research papers uk

Free to access articles

As well as  open access , we  provide  ‘free  to  access' journal  content . A rticles  denoted  b y a tick symbol  can be accessed free of charge.

Access for low-and middle-income countries

In keeping with our role as the UK's national academy of science, we offer  free access to readers in low-and middle-income countries .

Journal archive

The Royal Society’s Philosophical Transactions , launched in 1665, was the world’s first scientific journal . It established the fundamental principles of scientific priority and peer review.

Most of our oldest content is now freely available, specifically, all papers older than 70 years. In addition, papers published between 10 years ago and either 12 months ago (biological sciences and history of science) or 24 months ago (physical sciences) from online issue publication are freely available. For Biographical Memoirs all issues are now freely available.

Please refer to the table below for an overview of the freely available content in our journals.

* 1 year for biological sciences ( including JRSI, Focus and N&R ) and 2 years for physical sciences. 

Package subscribers receive free lease access to the full journal archive, while their subscription remains active. If you are interested in purchasing perpetual access to the non-free content in our archive, please see our archive purchasing options

Open access week

Each year, during  Open Access Week , we make all of our published content free to access.

Related content

Open access publishing

The Royal Society supports open access publishing as part of our commitment to the widest possible…

Royal Society Open Access Equity

Free journal access and automatic APC waivers to researchers in low- and middle-income countries

History of scientific journals

Over 350 years of scientific publishing

Email updates

We promote excellence in science so that, together, we can benefit humanity and tackle the biggest challenges of our time.

Subscribe to our newsletters to be updated with the latest news on innovation, events, articles and reports.

What subscription are you interested in receiving? (Choose at least one subject)

Access to Research

Discover a world of published academic research at your local library

Access to Research gives free, walk-in access to over 30 million academic articles in participating public libraries across the UK. Start now by viewing which articles and journals are available from home, then find a participating library where you can view the full text. Share #AccessToResearch

• Art, Architecture, Biological Sciences, Business, Engineering, Environmental Science, Film, Health, History, Journalism, Languages, Politics, Philosophy, Physics, Religion, Social Sciences, Mathematics

Other Areas of interest

What, Why, Who? Find out more.

Which publishers are taking part?

Which libraries are participating?

About Open Access

Supported by

🇺🇦    make metadata, not war

Oxford University ranked number 1 in the  Times Higher Education (THE) World University Rankings  for the eighth year running, and at the heart of this success is our ground-breaking research and innovation.

AI at Oxford

Applying AI to society's greatest challenges and tackling its ethical issues

Brain and Mental Health

How Oxford experts are exploring the most complex object in the known universe

Oxford’s REF 2021 results show largest volume of world-leading research

The Research Excellence Framework assesses the quality of research in UK higher education.

Oxford's innovation case studies

Helping you to change the world

Oxford's Global Research Map

Explore Oxford's world-class research from pole to pole and in every continent

Oxford is world-famous for research excellence and home to some of the most talented people from across the globe. Our work helps the lives of millions, solving real-world problems through a huge network of partnerships and collaborations. The breadth and interdisciplinary nature of our research sparks imaginative and inventive insights and solutions.

Oxford profiles

Meet some of the talented people behind Oxford’s world-class research. Pushing forward the boundaries of knowledge, their work solves real world problems and creates a positive impact on our societies, economies and health.

Started in Oxford

The Oxford region is one of the most innovative in the UK, with new enterprises continuing to join a growing band of spinouts, startups and entrepreneurs.

Oxford's experience in Policy Engagement

Oxford’s researchers and academics have a wealth of experience in engaging with policymakers and contributing to policy impact.

Engaged research at Oxford

Have a look at some of the short films below for excellent examples of Public Engagement with Research (PER) activities that take place at Oxford which Inform/Inspire, Consult and Collaborate with the public.

Research Collaboration Values

Our approach to research collaboration and partnership is underpinned by five core values.

Oxford Impact Films

Watch our Research Impact films: 3-4 minute videos of how our research has benefitted policy, health, business and culture. 

Oxford leads Nature Positive Universities Alliance to reverse biodiversity decline

The Nature Positive Universities Alliance brings higher education institutions together to use their unique power and influence as drivers of positive change.

From research to action: How the Young Lives project is helping to protect girls from child marriage

Our early experiences can have a staggering impact on the rest of our lives – for better or worse.

New Academic Champion for Women and Diversity in Entrepreneurship

To support diversity in innovation and entrepreneurship, and to enhance the University’s commitment to these goals, Professor Stevens will work with the IDEA (Increasing Diversity in Enterprising Activities) programme and

AI, automation in the home and its impact on women

As we mark International Women’s Day, Professor Ekaterina Hertog spoke to us about AI, the increase of automation in the home and its impact on women and wider society.

Research to policy impact: strategies for translating findings into policy messages

Getting started in Policy Engagement: pathways to engagement

Experts call for responsible use of generative AI in adult social care

‘Adult social care is about supporting people to live independently and to protect fundamental human rights. Generative AI offers many potential benefits and opportunities to adult social care.

Boosting Policy Engagement Through OPEN Leaders

Urgent call for UK Government to develop a heat resilience strategy

A new device to detect cardiovascular disease

New research shows the Cerne Abbas Giant was a muster station for King Alfred’s armies

Research undertaken by Martin Papworth for the National Trust, showed that the Giant was carved in the Anglo-Saxon period not, as most people thought, in prehistory or more recently, yet the reason why he was made has remained a mystery.

Engaging communities in wildlife conservation through storybooks

Impact case study.

A Partnership in Learning by Doing: using research to engage policymakers to pave the way for electric car clubs in Oxfordshire

10 recommendations for best practice stakeholder engagement

Developing the next generation of wildlife conservation leaders

A Global Resilience Index: Supporting climate adaptation of global infrastructure systems

Tackling mental illness by supporting industry to develop new drugs

King Charles presents President Macron with Oxford University research on Voltaire’s work

The gift is an extract from a University of Oxford research project to produce and publish the first ever scholarly edition of the Œuvres complètes de Voltaire (Complete works of Voltaire) begun in 1968, completed over 50 years later, in 2022, made up of 205 volumes.

Development of a Malaria vaccine - R21/Matrix-M

Impact case studies.

Oxford University welcomes UK associate membership of Horizon Europe

Horizon Europe is the EU’s funding programme for research and innovation projects for the years 2021 to 2027.  The programme has a budget of €95.5 billion (£81bn). It is the successor to Horizon 2020 and the previous Framework Programmes for Research and Technological Development.

Latest research news

• Research strategy • Engagement and partnership strategy • Divisions

Latest campaigns

We are developing fundamental AI tools, applying AI to global challenges, and addressing the ethical issues of new technologies.

Experts at Oxford are expanding our understanding of brain health at a cellular level, exploring the impacts of mental health issues on the individual, and examining population-wide global health problems.

Other campaigns

Cancer at Oxford

Digital collections at oxford, true planet, coronavirus research.

• - Google Chrome

Intended for healthcare professionals

• Access provided by Google Indexer
• My email alerts
• BMA member login
• Username * Password * Forgot your log in details? Need to activate BMA Member Log In Log in via OpenAthens Log in via your institution
• International

Search form

• Advanced search
• Search responses
• Search blogs
• The BMJ research homepage: make an impact, change clinical practice

At a glance: Research

Association of ultra-processed food consumption with all cause and cause specific mortality

Clinical and healthcare use outcomes after cessation of long term opioid treatment due to prescriber workforce exit

Long acting progestogens versus combined oral contraceptive pill for preventing recurrence of endometriosis related pain

Guidance on terminology, application, and reporting of citation searching: the TARCiS statement

Comparative effectiveness of second line oral antidiabetic treatments among people with type 2 diabetes mellitus

Research papers, effect of the hpv vaccination programme on incidence of cervical cancer and grade 3 cervical intraepithelial neoplasia by socioeconomic deprivation in england: population based observational study, efficacy of psilocybin for treating symptoms of depression, reverse total shoulder replacement versus anatomical total shoulder replacement for osteoarthritis, effect of combination treatment with glp-1 receptor agonists and sglt-2 inhibitors on incidence of cardiovascular and serious renal events, prenatal opioid exposure and subsequent risk of neuropsychiatric disorders in children, temporal trends in lifetime risks of atrial fibrillation and its complications between 2000 and 2022, multiple adverse outcomes associated with antipsychotic use in people with dementia, tripod+ai statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods, predicting the risks of kidney failure and death in adults with moderate to severe chronic kidney disease, impact of large scale, multicomponent intervention to reduce proton pump inhibitor overuse, efficacy of a single low dose of esketamine after childbirth for mothers with symptoms of prenatal depression, glucagon-like peptide 1 receptor agonist use and risk of thyroid cancer, quantifying possible bias in clinical and epidemiological studies with quantitative bias analysis: common approaches and limitations, use of progestogens and the risk of intracranial meningioma, delirium and incident dementia in hospital patients, derivation and external validation of a simple risk score for predicting severe acute kidney injury after intravenous cisplatin, community based complex interventions to sustain independence in older people, quality and safety of artificial intelligence generated health information, current safeguards, risk mitigation, and transparency measures of large language models against the generation of health disinformation, assessing robustness to worst case publication bias using a simple subset meta-analysis, research news, covid-19: nearly 20% of patients receive psychiatric diagnosis within three months of covid, antibiotics are as good as surgery for appendicitis, study reports, black babies are less likely to die when cared for by black doctors, us study finds.

Submit your paper

Latest video

A boat shaped bleed, protective power of plants . . . and other stories, fibroids, hypertension, . . . and other stories.

Tate Papers

Tate Papers is an online research journal that publishes scholarly articles on British and modern international art, and on museum practice today. These areas reflect the breadth of Tate’s collection, exhibition programme and activities. Leading specialists from around the world contribute to Tate Papers, as do researchers working at Tate, and the journal aims to showcase a range of disciplinary approaches to the study of art and museums.

Information about Tate Papers journal, its Academic Advisory Board and statement on Open Access and copyright

How to Submit

We have temporarily paused open submissions. Read more about the changes we’re making to our submission processes

Current Issue

Tate papers no.35.

New research into the conservation of paintings by Modigliani, Tate conservators on their experiences of works by Ima-Abasi Okon, and viewing the museum from within Richard Bell’s Embassy

Past Issues

Tate papers no.34.

Papers on the São Paulo Biennial (1964–85), a focus on socially engaged art and technical analysis of a Burne-Jones altarpiece

Tate Papers no.33

A group of papers on John Constable, and articles on Renato Guttuso, Jessa Fairbrother, Eva Hesse, and more

Tate Papers no.32

Focusing on ‘American art’, provincialism and transnationalism; and including articles on Joseph Beuys, Nam June Paik and experimental Philippine art

Tate Papers no.31

Joseph Beuys is the main focus of this issue, which also includes papers on Frank Bowling and Anthony Hill

Tate Papers no.30

A collection of papers from the ‘Positioning Nigerian Modernism’ conference, plus reflections on Tate Exchange

Tate Papers no.29

Including nineteenth-century sculptor John Gibson, practice-based research, former Tate director John Rothenstein’s curatorial vision and Chris Ofili’s The Upper Room

Tate Papers no.28

Examining seven works – three by Picasso, four by Picabia – all with paintings hidden beneath paintings

Tate Papers no.27

This issue features exchanges between British and American artists: David Hockney, Sargent, Joseph Pennell and more

Tate Papers no.26

Richard Hamilton, Bruce Nauman, Hungarian neo-avant-garde and contemporary art, and Meyer Schapiro

Tate Papers no.25

The politics of collaboration, Turner, Hepworth and the value of exhibitions

Tate Papers no.24

The internationalism of pop art, William Hazlitt, Paul Thek and Lynn Hershman Leeson

Tate Papers no.23

Art and ideas from Asia, politics, action and the body, Mark Rothko, Jay DeFeo, and visual experience in British art

Tate Papers no.22

Ludic exhibitions, Hepworth, learning at Tate, Victorian domestic art, deserts and ‘the end’

Tate Papers no.21

Paolozzi, Schendel, Sylvester, contemporary Chinese art, and photography and the American West

Tate Papers no.20

Adrian Stokes, Kenneth Clark, Barbara Hepworth and a newly restored sixteenth-century portrait

Sorry, no image available

Tate Papers no.19

August Sander, Edward Burra, Akram Zaatari, learning and institutional critique, and Tate’s Digital Strategy

Tate Papers no.18

A group of papers on ‘involuntary drawing’, and others on Anthony Van Dyck, John Everett Millais, Allan Sekula

Tate Papers no.17

A group of papers from the ‘Art & Environment’ conference at Tate, as well as Francis Bacon, Edgar Degas and Van Dyck

Tate Papers no.16

The life and legacy of critic Lawrence Alloway and a report on learning initiatives associated with the ARTIST ROOMS collection

Tate Papers no.15

Carsten Höller, Henry Moore, David Musgrave, and articles from the conference Interpretation, Theory & the Encounter at Tate Britain

Tate Papers no.14

William Blake, drawing as a modern art practice, the sublime, Naum Gabo, Liubov Popova and Richard Hamilton

Tate Papers no.13

Articles from ‘Wrong From the Start’: Modernism and the Sublime , Anna Cutler on learning in cultural institutions, Tate’s online strategy

Tate Papers no.12

Including papers from Landmark Exhibitions: Contemporary Art Shows since 1968 , and Rodchenko, The Other Story , art spaces in Beirut, and more

Tate Papers no.11

The artist as educator, Rothko interpretation at Tate, Sally Tallant on integrated programming, and more

Tate Papers no.10

Andrei Tarkovsky, a group of articles on Cy Twombly, the evolution of Tate’s peer-led youth group Tate Forum, and more

Tate Papers no.9

A collection of articles originating from the Archival Impulse Study Day at Tate Britain, as well as Alfredo Jaar, Constable and Hans Hartung

Tate Papers no.8

A collection of papers produced for Inherent Vice: The Replica and its Implications in Modern Sculpture , at Tate Modern

