in 2022), biological and biomedical sciences (13%), and social sciences (5%) ( Table SPBS-2 , Table SPBS-5 , Table SPBS-8 , and Table SPBS-16 ). In the United States, the European Union (EU-27), and Japan, health sciences publication output in 2022 far exceeded that of any other field. Table SPBS-17 through Table SPBS-31 )." data-bs-content="The use of whole counting or fractional counting to tally the publication output of nations can change the calculated publication count based on the degree to which a region, country, or economy is involved in international collaborations. Under whole counting, a nation receives credit for any publication with an author from that nation. Under fractional counting, the nation’s credit for a publication is prorated based on the share of the publication’s coauthors who are located in that nation ( Table SPBS-17 through Table SPBS-31 )." data-endnote-uuid="05c8023f-5250-4386-96cf-f321c7f293c4">​ The use of whole counting or fractional counting to tally the publication output of nations can change the calculated publication count based on the degree to which a region, country, or economy is involved in international collaborations. Under whole counting, a nation receives credit for any publication with an author from that nation. Under fractional counting, the nation’s credit for a publication is prorated based on the share of the publication’s coauthors who are located in that nation ( Table SPBS-17 through Table SPBS-31 ). Meanwhile, of the other top producers, publications from China were most highly concentrated in engineering (25%), and publications from India were published predominantly in computer and information sciences (21%) ( Figure PBS-4 ).

Distribution of national S&E research portfolios across scientific fields, by selected region, country, or economy: 2022

EU-27 = European Union.

Articles refer to publications from a selection of conference proceedings and peer-reviewed journals in S&E fields from Scopus. Articles are classified by their year of publication and are assigned to a region, country, or economy on the basis of the institutional address(es) of the author(s) listed in the article. Articles are credited on a fractional count basis (i.e., for articles from multiple countries, each country receives fractional credit on the basis of the proportion of its participating authors). See Table SPBS-1 for countries included in the EU; beginning in 2020, the United Kingdom was no longer a member of the EU. See Table SPBS-3 through Table SPBS-16 for data on all regions, countries, and economies by each S&E field.

Fields within life sciences were dominant in the United States in 2022, with more than half of all U.S. publications in health sciences (37%) or biological and biomedical sciences (14%) ( Figure PBS-5 ). There were fewer U.S. publications in engineering (11%), computer and information sciences (7%), and physics (5%). In comparison with the United States, China had a stronger focus on publications in engineering and in the physical sciences and information sciences. In 2022, 25% of China’s publications were in engineering, 11% were in computer and information sciences, and 9% were in physics ( Figure PBS-6 ). Compared with the United States, China had a lower percentage of its publications in health sciences (14%) and biological and biomedical sciences (12%). In 2022, China also had a much lower percentage of its publications in social sciences (1%) when compared with the United States (8%).

U.S. S&E publication portfolio, by field of science: 2022

Articles refer to publications from a selection of conference proceedings and peer-reviewed journals in S&E fields from Scopus. Articles are classified by their year of publication and are assigned to a region, country, or economy on the basis of the institutional address(es) of the author(s) listed in the article. Articles are credited on a fractional count basis (i.e., for articles from multiple countries, each country receives fractional credit on the basis of the proportion of its participating authors). See Table SPBS-3 through Table SPBS-16 for data on all regions, countries, and economies by each S&E field.

S&E publication portfolio from China, by field of science: 2022

Articles refer to publications from a selection of conference proceedings and peer-reviewed journals in S&E fields from Scopus. Articles are classified by their year of publication and are assigned to a region, country, or economy on the basis of the institutional address(es) of the author(s) listed in the article. Articles are credited on a fractional count basis (i.e., for articles from multiple countries, each country receives fractional credit on the basis of the proportion of its participating authors). See Table SPBS-3 through Table SPBS-16 for data on all regions, countries, and economies and by each S&E field.