Tate Papers no.7

Gordon Matta-Clark and Le Corbusier, reconstructing artists’ oil painting materials, Augustus Leopold Egg, Josef Albers and Eva Hesse

Tate Papers no.6

Alfred Watkins, Rodney Graham, Edward Hopper, time-based media conservation, cleaning acrylic emulsion paint, Wassily Kandinsky

Tate Papers no.5

John Constable, Frances Hodgkins, Marcel Duchamp, Walter Sickert and seaside Pierrots, Jack Burnham’s systems aesthetics

Tate Papers no.4

Joseph Beuys, Emila Medková, the Schools Programme at Tate Modern, Thomas Gainsborough, Eileen Agar

Tate Papers no.3

Victor Burgin, the work of education curators at Tate, Thomas Guest and Paul Nash, Marcel Duchamp and Richard Hamilton

Tate Papers no.2

Teaching interpretation, New Media Art, painting conservation, E.A. Hornel, relational aesthetics, Donald Judd, and more

Tate Papers no.1

Art & Language, Doris Salcedo, Anthony Van Dyck, Gary Hill, Joseph Cornell, Henry Fuseli, Michael Landy, Bernd and Hilla Becher

Banner image credit: Detail of damaged lower panel of Marcel Duchamp’s The Bride Stripped Bare by Her Bachelors, Even (The Large Glass) 1915–23, Estate of Marcel Duchamp / ADAGP, Paris, Photo © Bryony Bery

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: 07 June 2023

CORE: A Global Aggregation Service for Open Access Papers

• Petr Knoth   ORCID: orcid.org/0000-0003-1161-7359 1 ,
• Drahomira Herrmannova   ORCID: orcid.org/0000-0002-2730-1546 1   nAff2 ,
• Matteo Cancellieri 1 ,
• Lucas Anastasiou 1 ,
• Nancy Pontika 1 ,
• Samuel Pearce 1 ,
• Bikash Gyawali 1 &
• David Pride 1  

Scientific Data volume  10 , Article number:  366 ( 2023 ) Cite this article

7121 Accesses

76 Altmetric

Metrics details

• Research data

This paper introduces CORE, a widely used scholarly service, which provides access to the world’s largest collection of open access research publications, acquired from a global network of repositories and journals. CORE was created with the goal of enabling text and data mining of scientific literature and thus supporting scientific discovery, but it is now used in a wide range of use cases within higher education, industry, not-for-profit organisations, as well as by the general public. Through the provided services, CORE powers innovative use cases, such as plagiarism detection, in market-leading third-party organisations. CORE has played a pivotal role in the global move towards universal open access by making scientific knowledge more easily and freely discoverable. In this paper, we describe CORE’s continuously growing dataset and the motivation behind its creation, present the challenges associated with systematically gathering research papers from thousands of data providers worldwide at scale, and introduce the novel solutions that were developed to overcome these challenges. The paper then provides an in-depth discussion of the services and tools built on top of the aggregated data and finally examines several use cases that have leveraged the CORE dataset and services.

Similar content being viewed by others

Augmenting large language models with chemistry tools

OpenFold: retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization

GAGE-seq concurrently profiles multiscale 3D genome organization and gene expression in single cells

Introduction.

Scientific literature contains some of the most important information we have assembled as a species, such as how to treat diseases, solve difficult engineering problems, and answer many of the world’s challenges we are facing today. The entire body of scientific literature is growing at an enormous rate with an annual increase of more than 5 million articles (almost 7.2 million papers were published in 2022 according to Crossref, the largest Digital Object Identifier (DOI) registration agency). Furthermore, it was estimated that the amount of research published each year increases by about 10% annually 1 . At the same time, an ever growing amount of research literature, which has been estimated to be well over 1 million publications per year in 2015 2 , is being published as open access (OA), and can therefore be read and processed with limited or no copyright restrictions. As reading this knowledge is now beyond the capacities of any human being, text mining offers the potential to not only improve the way we access and analyse this knowledge 3 , but can also lead to new scientific insights 4 .

However, systematically gathering scientific literature to enable automated methods to process it at scale is a significant problem. Scientific literature is spread across thousands of publishers, repositories, journals, and databases, which often lack common data exchange protocols and other support for inter-operability. Even when protocols are in place, the lack of infrastructure for collecting and processing this data, as well as restrictive copyrights and the fact that OA is not yet the default publishing route in most parts of the world further complicate the machine processing of scientific knowledge.

To alleviate these issues and support text and data mining of scientific literature we have developed CORE ( https://core.ac.uk/ ). CORE aggregates open access research papers from thousands of data providers from all over the world including institutional and subject repositories, open access and hybrid journals. CORE is the largest collection of OA literature–at the time of writing this article, it provides a single point of access to scientific literature collected from over ten thousand data providers worldwide and it is constantly growing. It provides a number of ways for accessing its data for both users and machines, including a free API and a complete dump of its data.

As of January 2023, there are 4,700 registered API users and 2,880 registered dataset and more than 70 institutions have registered to use CORE Recommender in their repository systems.

The main contributions of this work are the development of CORE’s continuously growing dataset and the tools and services built on top of this corpus. In this paper, we describe the motivation behind the dataset’s creation and the challenges and methods of assembling it and keeping it continuously up-to-date. Overcoming the challenges posed by creating a collection of research papers of this scale required devising innovative solutions to harvesting and resource management. Our key innovations in this area which have contributed to the improvement of the process of aggregating research literature include:

Devising methods to extend the functionality of existing widely-adopted metadata exchange protocols which were not designed for content harvesting, to enable efficient harvesting of research papers’ full texts.

Developing a novel harvesting approach (referred to here as CHARS) which allows us to continuously utilise the available compute resources while providing improved horizontal scalability, recoverability, and reliability.

Designing an efficient algorithm for scheduling updates of harvested resources which optimises the recency of our data while effectively utilising the compute resources available to us.

This paper is organised as follows. First, in the remainder of this section, we present several use cases requiring large scale text and data mining of scientific literature, and explain the challenges in obtaining data for these tasks. Next, we present the data offered by CORE and our approach for systematically gathering full text open access articles from thousands of repositories and key scientific publishers.

Terminology

In digital libraries the term record is typically used to denote a digital object such as text, image, or video. In this paper and when referring to data in CORE, we use the term metadata record to refer to the metadata of a research publication, i.e. the title, authors, abstract, project funding details, etc., and the term full text record to describe a metadata record which has an associated full text.

We use the term data provider to refer to any database or a dataset from which we harvest records. Data providers harvested by CORE include disciplinary and institutional repositories, publishers and other databases.

When talking about open access (OA) to scientific literature, we refer to the Budapest Open Access Initiative (BOAI) definition which defines OA as “free availability on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose” ( https://www.budapestopenaccessinitiative.org/read ). There are two routes to open access, 1) OA repositories and 2) OA journals. The first can be achieved by self-archiving (depositing) publications in repositories (green OA), and the latter by directly publishing articles in OA journals (gold OA).

Text and Data Mining of Scientific Literature

Text and data mining (TDM) is the discovery by a computer of new, previously unknown information, by automatically extracting information from different written resources ( http://bit.ly/jisc-textm ). The broad goal of TDM of scientific literature is to build tools that can retrieve useful information from digital documents, improve access to these documents, or use these documents to support scientific discovery. OA and TDM of scientific literature have one thing in common–they both aim to improve access to scientific knowledge for people. While OA aims to widen the availability of openly available research, TDM aims to improve our ability to discover, understand and interpret scientific knowledge.

TDM of scientific literature is being used in a growing number of applications, many of which were until recently not viable due to the difficulties associated with accessing the data from across many publishers and other data providers. Because many use cases involving text and data mining can only realise their full potential when they are executed on an as large corpus of research papers as possible, these data access difficulties have rendered many of the uses cases described below very difficult to achieve. For example, to reliably detect plagiarism in newly submitted publications it is necessary to have access to an always up-to-date dataset of published literature spanning all disciplines. Based on data needs, scientific literature TDM use cases can be broadly categorised into the following two categories, which are shown in Fig.  1 :

A priori defined sample use cases: Use cases which require access to a subset of scientific publications that can be specified prior to the execution of the use case. For example, gathering the list of all trialled treatments for a particular disease in the period 2000–2010 is a typical example of such a use case.

Undefined sample use cases: Use cases which cannot be completed using data samples that are defined a priori. The execution of such use cases might require access to data not known prior to the execution or may require access to all data available. Plagiarism detection is a typical example of such use case.

Example uses cases of text and data mining of scientific literature. Depending on data needs, TDM uses can be categorised into a) a priori defined sample use cases, and b) undefined sample use cases. Furthermore, TDM use cases can broadly be categorised into 1) indirect applications which aim to improve access to and organisation of literature and 2) direct applications which focus on answering specific questions or gaining insights.

However, there are a number of factors that significantly complicate access to data for these applications. The needed data is often spread across many publishers, repositories, and other databases, often lacking interoperability (these factors will be further discussed in the next section). Consequently, researchers and developers working in these areas typically invest a considerable amount of time in corpus collection, which could be up to 90% of the total investigation time 5 . For many, this task can even prove impossible due to technical restrictions and limitations of publisher platforms, some of which will be discussed in the next section. Consequently, there is a need for a global, continuously updated, and downloadable dataset of full text publications to enable such analysis.

Challenges in machine access to scientific literature

Probably the largest obstacle to the effective and timely retrieval of relevant research literature is that it may be stored in a wide variety of locations with little to no interoperability: repositories of individual institutions, publisher databases, conference and journal websites, pre-print databases, and other locations, each of which typically offers different means for accessing their data. While repositories often implement a standard protocol for metadata harvesting, the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH), publishers typically allow access to their data through custom made APIs, which are not standardised and are subject to changes 6 . Other data sources may provide static data dumps in a variety of formats or not offer programmatic access to their data at all.

However, even when publication metadata can be obtained, other steps involved in the data collection process complicate the creation of a final dataset suitable for TDM applications. For example, the identification of scientific publications within all downloaded documents, matching these publications correctly to the original publication metadata, and their conversion from formats used in publishing, such as the PDF format, into a textual representation suitable for text and data mining, are just some of the additional difficulties involved in this process. The typical minimum steps involved in this process are illustrated in Fig.  2 . As there are no widely adopted solutions providing interoperability across different platforms, custom harvesting solutions need to be created for each.

Example illustration of the data collection process. The figure depicts the typical minimum steps which are necessary to produce a dataset for TDM of scientific literature. Depending on the use case, tens or hundreds of different data sources may need to be accessed, each potentially requiring a different process–for example accessing a different set of API methods or a different process for downloading publication full text. Furthermore, depending on the use case, additional steps may be needed, such as extraction of references, identification of duplicate items or detection of the publication’s language. In the context of CORE, we provide the details of this process in Section Methods.

Challenges in systematically gathering open access research literature

Open access journals and repositories are increasingly becoming the central providers of open access content, in part thanks to the introduction of funder and institutional open access policies 7 . Open access repositories include institutional repositories such as the University of Cambridge Repository https://www.repository.cam.ac.uk/ , and subject repositories such arXiv https://arxiv.org/ . As of February 2023, there are 6,015 open access repositories indexed in the Directory of Open Access Repositories http://v2.sherpa.ac.uk/opendoar/ (OpenDOAR), as well as 18,935 open access journals indexed in the Directory of Open Access Journals https://doaj.org/ (DOAJ). However, open access research literature can be stored in a wide variety of other locations, including publisher and conference websites, individual researcher websites, and elsewhere. Consequently, a system for harvesting open access content needs to be able to harvest effectively from thousands of data providers. Furthermore, a large number of open access repositories (69.4% of repositories indexed in OpenDOAR as of January 2018) expose their data through the OAI-PMH protocol while often not providing any alternatives. An open access harvesting system therefore also needs to be able to effectively utilise OAI-PMH for open access content harvesting. However, these two requirements–harvesting from thousands of data providers and utilising OAI-PMH for content harvesting–pose a number of significant scalability challenges.

Challenges related to harvesting from thousands of data providers

Open access data providers vary greatly in size, with some hosting millions of documents while others host a significantly lower number. New documents are added and old documents are often updated by data providers daily.

Different geographic locations and internet connection speeds may result in vastly differing times needed to harvest information from different providers, even when their size in terms of publication numbers is the same. As illustrated in Table  1 , there are also a variety of OAI-PMH implementations across commonly used repository platforms providing significantly different harvesting performance. To construct this table, we analysed OAI-PMH metadata harvesting performances of 1,439 repositories in CORE, covering eight different repository platforms. It should be noted that the OAI-PMH protocol only necessitates metadata to be expressed in the Dublin Core (DC) format. However, it also can be extended to express the metadata in other formats. Because the Dublin-Core standard is constrained to just 15 elements, it is not uncommon for OAI-PMH repositories to also use and extended metadata format such as Rioxx ( https://rioxx.net ) or the OpenAIRE Guidelines ( https://www.openaire.eu/openaire-guidelines-for-literature-institutional-and-thematic-repositories ).

Additionally, harvesting is limited not only by factors related to the data providers, but also by the compute resources (hardware) available to the aggregator. As many use cases listed in the Introduction, such as in plagiarism detection or systematic review automation, require access to very recent data, ensuring that the harvested data stays recent and that the compute resources are utilised efficiently both pose significant challenges.

To overcome these challenges, we designed the CORE Harvesting System (CHARS) which relies on two key principles. The first is the application of the microservices software principles to open access content harvesting 8 . The second is our strategy we denote pro-active harvesting , which means that providers are scheduled automatically according to current need. This strategy is implemented in the harvesting Scheduler (Section CHARS_architecture). The Scheduler uses a formula we designed for prioritising data providers.

The combination of the Scheduler with CHARS microservices architecture enables us to schedule harvesting according to current compute resource utilisation, thus greatly increasing our harvesting efficiency. Since switching from a fixed-schedule approach described above to pro-active harvesting, we have been able to greatly improve the data recency of our collection as well as to increase the size of the collection threefold within the span of three years.

Challenges related to the use of OAI-PMH protocol for content harvesting

As explained above, OAI-PMH is currently the standard method for exchanging data across repositories. While the OAI-PMH protocol was originally been designed for metadata harvesting only, it has been, due to its wide adoption and lack of alternatives, used as an entry point for full text harvesting. Full text harvesting is achieved by extracting URLs from the metadata records collected through OAI-PMH, the extracted URLs are then used to discover the location of the actual resource 9 . However, there are a number of limitations of the OAI-PMH protocol which make it unsuitable for large-scale content harvesting:

It directly supports only metadata harvesting, meaning additional functionality has to be implemented in order to use it for content harvesting.

The location of full text links in the OAI-PMH metadata is not standardised and the OAI-PMH metadata records typically contain multiple links. From the metadata it is not clear which of these links points to the described representation of the resource and in many cases none of them does so directly. Therefore, all possible links to the resource itself have to be extracted from the metadata and tested to identify the correct resource. Furthermore, OAI-PMH does not facilitate any validation for ensuring the discovered resource is truly the described resource. In order to overcome this issues, the adoption of the RIOXX https://rioxx.net/ metadata format or the OpenAIRE guidelines https://guidelines.openaire.eu/ has been promoted. However, the issue of unambiguously connecting metadata records and the described resource is still present.

The architecture of the OAI-PMH protocol is inherently sequential, which makes it ill-suited for harvesting from very large repositories. This is because the processing of large repositories cannot be parallelised and it is not possible to recover the harvesting in case of failures.

Scalability across different implementations of OAI-PMH differs dramatically. Our analysis (Table  1 ) shows that performance can differ significantly also when only a single repository software is considered 10 .

Other limitations include difficulties in incremental harvesting, reliability issues, metadata interoperability issues, and scalability issues 11 .

We have designed solutions to overcome a number of these issues, which have enabled us to efficiently and effectively utilise OAI-PMH to harvest open access content from repositories. We present these solutions in Section Using OAI-PMH for content harvesting. While we currently rely on a variety of solutions and workarounds to enable content harvesting through OAI-PMH, most of the limitations listed in this section could also be addressed by adopting more sophisticated data exchange protocols, such as the ResourceSync ( http://www.openarchives.org/rs/1.1/resourcesync ) protocol which was designed with content harvesting in mind 10 and the adoption in the systems of data providers we support.

Our solution

In the above sections we have highlighted a critical need for many researchers and organisations globally for large-scale always up-to-date seamless machine access to scientific literature originating from thousands of data providers at full text level. Providing this seamless access has become both a defining goal and a feature of CORE and has enabled other researchers to design and test innovative methods on CORE data, often powered by artificial intelligence processes. In order to put together this vast continuously updated dataset, we had to overcome a number of research challenges, such as those related to the lack of interoperability, scalability, regular content synchronisation, content redundancy and inconsistency. Our key innovation in this area is the improvement of the process of aggregating research literature , as specified in the Introduction section.

This underpinning research has allowed CORE to become a leading provider of open access papers. The amount of data made available by CORE has been growing since 2011 12 and is continuously kept up to date. As of February 2023, CORE provides access to over 291 million metadata records and 32.8 million full text open access articles, making it the world’s largest archive of open access research papers, significantly larger than PubMed, arXiv and JSTOR datasets.

Whilst there are other publication databases that could be initially viewed as similar to CORE, such as BASE or Unpaywall, we will demonstrate the significant differences that set CORE apart and show how it provides access to a unique, harmonised corpus of Open Access literature. A major difference between these existing services is that CORE is completely free to use for the end user, it hosts full text content, and offers several methods for accessing its data for machine processing. Consequently, it removes the need to harvest and pre-process full text for text mining, since CORE provides plain text access to the full texts via its raw data services, eliminating the need for text and data miners to work on PDF formats. A detailed comparison of other publication databases is provided in the Discussion. In addition, CORE enables building powerful services on top of the collected full texts, supporting all the categories of use cases outlined in the Use cases section.

As of today, CORE provides three services for accessing its raw data: API, dataset, and a FastSync service. The CORE API provides real-time machine access to both metadata and full texts of research papers. It is intended for building applications that need reliable access to a fraction of CORE data at any time. CORE provides a RESTful API. Users can register for an API key to access the service. Full documentation and Python notebooks containing code examples can be found on the CORE documentation pages online ( https://api.core.ac.uk/docs/v3 ). The CORE Dataset can be used to download CORE data in bulk. Finally, CORE FastSync enables third party systems to keep an always up to date copy of all CORE data within their infrastructure. Content can be transferred as soon as it becomes available in CORE using a data synchronisation service on top of the ResourceSync protocol 13 optimised by us for improved synchronisation scalability with an on-demand resource dumps capability. CORE FastSync provides fast, incremental and enterprise data synchronisation.

CORE is the largest up-to-date full text open access dataset as well as one of the most widely used services worldwide supporting access to freely available research literature. CORE regularly releases data dumps licensed as ODC-By, making the data freely available for both commercial and non-commercial purposes. Access to CORE data via the API is provided freely to individuals conducting work in their own personal capacity and to public research organisations for unfunded research purposes. CORE offers licenses to commercial organisations wanting to use CORE services to obtain a convenient way of accessing CORE data with a guaranteed level of service support. CORE is operated as a not-for-profit entity by The Open University and this business model makes it possible for CORE to remain free for the >99.99% of its users.

A large number of commercial organisations have benefited from these licenses in areas as diverse as plagiarism detection in research, building specialised scholarly publication search engines, developing scientific assistants and machine translation systems and supporting education etc. https://core.ac.uk/about/endorsements/partner-projects . The CORE data services–CORE API and Dataset, have been used by over 7,000 experts to analyse data, develop text-mining applications and to embed CORE into existing production systems.

Additionally, more than 70 repository systems have registered to use the CORE Recommender and the service is notably used by prestigious institutions, including the University of Cambridge and by popular pre-prints services such as arXiv.org. Other CORE services are the CORE Discovery and the CORE Repository Dashboard. The first was released on July 2019 and at the time of writing it has more than 5000 users. The latter is a tool designed specifically for repository managers which provides access to a range of tools for managing the content within their repositories. The CORE Repository Dashboard is currently used by 499 users from 36 countries.

In the rest of this paper we describe the CORE dataset and the methods of assembling it and keeping it continuously up-to-date. We also present the services and tools built on top of the aggregated corpus and provide several examples of how the CORE dataset has been used to create real-world applications addressing specific use-cases.

As highlighted in the Introduction, CORE is a continuously growing dataset of scientific publications for both human and machine processing. As we will show in this section, it is a global dataset spanning all disciplines and containing publications aggregated from more than ten thousand data providers including disciplinary and institutional repositories, publishers, and other databases. To improve access to the collected publications, CORE performs a number of data enrichment steps. These include metadata and full text extraction, language and DOI detection, and linking with other databases. Furthermore, CORE provides a number of services which are built on top of the data: a publications recommender ( https://core.ac.uk/services/recommender/ ), CORE Discovery service ( https://core.ac.uk/services/discovery/ ) (a tool for discovering OA versions of scientific publications), and a dashboard for repository managers ( https://core.ac.uk/services/repository-dashboard/ ).

Dataset size

As of February 2023, CORE is the world’s largest dataset of open access papers (comparison with other systems is provided in the Discussion). CORE hosts over 291 million metadata records including over 34 million articles with full text written in 82 languages and aggregated from over ten thousand data providers located in 150 countries. Full details of CORE Dataset size are presented in Table  2 . In the table, “Metadata records” represent all valid (not retracted, deleted, or for some other reason withdrawn) records in CORE. It can be seen that about 13% of records in CORE contain full text. This number represents records for which a manuscript was successfully downloaded and converted to plain text. However, a much higher proportion of records contains links to additional freely available full text articles hosted by third-party providers. Based on analysing a subset of our data, we estimate that about 48% of metadata records in CORE fall into this category, indicating that CORE is likely to contain links to open access full texts for 139 million articles. Due to the nature of academic publishing there will be instances where multiple versions of the same paper are deposited in different repositories. For example, an early version of an article can be deposited by an author to a pre-print server such as arXiv or BiorXiv and then a later version uploaded to an institutional repository. Identifying and matching these different versions is a significant undertaking. CORE has carried out research to develop techniques based on locality sensitive hashing for duplicates identification 8 and integrated these into its ingestion pipeline to link versions of papers from across the network of OA repositories and group these under a single works entity. The large number of records in CORE translates directly into the size of the dataset in bytes as the uncompressed version of the dataset including PDFs is about 100 TB. The compressed version of the CORE dataset with plain texts only amounts to 393 GB and uncompressed to 3.5 TBs.

Recent studies have estimated that around 24%–28% of all articles are available free to read 2 , 14 . There are a number of reasons why the proportion of full text content in CORE is lower than these estimates. The main reason is likely that a significant proportion of the free to read articles represents content hosted on platform with many restrictions for machine accessibility, i.e. some repositories severely restrict or fully prohibit content harvesting 9 .

The growth of CORE has been made possible thanks to the introduction of a novel harvesting system and the creation of an efficient harvesting scheduler, both of which are described in the Methods section. The growth of metadata and full text records in CORE is shown in Fig.  3 . Finally, Fig.  4 shows age of publications in CORE.

Growth of records in CORE per month since February 2012. “Full text growth” represents growth of records containing full text, while “Metadata growth” represents growth of records without full text, i.e. the two numbers do not overlap. The two area plots are stacked on top of each other, their sum therefore represents the total number of records in CORE.

Age of publications in CORE. Similarly as in Fig.  3 , the “Metadata” and “Full text” records bars are stacked on top of each other.

Data sources and languages

As of February 2023, CORE was aggregating content from 10,744 data sources. These data sources include institutional repositories (for example the USC Digital Library or the University of Michigan Library Repository), academic publishers (Elsevier, Springer), open access journals (PLOS), subject repositories, including those hosting eprints (arXiv, bioRxiv, ZENODO, PubMed Central) and aggregators (e.g. DOAJ). The ten largest data sources in CORE are shown in Table  3 . To calculate the total number of data providers in CORE, we consider aggregators and publishers as one data source despite each aggregating data from multiple sources. A full list of all data providers can be found on the CORE website. ( https://core.ac.uk/data-providers ).

The data providers aggregated by CORE are located in 150 different countries. Figure  5 shows the top ten countries in terms of number of data providers aggregated by CORE from each country alongside the top ten languages. The geographic spread of repositories is largely reflective of the size of the research economy in those countries. We see the US, Japan, Germany, Brazil and the UK all in the top six. One result that at first may appear surprising is the significant number of repositories in Indonesia, enough to place them at the top of the list. An article in Nature in 2019 showed that Indonesia may be the world’s OA leader, finding that 81% of 20,000 journal articles published in 2017 with an Indonesia-affiliated author are available to read for free somewhere online. ( https://www.nature.com/articles/d41586-019-01536-5 ). Additionally, there are a large number of Indonesian open-access journals registered with Crossref. This subsequently leads to a much higher number of individual repositories in this country.

Top ten languages and top ten provider locations in CORE.

As part of the enrichment process, CORE performs language detection. Language is either extracted from the attached metadata where available or identified automatically from full text in case it is not available in metadata. More than 80% of all documents with language information are in English. Overall, CORE contains publications in a variety of languages, the top 10 of which are shown in Fig.  5 .

Document types

The CORE dataset comprises a collection of documents gathered from various sources, many of which contain articles of different types. Consequently, aside of research articles from journals and conferences, it includes other types of research outputs such as research theses, presentations, and technical reports. To distinguish different types of articles, CORE has implemented a method of automatically classifying documents into one of the following four categories 15 : (1) research article, (2) thesis, (3) presentation, (4) unknown (for articles not belonging into any of the previous three categories). This method is based on a supervised machine learning model trained on article full texts. Figure  6 shows the distribution of articles in CORE into these four categories. It can be seen that the collection aggregated by CORE consists predominantly of research articles. We have observed in the data collected from repositories that the vast majority of research theses deposited in repositories has full text associated with the metadata. As this is not always the case for research articles, and as Fig.  6 is produced on articles with full text only, we expect that the proportion of research articles compared to research theses in CORE is actually higher across the entire collection.

Distribution of document types.

Research disciplines

To analyse the distribution of disciplines in CORE we have leveraged a third-party service. Figure  7 shows a subject distribution of a sample of 20,758,666 publications in CORE. For publications with multiple subjects we count the publication towards each discipline.

Subject distribution of a sample of 20,758,666 CORE publications.

The subject for each article was obtained using Microsoft Academic ( https://academic.microsoft.com/home ) prior to its retirement in November 2021. Our results are consistent with other studies, which have reported Biology, Medicine, and Physics to be the largest disciplines in terms of number of publications 16 , 17 , suggesting that the distribution of articles in CORE is representative of research publications in general.

Additional CORE Tools and Services

CORE has built several additional tools for a range of stakeholders including institutions, repository managers and researchers from across all scientific domains. Details of usage of these services is covered in the Uptake of CORE section.

The Dashboard provides a suite of tools for repository management, content enrichment, metadata quality assessment and open access compliance checking. Further, it can provide statistics regarding content downloads and suggestions for improving the efficiency of harvesting and the quality of metadata.

CORE Discovery helps users to discover freely accessible copies of research papers. There are several methods for interacting with the Discovery tool. First, as a plugin for repositories, enriching metadata only pages in repositories with links to open access copies of full text documents. Second, via a browser extension for researchers and anyone interested in reading scientific documents. And finally as an API service for developers.

Recommender

The recommender is a plugin for repositories, journal systems and web interfaces that provides suggestions on relevant articles to the one currently displayed. Its purpose is to support users in discovering articles of interest from across the network of open access repositories. It is notably used by prestigious institutions, including the University of Cambridge and by popular pre-prints services such as arXiv.org.