All the leading regions, countries, or economies saw an increase in their output of health sciences publications between 2010 and 2022. This increase is to be expected, given the context of increasing publication rates in general over that period, with overall number of publications increasing by 71% ( Table SPBS-2 ), while publications in health sciences increased by 66% ( Table SPBS-10 ). Russia had the highest relative growth rate among the 20 leading regions, countries, or economies in health sciences, increasing its publication output by almost 450% between 2010 and 2022 ( Table SPBS-10 ). China and Iran each increased their output of health sciences publications by more than 250% over this period, while India’s health sciences publication output increased by more than 180%. The United States increased its output of health sciences publications by 32% over this period, while Germany, France, the United Kingdom, and Japan had the smallest increases, each with less than 20%.

Leading regions, countries, or economies also saw increases in engineering publications. The fastest growing between 2010 and 2022 were India (up 378%) and Russia (up 230%) ( Table SPBS-8 ). China increased its output of engineering publications by 176% from 2010 to 2022, while France, the United States, and Japan all saw declines in newly published engineering articles per year over this period (3%, 13%, and 26%, respectively).

In the United States, publication output varied from that of other regions, countries, or economies with respect to scientific fields. Of the fields not already mentioned, the fastest growing from 2010 to 2022 were psychology (up 39% from 2010 to 2022) and the social sciences (up 38%) ( Figure PBS-7 ). Meanwhile, fields with the largest decreases in U.S. publications included physics (down 31% from 2010 to 2022) and materials science (down 16%).

Index of U.S. publications, by field: 2010–22

Using funding acknowledgments to track federally funded research over time.

Federally funded research is an important component of the research ecosystem and is often envisioned as a means of supporting science performed for public benefit that may not otherwise be motivated by commercial interest (Bornmann 2013; Stephan 2012; Yin et al. 2022). Federal research funding supports applied and basic research (see Indicators 2022 report “ Research and Development: U.S. Trends and International Comparisons ”) and has long been linked to successful expansions in scientific production—through the increased productivity and impact of individual researchers and laboratories (Ebadi and Schiffauerova 2016) and the national scale (Leydesdorff and Wagner 2009). This sidebar explores funding acknowledgments, as recorded in Scopus, as an emerging source to help illustrate the extent to which published research is supported by federal agencies and the trends in federally funded research. Specifically, the share of published research acknowledging support by federal funding was highest in chemistry and smaller in other fields, such as the social sciences ( Table PBS-A ). These differences may be driven by factors such as the resource costs to conduct research and by field differences, such as the overall frequency of publication, team size, and cultural differences among the disciplines. The time period analyzed in this sidebar is 2018–22, unless otherwise indicated.

Funding acknowledgments can shed light on the ability and priorities of federal funding to support discovery as measured by peer-reviewed journal articles and conference proceedings. However, some benefits and limitations of this emerging data source are important to highlight so as to accurately interpret these trends. Each peer-reviewed journal article and conference proceeding in the Scopus database includes a field for funding acknowledgments that are extracted by algorithmic (software) means. In some cases where the acknowledgments field is incomplete, funding information from agencies is also used to identify funded publications in Scopus. Using this field, it is possible to observe the conversion of federal funds to published research outputs, but a direct linkage between funding inputs and published discoveries remains challenging. First, extraction of this information into a structured field is a relatively new effort and is most complete for the most recent 4 years. Figure SPBS-1 shows how funding acknowledgment sections have grown in coverage since 2003 and that funding information was indexed for 68% of all publications in 2022. * Many factors may have contributed to this growth in addition to improved extraction, including increasing pressure and requirements from funders to include funding acknowledgments, standardization of acknowledgment language, and incentives to demonstrate high publication output—because future funding is tied to past conversion of funds into publications—while receiving funding. † Last, this inquiry helps explore research that acknowledges any federal funding but does not only account for publications that source all their funding from a single source. In practice, a publication may be generated using funding from multiple sources within the federal government, or from additional sources in state government, local government, or the private sector.