OAI Resolver

An OAI (Open Archives Initiative) identifier is a unique identifier of a metadata record. OAI identifiers are used in the context of repositories using the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH). OAI Identifiers are viable persistent identifiers for repositories that can be, as opposed to DOIs, minted in a distributed fashion and cost-free, and which can be resolvable directly to the repository rather than to the publisher. The CORE OAI Resolver can resolve any OAI identifier to either a metadata page of the record in CORE or route it directly to the relevant repository page. This approach has the potential to increase the importance of repositories in the process of disseminating knowledge.

Uptake of CORE

As of February 2023, CORE averages over 40 million monthly active users and is the top 10th website in the category Science and Education according to SimilarWeb ( https://www.similarweb.com/ ). There are currently 4,700 registered API users and 2,880 registered dataset users. The CORE Dashboard is currently used by 499 institutional repositories to manage their open access content, monitor content download statistics, manage issues with metadata within the repository and ensure compliance with OA funder policies, notably REF in the U.K. The CORE Discovery plugin has been integrated into 434 repositories and the browser extension has been downloaded by more than 5,000 users via the Google Chrome Web Store ( https://chrome.google.com/webstore/category/extensions ). The CORE Recommender has been embedded in 70 repository systems including the University of Cambridge and arXiv.

In this section we discuss differences between CORE and other open access aggregation services and present several real-word use cases where CORE was used to develop services to support science. In this section we also present our future plans.

Existing open access aggregation services

Currently there are a number of open access aggregation services available (Table  4 ), with some examples being BASE ( https://base-search.net/ ), OpenAIRE ( https://www.openaire.eu/ ), Unpaywall ( http://unpaywall.org/ ), Paperity ( https://paperity.org/ ). BASE (Bielfield Academic Search Engine) is a global metadata harvesting service. It harvests repositories and journals via OAI-PMH and exposes the harvested content through an API and a dataset. OpenAIRE is a network of open access data providers who support open access policies. Even though in the past the project focused on European repositories, it has recently expanded by including institutional and subject repositories from outside Europe. A key focus of OpenAIRE is to assist the European Council to monitor compliance of its open access policies. OpenAIRE data is exposed via an API. Paperity is a service which harvests publications from open access journals. Paperity harvests both metadata and full text but does not host full texts. SHARE (Shared Access Research Ecosystem) is a harvester of open access content from US repositories. Its aim is to assist with the White House Office of Science and Technology Policy (OSTP) open access policies compliance. Even though SHARE harvests both metadata and full text it does not host the latter. Unpaywall is not primarily a harvester, but rather collects content from Crossref, whenever a free to read available version can be retrieved. It processes both metadata and full text but does not host them. It exposes the discovered links to documents through an API.

CORE differs from these services in a number of ways. CORE is currently the largest database of full text OA documents. In addition, CORE offers via its API a rich metadata record for each item in its collection which includes additional enrichments, contrary, for example, to Unpaywall’s API, which focuses only on delivering to the user information as to whether a free to read version is available. CORE also provides the largest number of links to OA content. To simplify access to data for end users it provides a number of ways for accessing its collection. All of the above services are free to use for research purposes however both CORE and Unpaywall also offer services to commercial partners on a paid-for basis.

Existing publication databases

Apart from OA aggregation services, a number of other services exists for searching and downloading scientific literature (Table  5 ). One of the main publication databases is Crossref ( https://www.crossref.org/ ), an authoritative index of DOI identifiers. Its primary function is to maintain metadata information associated with each DOI. The metadata stored by Crossref includes both OA and non-OA records. Crossref does not store publication full text, but for many publications provides full text links. As of February 2023, 5.9 m records in Crossref were associated with an explicit Creative Commons license (we have used the Crossref API to determine this number). Although Crossref provides an API, it does not offer its data for download in bulk, or provide a data sync service.

The remaining services from Table  5 can be roughly grouped into the following two categories: 1) citation indices, 2) academic search engines and scholarly graphs. The two major citation indices are Elsevier’s Scopus ( https://www.elsevier.com/solutions/scopus ) and Clarivate’s Web of Science ( https://clarivate.com/webofsciencegroup/solutions/web-of-science/ ), both of which are premium subscription services. Google Scholar, the best known academic search engine does not provide an API for accessing its data and does not permit crawling its website. Semantic Scholar ( https://www.semanticscholar.org/ ) is a relatively new academic search service which aims to create an “intelligent academic search engine” 18 . Dimensions ( https://www.dimensions.ai/ ) is a service focused on data analysis. It integrates publications, grants, policy documents, and metrics. 1findr ( https://1findr.1science.com/home ) is a curated abstract indexing service. It provides links to full text, but no API or a dataset for download.

The added value of CORE

There are other services that claim to provide access to a large dataset of open access papers. In particular, Unpaywall 2 , claim to provide access to 46.4 million free to read articles, and BASE, who state they provide access to full texts of about 60% of their 300 million metadata records. However, these statistics are not directly comparable to the numbers we report and are a product of a different focus of these two projects. This is because both the analysis of BASE and now Unpaywall define “providing access to” in terms of having a list of URLs from which a human user can navigate to the full text of the resource. This means that both Unpaywall and BASE do not collect these full text resources, which is also why they do not face many of the challenges we described in the Introduction. Using this approach, we could say that the CORE Dataset provides access to approximately 139 million full texts, i.e. about 48% of our 291 million metadata records point to a URL from which a human can navigate to the full text. However, to people concerned with text and data mining of scientific literature, it makes little sense to count URLs pointing to many different domains on the Web as the number of full texts made available.

As a result, our 32.8 million statistic refers to the number of OA documents we identified, downloaded, extracted text from, validated their relationship to the metadata record and the full texts of which we host on the CORE servers and make available to others. In contrast, BASE and Unpaywall do not aggregate the full texts of the resources they provide access to and consequently do not offer the means to interact with the full texts of these resources or offer bulk download capability of these resources for text analytics over scholarly literature.

We have also integrated CORE data with the OpenMinTeD infrastructure, a European Commission funded project which aimed to provide a platform for text mining of scholarly literature in the cloud 6 .

A number of academia and industry partners have utilised CORE in their services. In this section we present three existing uses of CORE demonstrating how CORE can be utilised to support text and data mining use cases.

Since 2017, CORE has been collaborating with a range of scholarly search and discovery systems. These include Naver ( https://naver.com/ ), Lean Library ( https://www.leanlibrary.com/ ) and Ontochem ( https://ontochem.com/ ). As part of this work, CORE serves as a provider of full text copies of reserch papers to existing records in these systems (Lean Library) or even supplies both metadata and full texts for indexing (Ontochem, NAVER). This collaboration also benefits CORE’s data providers as it expands and increases the visibility of their content.

In 2019, CORE entered into a collaboration with Turnitin, a global leader in plagiarism detection software. By using the CORE FastSync service, Turnitin’s proprietary web crawler searches through CORE’s global database of open access content and metadata to check for text similarity. This partnership enables Turnitin to significantly enlarge its content database in a fast and efficient manner. In turn, it also helps protect open access content from misuse, thus protecting authors and institutions.

As of February 2023, CORE Recommender 19 is actively running in over 70 repositories including the University of Cambridge institutional repository and arXiv.org among others. The purpose of the recommender is to improve the discoverability of research outputs by providing suggestions for similar research papers both within the collection of the hosting repository and the CORE collection. Repository managers can install the recommender to advance the accessibility of other scientific papers and outreach to other scientific communities, since the CORE Recommender acts as a gate to millions of open access research papers. The recommender is integrated with the CORE search functionality and is also offered as a plugin for all repository software, for example EPrints, DSpace, etc. as well as open access journals and any other webpage. Based on the fact that CORE harvests open repositories, the recommender only displays research articles where the full text is available as open access, i.e. for immediate use, without access barriers or limited rights’ restrictions. Through the recommender, CORE promotes the widest discoverability and distribution of the open access scientific papers.

Future work

An ongoing goal of CORE is to keep growing the collection to become a single point of access to all of world’s open access research. However, there are a number of other ways we are planning to improve both the size and ease of access to the collection. The CORE Harvesting System was designed to enable adding new harvesting steps and enrichment tasks. There remains scope for adding more of such enrichments. Some of these are machine learning powered, such as classification of scientific citations 20 . Further, CORE is currently developing new methodologies to identify and link different versions of the same article. The proposed system, titled CORE Works, will leverage CORE’s central position in the OA infrastructure landscape and will link different versions of the same paper using a unique identifier. We will continue to keep linking the CORE collection to scholarly entities from other services, thereby making CORE data participate in a global scholarly knowledge graph.

In the Introduction section we focused on a a number of challenges researchers face when collecting research literature for text and data mining. In this section, we instead focus on the perspective of a research literature aggregator, i.e. a system whose goal is to continuously provide seamless access to research literature aggregated from thousands of data providers worldwide in a way that enables the resulting research publication collection to be used by others in production applications. We describe the challenges we had to overcome to build this collection and to keep it continuously up-to-date, and present the key technical innovations which allowed us to greatly increase the size of the CORE collection and become a leading provider of open access literature which we illustrate using our content growth statistics.

CORE Harvesting system (CHARS)

CORE Harvesting System (CHARS) is the backbone of our harvesting process. CHARS uses the Harvesting Scheduler (Section CHARS_architecture) to select data providers to be processed next. It manages all the running processes (tasks) and ensures the available compute resources are well utilised.

Prior to implementing CHARS, CORE was centralised around data providers rather than around individual tasks needed to harvest and process these data providers (e.g. metadata download and parsing, full text download, etc.). Consequently, even though the scaling up and the continuation of this system was possible, the infrastructure was not horizontally scalable and the architecture suffered from tight coupling of services. This was not consistent with CORE’s high availability requirements and was regularly causing problems in the complexity of maintenance. In response to these challenges, we designed CHARS using a microservices architecture, i.e. using small manageable autonomous components that work together as part of a larger infrastructure 21 . One of the key benefits of microservices-oriented architecture is that the implementation focus can be put on the individual components which can be improved and redeployed as frequently as needed and independently of the rest of the infrastructure. As the process of open access content harvesting can be inherently split into individual consecutive tasks, a microservices-oriented architecture presents a natural fit for aggregation systems like CHARS.

Tasks involved in open access content harvesting

The harvesting process can be described as a pipeline where each task performs a certain action and where the output of each task feeds into the next task. The input to this pipeline is a set of data providers and the final output is a system populated with records of research papers available from them. The main types of key tasks currently performed as part of CORE’s harvesting system are (Fig.  8 ):

Metadata download: The metadata exposed by a data provider via OAI-PMH are downloaded and stored in the file system (typically as an XML). The downloading process is sequential, i.e. a repository provides typically between 100–1,000 metadata records per request and a resumption token. This token is then used to provide the next batch. As a result, full harvesting can a significant amount of time (hours-days) for large data providers. Therefore, this process has been implemented to provide resilience to a range of communication failures.

Metadata extraction : Metadata extraction parses, cleans, and harmonises the downloaded metadata and stores them into the CORE internal data structure (database). The harmonisation and cleaning process addresses the fact that different data providers/repository platforms describe the same information in different ways (syntactic heterogeneity) as well as having different interpretations for the same information (semantic heterogeneity).

Full text download : Using links extracted from the metadata CORE attempts to download and store publication manuscripts. This process is non-trivial and is further described in the Using OAI-PMH for content harvesting section.

Information extraction : Plain text from the downloaded manuscripts is extracted and processed to create a semi-structured representation. This process includes a range of information extraction tasks, such as references extraction.

Enrichment : The enrichment task works by increasing both metadata and full text harvested from the data providers with additional data from multiple sources. Some of the enrichments are performed directly by specific tasks in the pipeline such as language detection and document type detection. The remaining enrichments that involve external datasets are performed externally and independently to the CHARS pipeline and ingested into the dataset as described in the Enrichments section.

Indexing : The final step in the harvesting pipeline is indexing the harvested data. The resulting index powers CORE’s services, including search, API and FastSync.

CORE Harvesting Pipeline. Each tasks’ output produces the input for the following task. In some cases the input is considered as a whole, for example all the content harvested from a data provider, while in other cases, the output is split in multiple small tasks performed on a record level.

Scalable infrastructure requirements

Based on the experience obtained while developing and maintaining our harvesting system as well as taking into consideration the features of the CiteSeerX 22 architecture, we have defined a set of requirements for a scalable harvesting infrastructure 8 . These requirements are generic and apply to any aggregation or digital library scenario. These requirements informed and are reflected in the architecture design of CHARS (Section CHARS architecture):

Easy to maintain: The system should be easy to manage, maintain, fix, and improve.

High levels of automation: The system should be completely autonomous while allowing manual interaction.

Fail fast: Items in the harvesting pipeline should be validated immediately after a task is performed, instead of having only one and final validation at the end of the pipeline. This has the benefit of recognising issues and enabling fixes earlier in the process.

Easy to troubleshoot: Possible code bugs should be easily discerned.

Distributed and scalable: The addition of more compute resources should allow scalability, be transparent and replicable.

No single point of failure: A single crash should not affect the whole harvesting pipeline, individual tasks should work independently.

Decoupled from user-facing systems: Any failure in the ingestion processing services should not have an immediate impact on user-facing services.

Recoverable: When a harvesting task stops, either manually or due to a failure, the system should be able to recover and resume the task without manual intervention.

Performance observable: The system’s progress must be properly logged at all times and overlay monitoring services should be set up to provide a transparent overview of the services’ progress at all times, to allow early detection of scalability problems and identification of potential bottlenecks.

CHARS architecture

An overview of CHARS is shown in Fig.  9 . The system consists of the following main software components:

Scheduler: it becomes active when a task finishes. It monitors resource utilisation and selects and submits data providers to be harvested.

Queue (Qn): a messaging system that assists with communication between parts of the harvesting pipeline. Every individual task, such as metadata download, metadata parsing, full text download, and language detection, has its own message queue.

Worker (W i ): an independent and standalone application capable of executing a specific task. Every individual task has its own set of workers.

CORE Harvesting System.