U.S. S&E publications with and without acknowledgments of U.S. federal funding: 2003–22

Articles are classified by their year of publication and are assigned to a region, country, or economy on the basis of the institutional address(es) of the author(s) listed in the article. Whole counting is used. An article is considered to be federally funded if the funding information tied with the publication record in Scopus links it with one of the U.S. federal agencies. Not all Scopus publications have funding information available, and coverage has evolved with time. For more information, see Figure SPBS-1 . For a breakdown of federally funded papers by funding agency, see Table SPBS-90 .

National Center for Science and Engineering Statistics; Science-Metrix; Elsevier, Scopus abstract and citation database, accessed April 2023.

Figure PBS-A tracks the growth of federally funded publications relative to the total research production in the United States. Other than a small downturn from 2021 to 2022, every year has seen an increase from the previous year in the number of publications that acknowledge funding support from federal agencies. The most comprehensive data from the past 4 years show variation among subject areas in the percentage of publications that acknowledge federal support. Table PBS-A shows number and share of publications appearing between 2018 and 2022 that acknowledged funding from federal sources and those acknowledging funding from other sources. During this time, more than 50% of publications in the following subject fields acknowledged federal funding support: chemistry (55% of publications), biological and biomedical sciences (53%), astronomy and astrophysics (53%), and physics (52%). Only two subject areas have less than 30% of publications with federal funding acknowledged: agricultural sciences (28%), and social sciences (15%). Otherwise, all other fields had between 30% and 50% of their publications acknowledging federal funding.

U.S. S&E publications, by U.S. federal funding status and field: 2018–22

In conclusion, federal funding plays an important role in the current research environment in the United States. Of the 606,144 articles published in journals and conference proceedings in 2022, 35% acknowledged support from federal agencies ( Figure PBS-A ). Ultimately, acknowledgment of federal funding can help show trends in the conversion of grants into published research over time and show variation at the subject or field level.

* Missing data in funding fields in a Scopus entry may mean that the research did not receive funding, the authors did not cite any funding despite receiving it, or the algorithm was unable to extract the acknowledgment. Of the articles from 2003 that had an entry for funding acknowledgment (27% had text in the funding field in Scopus), around 76% acknowledged a federal funding source. Comparatively, of the publications in 2022 with indexed funding information (68%), 52% acknowledged a federal source. The growth of coverage of funding not being tied to federal funding acknowledgments provides evidence that the data source has become more dependable over time. Data for the percentage of publications with indexed funding sources by year and field can be found in Figure SPBS-1 .

† Table SPBS-90 displays the number of articles and conference proceedings acknowledging federal funding at the agency level and sub-agency level. These counts represent the number of supported articles as acknowledged and attributed in Scopus from 2003 to 2022.

Output and Open Access

There is growing support for the availability of S&E publications through OA sources among government and private funders, institutions, and scientists themselves. Some of these funders have imposed requirements on their grantees to publish their research results in OA journals. In the United States, the Office of Science and Technology Policy announced that all federal agencies should update their public access policies as soon as possible to ensure that results of their funded research are publicly available, with full implementation of these policies by the end of 2025 (Brainard and Kaiser 2022). Meanwhile, restricted access to scientific literature may impede researchers’ ability to stay informed (Larivière and Sugimoto 2018; Piwowar et al. 2018). As alternatives to traditional closed-access journals (where readers must subscribe to gain access or pay per article), articles may be made OA through several avenues, with different levels of availability and durability.

There are four commonly defined types of OA: Gold, Hybrid, Bronze, and Green. Gold OA denotes articles published in journals that are entirely OA as a matter of journal policy. Hybrid OA denotes articles for which the authors have elected to pay a fee for publication as OA rather than as closed access. Bronze OA denotes articles that appear as OA after an embargo period of closed access or articles that appear available as OA despite lacking license information to guarantee OA in the long term. Green OA denotes articles that are self-archived by authors in OA repositories, which are often maintained, curated, and administered by universities or other institutions. The Hybrid and Bronze categories have been combined as Other Journal-Based OA in this report because of their similar structure as journal-hosted types of OA that allow only conditional—and potentially revocable—OA.