A complete harvest of a data provider can be described as follows. When an existing task finishes, the scheduler is activated and informed of the result. It then uses the formula described in Appendix A to assign a score to each data provider. Depending on current resource utilisation, i.e. if there are any idle workers, and the number of data providers already scheduled for harvesting, the data provider with the highest score is then placed in the first queue Q 1 which contains data providers scheduled for metadata download. Once one of the metadata download workers W i -W j becomes available, a data provider is taken out of the queue and a new download of its metadata starts. Upon completion, the worker notifies the scheduler and, if the task is completed successfully, places the data provider in the next queue. This process continues until the data provider passes through the entire pipeline.

While some of the tasks in the pipeline need to be performed at the granularity of data providers, specifically metadata download and parsing, other tasks, such as full text extraction and language detection, can be performed at the granularity of individual records. While these tasks are originally scheduled at the granularity of data providers, only the individual records of a selected data provider which require processing are subsequently independently placed in the appropriate queue. Workers assigned to these tasks then process the individual records in the queue and they move through the pipeline once completed.

A more detailed description of CHARS, which includes technologies used to implement it, as well as other details can be found in 8 .

The harvesting scheduler is a component responsible for identifying data providers which need to be harvested next and placing these data providers in the harvesting queue. In the original design of CORE, our harvesting schedule was created manually, assigning the same harvesting frequency to every data provider. However, we found this approach inefficient as it does not scale due to the varying data providers size, differences in the update frequency of their databases and the maximum data delivery speeds of their repository platforms. To address these limitations, we designed the CHARS scheduler according to our new concept of “pro-active harvesting.” This means that the scheduler is event driven. It is triggered whenever the underlying hardware infrastructure has resources available to determine which data provider should be harvested next. The underlying idea is to maximise the number of ingested documents over a unit of time. The pseudocode and the formula we use to determine which repository to harvest next is described in Algorithm 1.

The size of the metadata download queue, i.e. the queue which represents an entry into the harvesting pipeline, is kept limited in order to keep the system responsive to the prioritisation of data providers. A long queue makes prioritising data providers harder, as it is not known beforehand how long the processing of a particular data provider will take. An appropriate size of the queue ensures a good balance between the reactivity and utilisation of the available resources.

Using OAI-PMH for content harvesting

We now describe the third key technical innovation which enables us to harvest full text content (as opposed to just metadata) from data providers using the OAI-PMH protocol. This process represents one step in the harvesting pipeline (Fig.  9 ), specifically, the third step which is activated after data provider metadata have been downloaded and parsed.

The OAI-PMH protocol was originally designed for metadata harvesting only, but due to its wide adoption and lack of alternatives it has been used as an entry point for full text harvesting from repositories. Full text harvesting is achieved by using URLs found in the metadata records to discover the location of the actual resource and subsequently downloading it 9 . We summarised the key challenges of this approach in the Challenges related to the use of OAI-PMH protocol for content harvesting section. The algorithm follows a depth first search strategy with prioritisation and finishes as soon as the first matching document is found.

The procedure works in the following way. First, all metadata records from a selected data provider with no full text are collected. Those records for which full text download was attempted within the retry period ( RP ) (usually six months) are filtered out. This is to avoid repeatedly downloading URLs that do not lead to the sought after documents. The downside of this approach is that if a data provider updates a link in the metadata, it might take up to the duration of the retry period to acquire the full text.

Algorithm 1

Next, the records are further filtered using a set of rules and heuristics we developed to a) increase the chances of identifying the URL leading to the described document quickly and b) to ensure that we identify the correct document. These filtering rules include:

Accepted file extensions: URLs are filtered according to a list of accepted file extensions. URLs ending with extensions such as .pptx that clearly indicate that the URL does not link to the required resource are removed from the list.

Same domain policy: URLs in the OAI-PMH metadata can link to any resources and domains. For example, a common practice is to provide a link to the associated presentation, dataset, or another related resource. As these are often stored in external databases, filtering out all URLs that lead to an external domain, i.e. domain different than the domain of the data provider, presents a simple method of avoiding the download of resources which with very high likelihood do not represent the target document. Exceptions include dx.doi.org and hdl.handle.net domains whose purpose is to provide a persistent identifier pointing to the document. The same domain policy is disabled for data providers which are aggregators and link to many different domains by design.

Provider-specific crawling heuristics: Many data providers follow a specific pattern when composing URLs. For example, a link to a full text document may be composed of the following parts: data provider URL  +  record handle  +  .pdf . For data providers utilising such patterns, URLs may be composed automatically where the relevant information (record handle) is known to us from the metadata. These generated URLs are then added to the list of URLs obtained from the metadata.

Prioritising certain URLs: As it is more likely for PDF URL to contain the target record than for an HTML URL, the final step is to sort URLs according to file and URL type. Highest priority is assigned to URLs that uses repository software specific patterns to identify full text, document, and PDF filetypes, while the lowest priority is assigned to hdl.handle.net URLs.

The system then attempts to request the document at each URL and download it. After each download, checks are performed to determine whether the downloaded document represents the target record. Currently, the downloaded document has to be a valid PDF with a title matching the original metadata record. If the target record is identified, the downloaded document is stored and the download process for that record ends. If the downloaded document contains an HTML page, URLs are extracted from this page and filtered using the same method mentioned above. This is because it is common in some of the most widely used repository systems such as DSpace for the documents not to be directly referenced from within the metadata records. Instead, the metadata records typically link to an HTML overview page of the document. To deal with this problem, we use the concept of harvesting levels. A maximum harvesting level corresponds to the maximum search depth for the referenced document. The algorithm finishes either as soon as the first matching document is found or after all the available URLs up to the maximum harvesting level have been exhausted. Algorithm 2 describes our approach for collecting the full texts using the OAI-PMH protocol. The algorithm follows a depth first search strategy with prioritisation and finishes as soon as the first matching document is found.

Algorithm 2

CHARS limitations

Despite overcoming the key issues to scalable harvesting of content from repositories, there still remains a number of important challenges. The first relates to the difficulty of estimating the optimal number of workers in our system to run efficiently. While the worker allocation is still largely established empirically, we are investigating more sophisticated approaches based on formal models of distributed computation, such as Petri Nets. This will allow us to investigate new approaches to dynamically allocating and launching workers to optimise the usage of our resources.

Enrichments

Conceptually, two types of enrichment processes are used within CORE: 1) an online enrichment process enriching a single record at the time of it being processed by the CHARS pipeline and 2) a periodic offline enrichment process which enriches a record based on information in external datasets (Fig.  10 ).

CORE Offline Enrichments.

Online enrichments

Online enrichments are fully integrated into the CHARS pipeline described earlier in this section. These enrichments generally involve the application of machine learning models and rule-based tools to gather additional insights about the record, such as language detection, document type detection. As opposed to offline enrichments, online enrichments are always performed just once for a given record. The following is a list of the current enrichments performed online:

Article type detection: A machine learning algorithm assigns each publication one of the following four types: presentation, thesis, research paper, other. In the future we may include other types.

Language identification: This task uses third-party libraries to identify the language based on the full text of a document. The resulting language is then compared to the one provided by the metadata record. Some heuristics are applied to disambiguate and harmonise languages.

Offline enrichments

Offline enrichments are carried out by means of gathering a range of information from large third-party scholarly datasets (research graphs). Such information includes metadata that do not necessarily change, such as a DOI identifier, as well as metadata that evolve, such as the number of citations. Especially due to the latter, CORE performs offline enrichments periodically, i.e. all records in CORE go through this process repeatedly at specified time intervals (currently once per month).

The process is depicted in Fig.  10 . The initial mapping of a record is carried out using a DOI, if available. However, as the majority of records from repositories do not come with a DOI, we carry out a matching process against the Crossref database using a subset of metadata fields including title, authors and year. Once the mapping is performed, we can harmonise fields as well as gather a wide range of additional useful data from relevant external databases, thereby enriching the CORE record. Such data include, ORCID identifiers, citation information, additional links to freely available full texts, field of study information and PubMed identifiers. Our solution is based on a set of map-reduce tasks to enrich the dataset and implemented on a Cloudera Enterprise Data Hub ( https://www.cloudera.com/products/enterprise-data-hub.html ) 23 , 24 , 25 , 26 .

Data availability

CORE provides several large data dumps of the processed and aggregated data under the ODC-BY licence ( https://core.ac.uk/documentation/dataset ). The only condition for both commercial and non-commercial reuse of these datasets is to acknowledge the use of CORE in their outputs. Additionally, CORE makes its API and most recent data dump freely available to registered individual users and researchers. Please note that CORE claims no rights in the aggregated content itself which is open access and therefore freely available to everyone. All CORE data rights correspond to the sui generis database rights of the aggregated and processed collection.

Licences for CORE services, such as the API and FastSync, are available for commercial users wishing to benefit from convenient access to CORE data with guaranteed level of customer support. The organisation running CORE, i.e. The Open University, is a charitable organisation fully committed to the Open Research mission. CORE is a signatory of the Principles of Open Scholarly Infrastructure (POSI) ( https://openscholarlyinfrastructure.org/posse ). No profit generation is practised. Instead, CORE’s income from licences to commercial parties is used solely to provide sustainability by means of enabling CORE to become less reliant on unstable project grants, thus offsetting and reducing the cost of CORE to the taxpayer. This is done in full compliance with the principles and best practices of sustainable open science infrastructure.

Code availability

CORE consists of multiple services. Most of our source code is open source and available in our public repository on GitHub ( https://github.com/oacore/ ). As of today, we are unfortunately not yet able to provide the source code to our data ingestion module. However, as we want to be as transparent as possible with our community, we have documented in this paper the key algorithms and processes which we apply using pseudocode.

Bornmann, L. & Mutz, R. Growth rates of modern science: A bibliometric analysis based on the number of publications and cited references. JASIST 66 (11), 2215–2222 (2015).

CAS   Google Scholar  

Piwowar, H. et al . The State of OA: A large-scale analysis of the prevalence and impact of Open Access articles. PeerJ 6 , e4375 (2018).

Article   PubMed   PubMed Central   Google Scholar  

Saggion, H. & Ronzano, F. Scholarly data mining: making sense of scientific literature. 2017 ACM/IEEE Joint Conference on Digital Libraries (JCDL) : 1–2 (2017).

Kim, E. et al . Materials synthesis insights from scientific literature via text extraction and machine learning. Chemistry of Materials 29 (21), 9436–9444 (2017).

Article   CAS   Google Scholar  

Jacobs, N. & Ferguson, N. Bringing the UK’s open access research outputs together: Barriers on the Berlin road to open access. Jisc Repository (2014).

Knoth, P., Pontika, N. Aggregating Research Papers from Publishers’ Systems to Support Text and Data Mining: Deliberate Lack of Interoperability or Not? In: INTEROP2016 (2016).

Herrmannova, D., Pontika, N. & Knoth, P. Do Authors Deposit on Time? Tracking Open Access Policy Compliance. Proceedings of the 2019 ACM/IEEE Joint Conference on Digital Libraries , Urbana-Champaign, IL (2019).

Cancellieri, M., Pontika, N., Pearce, S., Anastasiou, L. & Knoth, P. Building Scalable Digital Library Ingestion Pipelines Using Microservices. Proceedings of the 11th International Conference on Metadata and Semantics Research (MTSR 2017) : 275–285. Springer (2017).

Knoth, P. From open access metadata to open access content: two principles for increased visibility of open access content. Proceedings of the 2013 Open Repositories Conference , Charlottetown, Prince Edward Island, Canada (2013).

Knoth, P.; Cancellieri, M. & Klein, M. Comparing the Performance of OAI-PMH with ResourceSync. Proceedings of the 2019 Open Repositories Conference , Hamburg, Germany (2019).

Kapidakis, S. Metadata Synthesis and Updates on Collections Harvested Using the Open Archive Initiative Protocol for Metadata Harvesting. Digital Libraries for Open Knowledge. TPDL 2018. Lecture Notes in Computer Science 11057 , 16–31 (2018).

Google Scholar  

Knoth, P. and Zdrahal, Z. CORE: three access levels to underpin open access. D-Lib Magazine 18 (11/12) (2012).

Haslhofer, B. et al . ResourceSync: leveraging sitemaps for resource synchronization. Proceedings of the 22nd International Conference on World Wide Web : 11–14 (2013).

Khabsa, M. & Giles, C. L. The number of scholarly documents on the public web. PLOS One 9 (5), e93949 (2014).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Charalampous, A. & Knoth, P. Classifying document types to enhance search and recommendations in digital libraries. Research and Advanced Technology for Digital Libraries. TPDL 2017. Lecture Notes in Computer Science 10450 , 181–192 (2017).

Rosvall, M. & Bergstrom, C. T. Maps of random walks on complex networks reveal community structure. Proceedings of the National Academy of Sciences 105 (4), 1118–1123 (2008).

Article   ADS   CAS   Google Scholar  

D’Angelo, C. A. & Abramo, G. Publication rates in 192 research fields of the hard sciences. Proceedings of the 15th ISSI Conference : 915–925 (2015).

Ammar, W. et al . Construction of the Literature Graph in Semantic Scholar. Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies , Volume 3 (Industry Papers): 84–91 (2018).

Knoth, P. et al . Towards effective research recommender systems for repositories. Open Repositories , Bozeman, USA (2017).

Pride, D. & Knoth, P. An Authoritative Approach to Citation Classification. Proceedings of the 2020 ACM/IEEE Joint Conference on Digital Libraries (JCDL 2020), Virtual–China (2020).

Newman, S. Building microservices: designing fine-grained systems. O’Reilly Media, Inc. (2015).

Li, H. et al . CiteSeer χ : a scalable autonomous scientific digital library. Proceedings of the 1st International Conference on Scalable Information Systems , ACM (2006).

Bastian, H., Glasziou, P. & Chalmers, I. Seventy-five trials and eleven systematic reviews a day: how will we ever keep up? PLoS medicine 7 (9), e1000326 (2010).

Shojania, K. G. et al . How quickly do systematic reviews go out of date? A survival analysis. Annals of internal medicine 147 (4), 224–233 (2007).

Article   PubMed   Google Scholar  

Tsafnat, G. et al . Systematic review automation technologies. Systematic reviews 3 (1), 74 (2014).

Harzing, A.-W. & Alakangas, S. Microsoft Academic is one year old: The Phoenix is ready to leave the nest. Scientometrics 112 (3), 1887–1894 (2017).

Article   Google Scholar  

Download references

Acknowledgements

We would like to acknowledge the generous support of Jisc, under a number of grants and service contracts with The Open University. These included projects CORE, ServiceCORE, UK Aggregation (1 and 2) and DiggiCORE, which was co-funded by Jisc with NWO. Since 2015, CORE has been supported in three iterations under the Jisc Digital Services–CORE (JDSCORE) service contract with The Open University. Within Jisc, we would like to thank primarily the CORE project managers, Andy McGregor, Alastair Dunning, Neil Jacobs and Balviar Notay. We would also like to thank the European Commission for funding that contributed to CORE, namely OpenMinTeD (739563) and EOSC Pilot (654021). We would like to show our gratitude to all current CORE Team members who contributed to CORE but are not authors of the manuscript, namely Valeriy Budko, Ekaterine Chkhaidze, Viktoriia Pavlenko, Halyna Torchylo, Andrew Vasilyev and Anton Zhuk. We would like to show our gratitude to all past CORE Team members who have contributed to CORE over the years, namely Lucas Anastasiou, Giorgio Basile, Aristotelis Charalampous, Josef Harag, Drahomira Herrmannova, Alexander Huba, Bikash Gyawali, Tomas Korec, Dominika Koroncziova, Magdalena Krygielova, Catherine Kuliavets, Sergei Misak, Jakub Novotny, Gabriela Pavel, Vojtech Robotka, Svetlana Rumyanceva, Maria Tarasiuk, Ian Tindle, Bethany Walker and Viktor Yakubiv, Zdenek Zdrahal and Anna Zelinska.

Author information

Drahomira Herrmannova

Present address: Oak Ridge National Laboratory Oak Ridge, Oak Ridge, TN, USA

Authors and Affiliations

Knowledge Media Institute, The Open University Walton Hall, Milton Keynes, UK

Petr Knoth, Drahomira Herrmannova, Matteo Cancellieri, Lucas Anastasiou, Nancy Pontika, Samuel Pearce, Bikash Gyawali & David Pride

You can also search for this author in PubMed   Google Scholar

Contributions

P.K. is the Founder and Head of CORE. He conceived the idea and has been the project lead since the start in 2011. He researched and created the first version of CORE, acquired funding, built the team, and has been managing and leading all research and development. M.C., L.A., S.P. and P.K. designed, worked out all technical details, and implemented significant parts of the system including CHARS, the harvesting scheduler, and the OAI-PMH content harvesting method. All authors contributed to the maintenance, operation and improvements of the system. D.H. drafted the initial version of the manuscript based on consultations with P.K. D.P. and P.K. wrote the final manuscript with additional input from L.A. and N.P. D.H., M.C. and L.A. performed the data analysis for the paper and D.H. produced the figures. D.H., D.P., B.G. and L.A. participated in research activities and tasks related to CORE following the instructions and directly supervised by P.K.

Corresponding author

Correspondence to Petr Knoth .

Ethics declarations

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 license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license 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 license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Knoth, P., Herrmannova, D., Cancellieri, M. et al. CORE: A Global Aggregation Service for Open Access Papers. Sci Data 10 , 366 (2023). https://doi.org/10.1038/s41597-023-02208-w

Download citation

Received : 18 May 2021

Accepted : 03 May 2023

Published : 07 June 2023

DOI : https://doi.org/10.1038/s41597-023-02208-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

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.

Our use of cookies

We use necessary cookies to make our site work (for example, to manage your session). We’d also like to use some non-essential cookies (including third-party cookies) to help us improve the site. By clicking ‘Accept recommended settings’ on this banner, you accept our use of optional cookies.

Necessary cookies

Necessary cookies enable core functionality on our website such as security, network management, and accessibility. You may disable these by changing your browser settings, but this may affect how the website functions.

Analytics cookies

We use analytics cookies so we can keep track of the number of visitors to various parts of the site and understand how our website is used. For more information on how these cookies work please see our Cookie policy .

Related links Related links

• Bank Underground - our staff blog
• BoE Research on Twitter/X
• Research datasets

In this section

• Bank of England Agenda for Research
• Researchers
• Research careers
• Staff working papers
• Research funding and fellowships
• Research Visitor Programme

Our research agenda

Our research activity is centred around the Bank of England Agenda for Research (BEAR) . This multi-year document is organised around five themes and includes a set of priority topics that we update once a year.

Sign up to receive email notifications

Our researchers.

Our researchers carry out diverse and cutting-edge research in business areas across the Bank, and the Research Hub centrally coordinates research activities on BEAR. Find out more about our Research careers .

Researchers listed

Our research events

Recent events listed

Our Staff Working papers

We share our work externally to encourage comments and debate. Our new research is released in our ‘Staff Working papers’ series.

Staff Working papers listed

Our external publications

External publications in refereed journals 2015-2023.

• Dating the lender of last resort (Michael Anson, David Bholat, Kilian Rieder and Ryland Thomas), The Economic Journal, Vol 133
• Tracking foreign capital: the effect of capital inflows on bank lending in the UK (Christiane Kneer and Alexander Raabe), IMF Economic Review 
• Quality is our asset: the international transmission of liquidity regulation (Dennis Reinhardt, Stephen Reynolds, Rhiannon Sowerbutts and Carlos van Hombeeck), Journal of Banking and Finance, Vol. 154
• Global capital flows and the role of macroprudential policy (Sudipto Karmakar and Diogo Lima), Journal of Financial Stability, Vol. 67
• House price dynamics, optimal LTV limits and the liquidity trap (Andrea Ferrero, Richard Harrison and Benjamin Nelson), The Review of Economic Studies, Vol. 90
• The more the merrier? Evidence on the value of multiple requirements in bank regulation (Marcus Buckmann, Paula Gallego Marquez, Mariana Gimpelewicz, Sujit Kapadia and Katie Rismanchi), Journal of Banking and Finance, Vol. 149
• Global banks and synthetic funding: the benefits of foreign relatives (Fernando Eguren-Martin, Matias Ossandon Busch and Dennis Reinhardt), Journal of Money, Credit and Banking
• Dating business cycles in the United Kingdom, 1700-2010 (Stephen Broadberry, Jagjit Chadha, Jason Lennard and Ryland Thomas), The Economic History Review, Vol. 76
• Sectoral shocks and monetary policy in the United Kingdom (Huw Dixon, Jeremy Franklin and Stephen Millard), Oxford Bulletin of Economics and Statistics, Vol. 85
• Targeting financial stability: macroprudential or monetary policy? (David Aikman, Julia Giese, Sujit Kapadia and Michael McLeay), International Journal of Central Banking, Vol. 19
• Updated estimates of UK GDP from the income side, 1841-1920 (Solomos Solomou and Ryland Thomas), Economic History Review, Vol. 76
• The impact of non-tariff barriers on trade and welfare (Swati Dhingra, Rebecca Freeman and Hanwei Huang), Economica, Vol. 90
• Improving variance forecasts: the role of realized variance features (Ioannis Papantonis, Leonidas Rompolis and Elias Tzavalis), International Journal of Forecasting, Vol. 39
• Monetary policy inertia and the paradox of flexibility (Dario Bonciani and Joonseok Oh), Journal of Economic Dynamics and Control, Vol. 151
• The cyclicality of bank credit losses and capital ratios under expected loss model (Mahmoud Fatouh and Simone Giansante), Annals of Operations Research, Vol. 330
• Divided committees and strategic vagueness (Nicole Baerg and Colin Krainin), European Journal of Political Economy, Vol. 74
• AI model risk: what the current model risk management framework can teach us about managing the risks of AI models (Catarina Souza), Journal of Financial Compliance, Vol. 6
• The pandemic’s promising lessons for the future of green public banking (Victoria Stadheim, Serdar Sengul and Thomas Marois), UCL Institute for Innovation and Public Purpose, Policy Brief 18
• Clients' connections: measuring the role of private information in decentralized markets (Peter Kondor and Gabor Pinter), The Journal of Finance, Vol. 77
• Systemic financial stress and macroeconomic amplifications in the United Kingdom (Somnath Chatterjee, Jeremy Chiu, Thibaut Duprey and Sinem Hacioglu-Hoke), Oxford Bulletin of Economics and Statistic, Vol. 84
• Impact of IFRS 9 on the cost of funding of banks in Europe (Mahmoud Fatouh, Robert Bock and Jamal Ouenniche), Journal of Banking Regulation, Vol. 24
• Shareholder risk-taking incentives in the presence of contingent capital (Mahmoud Fatouh and Ayowande A. McCunn), Journal of Financial Regulation and Compliance, Vol. 30
• The transmission of monetary policy shocks (Silvia Miranda-Agrippino and Giovanni Ricco), American Economic Journal: Macroeconomics, Vol. 13
• To ask or not to ask? Bank capital requirements and loan collateralization (Hans Degryse, Artashes Karapetyan and Sudipto Karmakar), Journal of Financial Economics, Vol. 142
• The physics of financial networks (Marco Bardoscia, Paolo Barucca, Stefano Battiston, Fabio Caccioli, Giulio Cimini, Diego Garlaschelli, Fabio Saracco, Tiziano Squartini and Guido Caldarelli), Nature Reviews – Physics, Vol. 3
• Central bank swap lines: evidence on the effects of the lender of last resort (Saleem Bahaj and Ricardo Reis), The Review of Economic Studies, Vol. 89 
• The assessment: artificial intelligence and financial services (David Bholat and Daniel Susskind), Oxford Review of Economic Policy, Vol. 37
• Is the UK’s productivity puzzle mostly driven by occupational mismatch? An analysis using big data on job vacancies (Arthur Turrell, Bradley Speigner, David Copple, Jyldyz Djumalieva and James Thurgood), Labour Economic, Vol. 71
• Adaptive learning and labor market dynamics (Federico Di Pace, Kaushik Mitra and Shoujian Zhang), Journal of Money, Credit and Banking,  Vol. 53
• Comparing minds and machines: implications for financial stability (Marcus Buckmann, Andy Haldane and Anne-Caroline Hüser), Oxford Review of Economic Policy, Vol. 37
• Central bank digital currencies – design principles for financial stability (Michael Kumhof and Clare Noone), Economic Analysis and Policy, Vol. 71
• Credit default swaps and corporate bond trading (Robert Czech), Journal of Financial Intermediation, Vol. 48