The number of articles published annually in closed-access journals increased by 112% between 2003 and 2022 ( Figure PBS-8 ). Over the same period, annual publishing of Green OA articles increased by 228%, while Other Journal-Based OA articles (Hybrid and Bronze OA) increased by 198%. Gold OA articles (which are published in OA journals with no restrictions) had the largest percentage growth, from 19,089 articles in 2003 to 991,805 articles in 2022, an increase of over 5,000%. Hence, although the majority (77%) of S&E articles in 2003 whose access status is known were published in closed-access journals, fewer than half (49%) were in closed-access journals in 2022.

S&E publications, by publication access type: 2003–22

OA = open access.

Articles refer to publications from a selection of conference proceedings and peer-reviewed journals in S&E fields from Scopus. Articles are classified by their year of publication. OA types are mutually exclusive. For articles published under multiple OA types, the article will be counted as part of only the first type it matches in this list: Gold OA, Other Journal-Based OA, or Green OA. Summing all OA and closed-access article counts results in a smaller number of articles than for all S&E because the access status of some articles (e.g., those without digital object identifiers) cannot be reliably ascertained. Green articles are published in toll-access journals but archived in an OA archive, or "repository." These repositories may be discipline specific (like arXiv) or institutional repositories operated by universities or other institutions. Green articles may be published versions or preprints and can have any license or no license. Bronze (Other Journal-Based OA) articles are free to read on the publisher's website, without a license that grants any other rights. There may be a delay between publication and availability to read, and often articles can be removed unilaterally by the publisher. Hybrid (Other Journal-Based OA) articles are free to read at the time of publication, with an open license. These are usually published in exchange for an article processing charge. Gold articles have all the same characteristics as Hybrid articles but are published in all-OA journals, which are in turn called "Gold journals" or just "OA journals."

To conclude this section, the findings of the output analysis reveal the growth in scientific publications over time, with upper-middle-income economies exhibiting particularly large percentage increases. Meanwhile, the distribution of publications across scientific fields shows that life sciences dominated in the United States, Europe, and Japan, whereas publications in engineering and computer sciences dominated in China and India. In OA, the dramatic growth of Gold OA publications and the steady growth of publications in other OA categories show an increased shift toward open science. However, OA can impede the dissemination of some scientific research. Publishing research as OA often requires authors to pay article processing fees, which may be prohibitive for scientists in less-developed nations or whose funders do not subsidize those fees. https://www.elsevier.com/about/policies/pricing . Wiley APCs are at https://authorservices.wiley.com/author-resources/Journal-Authors/open-access/article-publication-charges.html . Springer Nature APCs are at https://www.springernature.com/gp/open-research/journals-books/journals ." data-bs-content="Many publishers make their article processing charges (APCs) known publicly. For example, a list of Elsevier APCs can be found at https://www.elsevier.com/about/policies/pricing . Wiley APCs are at https://authorservices.wiley.com/author-resources/Journal-Authors/open-access/article-publication-charges.html . Springer Nature APCs are at https://www.springernature.com/gp/open-research/journals-books/journals ." data-endnote-uuid="9a612a36-466d-4d60-af1a-57e93bd50e76">​ Many publishers make their article processing charges (APCs) known publicly. For example, a list of Elsevier APCs can be found at https://www.elsevier.com/about/policies/pricing . Wiley APCs are at https://authorservices.wiley.com/author-resources/Journal-Authors/open-access/article-publication-charges.html . Springer Nature APCs are at https://www.springernature.com/gp/open-research/journals-books/journals . The fees can be seen as shifting the costs of accessing research from readers and libraries to authors (Larivière and Sugimoto 2018).

Related Content