• Currency mispricing and dealer balance sheets (Gino Cenedese, Pasquale Della Corte and Tianyu Wang), The Journal of Finance, Vol. 76
• Banks, money, and the zero lower bound on deposit rates (Michael Kumhof and Xuan Wang), Journal of Economic Dynamics and Control, Vol. 132
• Unconventional monetary policies and the macroeconomy: the impact of the UK’s QE2 and funding for lending scheme (Rohan Churm, Michael Joyce, George Kapetanios and Konstantinos Theodoridis), The Quarterly Review of Economics and Finance, Vol. 80
• Capital requirements and mortgage pricing: evidence from Basel II (Matteo Benetton, Peter Eckley, Nicola Garbarino, Liam Kirwin and Georgia Latsi), Journal of Financial Intermediation, Vol. 48 
• Tail risk interdependence (Ching-Wai (Jeremy) Chiu, Arnold Polanski and Evarist Stoja), International Journal of Finance and Economics, Vol. 26
• The missing link: monetary policy and the labor share (Cristiano Cantore, Filippo Ferroni and Miguel León-Ledesma), Journal of the European Economic Association, Vol. 19
• The nexus between CEO incentives and analysts' earnings forecasts (Anna Bagntasarian and Emmanuel Mamatzakis), International Journal of Finance and Economics, Vol. 26
• The interaction between macroprudential policy and monetary policy: overview   (Matthieu Bussière, Jin Cao, Jakob de Haan, Robert Hills, Simon Lloyd, Baptiste Meunier, Justine Pedrono, Dennis Reinhardt, Sonalika Sinha, Rhiannon Sowerbutts and Konstantin Styrin), Review of International Economics, Special Issue Paper, Vol. 29
• Le Pont de Londres: interactions between monetary and prudential policies in cross‐border lending (Matthieu Bussière, Robert Hills, Simon Lloyd, Baptiste Meunier, Justine Pedrono, Dennis Reinhardt and Rhiannon Sowerbutts), Review of International Economics, Special Issue Paper, Vol. 29
• Shocks and labour cost adjustment: evidence from a survey of European firms (Thomas Y Mathä, Stephen Millard, Tairi Rõõm, Ladislav Wintr and Robert Wyszyński), Oxford Economic Papers, Vol. 73
• Regulatory effects on short-term interest rates (Angelo Ranaldo, Patrick Schaffner and Michalis Vasios), Journal of Financial Economics, Vol. 141
• UK house prices and three decades of decline in the risk-free real interest rate (David Miles and Victoria Monro), Economic Policy, Vol. 36 
• Inferring trade directions in fast markets (Simon Jurkatis), Journal of Financial Markets, Vol. 58
• Transparent modeling of influenza incidence: big data or a single data point from psychological theory? (Konstantinos V. Katsikopoulos, Özgür Şimşek, Marcus Buckmann and Gerd Gigerenzer), International Journal of Forecasting, Vol. 38
• Economic integration and unit labour costs (Sophie Piton), European Economic Review, Vol. 136
• High water, no marks? Biased lending after extreme weather (Nicola Garbarino and Benjamin Guin), Journal of Financial Stability, Vol. 54
• Optimal simple objectives for monetary policy when banks matter (Lien Laureys, Roland Meeks and Boromeus Wanengkirtyo), European Economic Review, Vol. 135
• The distributional impact of the pandemic (Sinem Hacıoğlu Hoke, Diego R. Känzig and Paolo Surico), European Economic Review, Vol. 134
• History dependence in the housing market (Philippe Bracke and Silvana Tenreyro), American Economic Journal: Macroeconomics, Vol. 13
• Solvency and wholesale funding cost interactions at UK banks   (Kieran Dent, Sinem Hacıoğlu Hoke and Apostolos Panagiotopoulos), Journal of Financial Stability, Vol. 52.
• Common asset holdings and systemic vulnerability across multiple types of financial institution   (Paolo Barucca, Tahir Mahmood and Laura Silvestri), Journal of Financial Stability, Vol. 52
• Macroprudential FX regulations: shifting the snowbanks of FX vulnerability? (Toni Ahnert, Kristin Forbes, Christian Friedrich and Dennis Reinhard), Journal of Financial Economics, Vol. 140
• Ambiguity, monetary policy, and trend inflation (Riccardo M. Masolo and Francesca Monti), Journal of the European Economic Association, Vol. 19
• Overnight indexed swap-implied interest rate expectations (Simon P. Lloyd), Finance Research Letters, Vol. 38
• Foreign banks and trade (Stijn Claessens and Neeltje van Horen), Journal of Financial Intermediation, Vol. 45
• The cost of human capital depreciation during unemployment (Lien Laureys), The Economic Journal, Vol. 131
• Common correlated effect cross‐sectional dependence corrections for non‐linear conditional mean panel models (Sinem Hacıoğlu Hoke and George Kapetanios), Journal of Applied Econometrics, Vol. 36
• Gilt auctions and secondary market dynamics (Lucas Marc Fuhrer and Julia Giese), Finance Research Letters, Vol. 38
• The impact of quantitative easing on UK bank lending: why banks do not lend to businesses? (Mahmoud Fatouh, Sheri Markose and Simone Giansante), Journal of Economic Behavior & Organization, Vol. 183
• Counterparty choice in the UK credit default swap market: an empirical matching approach (Gerardo Ferrara, Jun Sung Kim, Bonsoo Koo and Zijun Liu), Economic Modelling, Vol. 94
• Risk shocks and monetary policy in the new normal (Martin Seneca), International Journal of Central Banking, Vol. 16
• The leverage ratio and liquidity in the gilt and gilt repo markets   (Andreea Bicu-Lieb, Louisa Chen and David Elliott), Journal of Financial Markets, Vol. 48
• International evidence on shock-dependent exchange rate pass-through (Kristin Forbes, Ida Hjortsoe and Tsvetelina Nenova), IMF Economic Review, Vol. 68
• Intangible capital, stock markets and investments: implications for macroeconomic stability (Eddie Gerba), Journal of Accounting and Finance, Vol. 20
• Exchange rate passthrough at the micro and macro levels in a small open economy: evidence from several million unit values (John Lewis), Canadian Journal of Economics, Vol. 53
• Stress tests and the countercyclical capital buffer: The UK experience (Donald Kohn), The Manchester School, Vol. 88, Special issue: Proceedings of the Money, Macroeconomics and Finance Research Group, 2019 
• Economic uncertainty before and during the COVID-19 pandemic (Dave Altig, Scott Baker, Jose Maria Barrero, Nicholas Bloom, Philip Bunn, Scarlet Chen, Steven J. Davis, Julia Leather, Brent Meyer, Emil Mihaylov, Paul Mizen, Nicholas Parker, Thomas Renault, Pawel Smietanka and Gregory Thwaites), Journal of Public Economics, Vol. 191 
• Revisiting the global decline of the (non-housing) labor share (Germán Gutiérrez and Sophie Piton), American Economic Review: Insights, Vol. 2
• Time-varying cointegration with an application to the UK great ratios (George Kapetanios, Stephen Millard, Katerina Petrova and Simon Price), Economics Letters, Vol. 193
• The forced safety effect: how higher capital requirements can increase bank lending (Saleem Bahaj and Frederic Malherbe), The Journal of Finance, Vol. 75
• Identification of structural vector autoregressions by stochastic volatility (Dominik Bertsche and Robin Braun), Journal of Business and Economic Statistics
• COVID-19 and the financial system: a tale of two crises (Julia Giese and Andy Haldane), Oxford Review of Economic Policy, Vol. 36
• Estimating nominal interest rate expectations: overnight indexed swaps and the term structure (Simon P. Lloyd), Journal of Banking and Finance, Vol. 119
• Cross-border effects of regulatory spillovers: evidence from Mexico (Jagdish Tripathy), Journal of International Economics, Vol. 126
• Nonlinear effects of mortgage spreads over the business cycle   (Chak Hung Jack Cheng and Ching‐Wai (Jeremy) Chiu), Journal of Money, Credit and Banking, Vol. 52
• Uncertainty and economic activity: a multicountry perspective   (Ambrogio Cesa-Bianchi, M Hashem Pesaran and Alessandro Rebucci), The Review of Financial Studies, Vol. 33
• The evolution of competition in the UK deposit-taking sector, 1989–2013   (Sebastian de-Ramon and Michael Straughan), The European Journal of Finance, Vol. 26
• The international effects of global financial uncertainty shocks (Dario Bonciani and Martino Ricci), Journal of International Money and Finance, Vol. 109
• Home values and firm behavior (Saleem Bahaj, Angus Foulis and Gabor Pinter), American Economic Review, Vol. 110
• Risk-neutral densities: advanced methods of estimating nonnormal options underlying asset prices and returns (André Santos and João Guerra), Journal of Risk Model Validation, Vol. 14
• Network valuation in financial systems (Paolo Barucca, Marco Bardoscia, Fabio Caccioli, Marco D’Errico, Gabriele Visentin, Guido Caldarelli and Stefano Battiston), Mathematical Finance, Vol. 30
• Corporate decision making in the presence of political uncertainty: the case of corporate cash holdings   (William B. Hankins, Anna‐Leigh Stone, Chak Hung Jack Cheng and Ching‐Wai (Jeremy) Chiu), The Financial Review, Vol. 55
• The global financial cycle after Lehman (Silvia Miranda-Agrippino and Hélène Rey), AEA Papers and Proceedings, Vol. 110
• Firms’ price, cost and activity expectations: evidence from micro data (Lena Boneva, James Cloyne, Martin Weale and Tomasz Wieladek), The Economic Journal, Vol. 130
• Sovereign spreads in the euro area: cross border transmission and macroeconomic implication   (Saleem Bahaj), Journal of Monetary Economics, Vol. 110
• Official demand for US debt: implications for US real rates (Iryna Kaminska and Gabriele Zinna), Journal of Money, Credit and Banking, Vol. 52
• A welfare‐based analysis of international monetary policy spillovers at the zero lower bound  (Alex Haberis and Anna Lipińska), Journal of Money, Credit and Banking, Vol. 52
• Centralized trading, transparency, and interest rate swap market liquidity: evidence from the implementation of the Dodd-Frank Act   (Evangelos Benos, Richard Payne and Michalis Vasios), Journal of Financial and Quantitative Analysis, Vol. 55
• Monetary policy transmission in the United Kingdom: a high frequency identification approach (Ambrogio Cesa-Bianchi, Gregory Thwaites and Alejandro Vicondoa), European Economic Review, Vol. 123
• Monetary policy when households have debt: new evidence on the transmission mechanism   (James Cloyne, Clodomiro Ferreira and Paolo Surico), Review of Economic Studies, Vol. 87 
• Estimating the elasticity of intertemporal substitution using mortgage notches  (Michael Carlos Best, James Cloyne, Ethan Ilzetzki and Henrik Kleven), The Review of Economic Studies, Vol. 87
• The macroeconomic impact of shocks to bank capital buffers in the euro area   (Derrick Kanngiesser, Reiner Martin, Laurent Maurin and Diego Moccero), The B.E. Journal of Macroeconomics, Vol. 20
• The impact of the 1932 General Tariff: a difference-in-difference approach   (Simon P. Lloyd and Solomos Solomou), Cliometrica, Vol. 14
• Financial stability considerations and monetary policy (Anil Kashyap and Caspar Siegert), International Journal of Central Banking, Vol. 16
• Mapping bank securities across euro area sectors: comparing funding and exposure networks (Anne-Caroline Huser and Christoffer Kok), Journal of Network Theory in Finance, Vol. 5
• Inefficiencies in trade reporting for over-the-counter derivatives: is blockchain the solution? (Olatunji Jayeola), Capital Markets Law Journal, Vol. 15 

(Carlos Eduardo van Hombeeck), Journal of International Money and Finance, Vol. 101

• Deposit withdrawals from distressed banks: client relationships matter  (Martin Brown, Benjamin Guin and Stefan Morkoetter), Journal of Financial Stability, Vol. 46
• Three triggers? Negative equity, income shocks and institutions as determinants of mortgage default (Andrew Linn and Ronan C. Lyons), The Journal of Real Estate Finance and Economics, Vol. 61
• Central counterparty auction design (Gerardo Ferrara, Xin Li and Daniel Marszalec), Journal of Financial Market Infrastructures, Vol. 8
• Text mining letters from financial regulators to firms they supervise (David Bholat and James Brookes), Digital Scholarship in the Humanities, Vol. 35
• Identifying noise shocks: a VAR with data revisions (Riccardo M. Masolo and Alessia Paccagnini), Journal of Money, Credit and Banking, Vol. 51
• Enhancing central bank communications using simple and relatable information (David Bholat, Nida Broughton, Janna Ter Meer and Eryk Walczak), Journal of Monetary Economics, Vol. 108
• The long-run information effect of central bank communication (Stephen Hansen, Michael McMahon and Matthew Tong), Journal of Monetary Economics, Vol. 108
• Investor behaviour and reaching for yield: evidence from the sterling corporate bond market (Robert Czech and Matt Roberts‐Sklar), Financial Markets, Institutions and Instruments, Vol. 28
• Multiplex network analysis of the UK over‐the‐counter derivatives market (Marco Bardoscia, Ginestra Bianconi and Gerardo Ferrara), International Journal of Finance and Economics, Vol. 24
• Forward-looking solvency contagion (Marco Bardoscia, Paolo Barucca, Adam Brinley Codd and John Hill), Journal of Economic Dynamics and Control, Vol. 108
• Forecasting the UK economy with a medium-scale Bayesian VAR (Sílvia Domit, Francesca Monti and Andrej Sokol), International Journal of Forecasting, Vol. 35
• A new approach for detecting shifts in forecast accuracy (Ching-Wai (Jeremy) Chiu, Simon Hayes, George Kapetanios and Konstantinos Theodoridis), International Journal of Forecasting, Vol. 35
• The invisible hand of the government: moral suasion during the European sovereign debt crisis (Steven Ongena, Alexander Popov and Neeltje Van Horen), American Economic Journal: Macroeconomics, Vol. 11
• The determinants of credit union failure: insights from the United Kingdom (Jamie Coen, William B. Francis and May Rostom), International Journal of Central Banking, Vol. 15
• Lost in diversification (Marco Bardoscia, Daniele d'Arienzo, Matteo Marsili and Valerio Volpati), Comptes Rendus Physique, Vol. 20
• The effect of house prices on household borrowing: a new approach (James Cloyne, Kilian Huber, Ethan Ilzetzki and Henrik Kleven), American Economic Review, Vol. 109
• Enterprise vs. product logic: the industrial reorganisation corporation and the rationalisation of the British electrical/electronics industry  (Anthony Gandy and Roy Edwards), Business History, Vol. 61 
• A time-varying parameter structural model of the UK economy (George Kapetanios, Riccardo M. Masolo, Katerina Petrova and Matthew Waldron), Journal of Economic Dynamics and Control, Vol. 106
• Unstable diffusion indexes: with an application to bond risk premia (Daniele Massacci), Oxford Bulletin of Economics and Statistics, Vol. 81
• The Stochastic Lower Bound (Riccardo M. Masolo and Pablo E. Winant), Economics Letters, Vol. 180
• Some principles for regulating cyber risk (Anil K. Kashyap and Anne Wetherilt), AEA Papers and Proceedings, Vol. 109
• Editorial: International spillovers of monetary policy through global banks: introduction to the special issue (Claudia Buch, Matthieu Bussière, Menzie Chinn, Linda Goldberg and Robert Hills), Journal of International Money and Finance, Vol. 91
• The international transmission of monetary policy (Claudia M. Buch, Matthieu Bussière, Linda Goldberg and Robert Hills), Journal of International Money and Finance, Vol. 91
• The international transmission of monetary policy through financial centres: evidence from the United Kingdom and Hong Kong (Robert Hills, Kelvin Ho, Dennis Reinhardt, Rhiannon Sowerbutts, Eric Wong and Gabriel Wu), Journal of International Money and Finance, Vol. 90
• Drawing on different disciplines: macroeconomic agent-based models   (Andrew Haldane and Arthur Turrell), Journal of Evolutionary Economics, Vol. 29
• Deflation probability and the scope for monetary loosening in the United Kingdom (Alex Haberis, Riccardo M. Masolo and Kate Reinold), International Journal Of Central Banking, Vol. 15
• Fiscal rules and structural reforms (Rana Sajedi and Armin Steinbach), International Review of Law and Economics, Vol. 58
• Would macroprudential regulation have prevented the last crisis?< (David Aikman, Jonathan Bridges, Anil Kashyap and Caspar Siegert), Journal of Economic Perspectives, Vol. 33
• An elusive panacea? The impact of the regulatory valuation regime on insurers’ investment behaviour (Caterina Lepore, Misa Tanaka, David Humphry and Kallol Sen), International Review of Financial Analysis
• Short-time work in the Great Recession: firm-level evidence from 20 EU countries (Reamonn Lydon, Thomas Y. Mathä and Stephen Millard), IZA Journal of Labor Policy, Vol. 8
• Uncertain policy promises (Alex Haberis, Richard Harrison and Matt Waldron), European Economic Review, Vol. 111
• Predictive regressions under asymmetric loss: factor augmentation and model selection (Matei Demetrescua and Sinem Hacıoğlu Hoke), International Journal of Forecasting, Vol. 35
• Forecasting the UK economy: alternative forecasting methodologies and the role of off-model information (Lena Boneva, Nicholas Fawcett, Riccardo M. Masolo and Matt Waldron), International Journal of Forecasting, Vol. 35
• Finance and synchronization (Ambrogio Cesa-Bianchi, Jean Imbs and Jumana Saleheen), Journal of International Economics, Vol. 116
• Notes on the Underground: monetary policy in resource‐rich economies (Andrea Ferrero and Martin Seneca), Journal of Money, Credit and Banking, Vol. 51
• Credit traps and macroprudential leverage (Angus Foulis, Benjamin Nelson and Misa Tanaka), Journal of Money, Credit and Banking, Vol. 51
• Output gaps, inflation and financial cycles in the UK   (Marko Melolinna and Máté Tóth), Empirical Economics, Vol. 56
• Labor market frictions, monetary policy and durable goods   (Federico Di Pace and Matthias S. Hertweck), Review of Economic Dynamics, Vol. 32
• Foreign booms, domestic busts: the global dimension of banking crises  (Ambrogio Cesa-Bianchi, Fernando Eguren Martin and Gregory Thwaites), Journal of Financial Intermediation, Vol. 37
• Brexit and uncertainty: insights from the Decision Maker Panel (Nicholas Bloom, Philip Bunn, Scarlet Chen, Paul Mizen, Pawel Smietanka, Greg Thwaites and Garry Young), Fiscal Studies, Vol. 39
• Partisan conflict, policy uncertainty and aggregate corporate cash holdings (Chak Hung Jack Chenga, Ching-Wai (Jeremy) Chiu, William B. Hankins and Anna-Leigh Stone), Journal of Macroeconomics, Vol. 58
• Financial market volatility, macroeconomic fundamentals and investor sentiment (Ching-Wai (Jeremy) Chiu, Richard D.F. Harris, Evarist Stoja and Michael Chin), Journal of Banking & Finance, Vol. 92
• Leverage and risk-weighted capital requirements (Leonardo Gambacorta and Sudipto Karmakar), International Journal of Central Banking, Vol. 14
• Macroeconomic modelling at the Institute: hopes, challenges and a lasting contribution (Ray Barrell, Andrew Blake and Garry Young), National Institute Economic Review,  Issue 246
• Bank taxes, leverage, and risk (Kristoffer Milonas), Journal of Financial Services Research, Vol. 54
• The shocks matter: improving our estimates of exchange rate pass-through (Kristin Forbes, Ida Hjortsoe and Tsvetelina Nenova), Journal of International Economics, Vol. 114
• Macroeconomic uncertainty in South Africa (Christopher Redl), South African Journal of Economics, Vol. 86 
• Capital requirements, monetary policy and risk shifting in the mortgage market (Arzu Uluc and Tomasz Wieladek), Journal of Financial Intermediation, Vol. 35, Part B  
• Monetary and macroprudential policies under rules and discretion   (Lien Laureys and Roland Meeks), Economics Letters, Vol. 170
• Fiscal consequences of structural reform under constrained monetary policy (Rana Sajedi), Journal of Economic Dynamics and Control, Vol. 93
• Public debt and fiscal policy traps (Antoine Camous and Andrew Gimber), Journal of Economic Dynamics and Control, Vol. 93
• The time value of housing: historical evidence on discount rates (Philippe Bracke, Edward W. Pinchbeck and James Wyatt), The Economic Journal, Vol. 128
• Tailwinds from the East: how has the rising share of imports from emerging markets affected import prices? (John Lewis and Jumana Saleheen), Cambridge Journal of Economics, Vol. 42
• Funding constraints and liquidity in two-tiered OTC markets (Evangelos Benos and Filip Žikeš), Journal of Financial Markets, Vol. 39
• Measuring systemic risk in the European banking sector: a copula CoVaR approach (Emmanouil Karimalis and Nikos Nomikos), The European Journal of Finance, Vol. 24
• Climate change challenges for central banks and financial regulators (Emanuele Campiglio, Yannis Dafermos, Pierre Monnin, Josh Ryan-Collins, Guido Schotten and Misa Tanaka), Nature Climate Change, Vol. 8
• The consumption response to positive and negative income shocks (Philip Bunn, Jeanne Le Roux, Kate Reinold and Paolo Surico), Journal of Monetary Economics, Vol. 96
• Stabilising house prices: the role of housing futures trading (Arzu Uluc), The Journal of Real Estate Finance and Economics, Vol. 56, Issue 4
• International credit supply shocks (Ambrogio Cesa-Bianchi, Andrea Ferrero and Alessandro Rebucci), Journal of International Economics, Vol. 112
• Uncertainty, financial frictions, and nominal rigidities: a quantitative investigation (Ambrogio Cesa‐Bianchi and Emilio Fernandez‐Corugedo), Journal of Money, Credit and Banking, Vol. 50, Issue 4
• How does financial liberalisation affect the influence of monetary policy on the current account? (Ida Hjortsoe, Martin Weale and Tomasz Wieladek), Journal of International Money and Finance, Vol. 85
• The missing links: a global study on uncovering financial network structures from partial data   (Kartik Anand, Iman van Lelyveld, Ádám Banai, Soeren Friedrich, Rodney Garratt, Grzegorz Hałaj, Jose Fique, Ib Hansen, Serafín Martínez Jaramillo, Hwayun Lee, José Luis Molina-Borboa, Stefano Nobili, Sriram Rajan, Dilyara Salakhova, Thiago Christiano Silva, Laura Silvestri, Sergio Rubens Stancato de Souza), Journal of Financial Stability, Vol. 35
• On the theory of international currency portfolios (Michael Kumhof), European Economic Review, Vol. 101
• The impact of de-tiering in the United Kingdom’s large-value payment system (Evangelos Benos, Gerardo Ferarra and Pedro Gurrola-Perez), Journal of Financial Market Infrastructures, Vol. 6, No. 2/3
• The validation of filtered historical VaR models (Pedro Gurrola-Perez), Journal of Risk Model Validation, Vol. 12, No. 1
• Forecasting with VAR models: fat tails and stochastic volatility (Ching-Wai (Jeremy) Chiu, Haroon Mumtaz and Gabor Pinter), International Journal of Forecasting, Vol. 33, Issue 4
• Non-performing loans at the dawn of IFRS 9: regulatory and accounting treatment of asset quality (David Bholat, Rosa M. Lastra, Sheri M. Markose, Andrea Miglionico and Kallol Sen), Journal of Banking Regulation, Vol. 19, Issue 1
• Threshold-based forward guidance (Lena Boneva, Richard Harrison and Matt Waldron), Journal of Economic Dynamics and Control, Volume 90
• Optimal pay regulation for too-big-to-fail banks (John Thanassoulis and Misa Tanaka), Journal of Financial Intermediation, Vol. 33
• The impact of EU and non-EU immigration on British wages (Stephen Nickell and Jumana Saleheen), IZA Journal of Development and Migration
• Network reconstruction with UK CDS trade repository data (William Abel and Laura Silvestri), Quantitative Finance, Vol. 17, Issue 12
• When arm’s length is too far: relationship banking over the credit cycle (Thorsten Beck, Hans Degryse, Ralph De Haas and Neeltje van Horen), Journal of Financial Economics, Vol. 127, Issue 1
• An interdisciplinary model for macroeconomics (Andrew Haldane and Arthur Turrell), Oxford Review of Economic Policy, Vol. 34, Issue 1-2
• An agent-based model of corporate bond trading (Karen Braun-Munzinger, Zijun Liu and Arthur Turrell), Quantitative Finance, Vol. 18, Issue 4
• What do the prices of UK inflation-linked securities say on inflation expectations, risk premia and liquidity risks? (Iryna Kaminska, Zhuoshi Liu, Jon Relleen and Elisabetta Vangelista), Journal of Banking & Finance, Vol.  88
• Volatility in equity markets and monetary policy rate uncertainty (Iryna Kaminska and Matt Roberts-Sklar), Journal of Empirical Finance, Vol. 45
• Resolution of financial distress under agency frictions (Santiago Moreno-Bromberg and Quynh-Anh Vo), Journal of Banking and Finance, Vol. 82
• Initial margin model sensitivity analysis and volatility estimation (Melanie Houllier and David Murphy) Journal of Financial Market Infrastructures, Vol. 5, No. 4
• Interactions among High-Frequency Traders (Evangelos Benos, James Brugler, Erik Hjalmarsson and Filip Zikes) Journal of Financial and Quantitative Analysis, Vol. 52, Issue 4
• Statistical mechanics of complex economies (Marco Bardoscia, Giacomo Livan, Matteo Marsili), Journal of Statistical Mechanics: Theory and Experiment
• Managing market liquidity risk in central counterparties (Evangelos Benos, Pedro Gurrola-Perez and Michael Wood), Journal of Financial Market Infrastructures, Vol. 5, No. 4
• Macroprudential policy under uncertainty (Saleem Bahaj and Angus Foulis), International Journal of Central Banking, Vol. 13(3)
• Central Counterparties (CCPs) and the law of default management (Jo Braithwaite and David Murphy), Journal of Corporate Studies. Published online: 5 Jan 2017
• Household debt and the dynamic effects of income tax changes (James Cloyne and Paolo Surico), Review of Economic Studies, Vol. 84, Issue 1
• On a tight leash:  Does bank organizational structure matter for macroprudential spillovers? (Danisewicz, P, Reinhardt, D and Sowerbutts, R), Journal of International Economics, Vol. 109
• A non-cooperative foundation for the continuous Raiffa solution (Bram Driesen, Peter Eccles and Nora Wegner), International Journal of Game Theory, Vol. 46, Issue 4
• Resolving a multi-million dollar contract dispute with a Latin Square (William B Fairley, Peter J Kempthorne, Julie Novak, Scott McGarvie, Steve Crunk, Bee Leng Lee and Alan J Salzberg), The American Statistician, Vol. 71, Issue 3
• Assessing vulnerabilities to financial shocks in some key global economies (Jack Fisher and Lukasz Rachel), Journal of Risk Management in Financial Institutions, Vol. 10, No. 1
• Current account deficits during heightened risk: menacing or mitigating? (Kristin Forbes, Ida Hjortsoe and Tsvetelina Nenova), The Economic Journal, Vol. 127, Issue 601
• The spillovers, interactions, and (un)intended consequences of monetary and regulatory policies (Kristin Forbes, Dennis Reinhardt and Tomasz Wieladek), Journal of Monetary Economics, Vol. 85
• International banking and cross-border effects of regulation: lessons from the Netherlands (Jon Frost, Jakob de Haan and Neeltje van Horen), International Journal of Central Banking, Vol. 13(1)
• International banking and cross-border effects of regulation: lessons from the United Kingdom (Robert Hills, Dennis Reinhardt, Rhiannon Sowerbutts and Tomasz Wieladek), International Journal of Central Banking, Vol. 13, Supplement 1
• Least squares estimation of large dimensional threshold factor models (Daniele Massacci), Journal of Econometrics, Vol. 197, Issue 1
• Tail risk dynamics in stock returns: links to the macroeconomy and global markets connectedness (Daniele Massacci), Management Science, Vol 63, Issue 9
• I’ve got you under my skin: large central counterparty financial resources and the incentives they create (David Murphy), Journal of Financial Market Infrastructures, Vol. 5, No. 3
• Estimating ‘hedge and auction’ liquidation costs in central counterparties: a closeout risk approach (Luis Vicente, Fernando Cerezetti and Alan De Genaro), Journal of Financial Market Infrastructures, Vol. 6, No. 1
• The Bank of England/NMG Survey of household finances (G Anderson, P Bunn, A Pugh and A Uluc) Fiscal Studies, Vol. 37, Issue 1
• The role of collateral in supporting liquidity (Y Baranova, Z Liu and J Noss) Journal of Financial Market Infrastructures, Vol. 5, No. 1
• Price discovery and the cross-section of high-frequency trading (E Benos and S Sagade) Journal of Financial Markets, Vol. 30
• What moves international stock and bond markets? (G Cenedese and E Mallucci) Journal of International Money and Finance, Vol. 60
• What do stock markets tell us about exchange rates? (G Cenedese, R Payne, L Sarno and G Valente) Review of Finance, Vol. 20, Issue 3
• Does US partisan conflict matter for the euro area? (C H J Cheng, J Chiu and W B Hawkins) Economics Letters, Vol. 138
• The macroeconomic effects of monetary policy:  a new measure for the United Kingdom (J Cloyne and P Hürtgen) American Economic Journal: Macroeconomics, Vol. 8, No. 4
• CCPs in crisis: International Commodities Clearing House, New Zealand Futures and Options Exchange (R T Cox, D Murphy and E Budding) Journal of Financial Market Infrastructures, Vol. 4, No 3
• Factor complementarity and labour market dynamics (F Di Pace and S Villa) European Economic Review, Vol. 82
• Exchange rate regimes and current account adjustment:  an empirical investigation (F Eguren Martin) Journal of International Money and Finance, Vol. 65
• Using Shapley’s asymmetric power index to measure banks’ contributions to systemic risk (R Garratt, L Webber and M Willison) Journal of Network Theory in Finance, Vol. 2, No. 2
• Imbalances and fiscal policy in a monetary union (I Hjortsoe) Journal of International Economics, Vol. 102
• A new summary measure of inflation expectations (G Kapetanios, B Maule and G Young) Economics Letters, Vol. 149
• Evaluating the robustness of UK term structure decompositions using linear regression methods (S Malik and A Meldrum)  Journal of Banking and Finance, Vol. 67
• Integrating macro-prudential policy:  central banks as the ‘third force’ in EU financial reform (S McPhilemy) West European Politics, Vol. 39, No. 3
• The effect of labor and financial frictions on aggregate fluctuations (H Mumtaz and F Zanetti) Macroeconomic Dynamics, Vol. 20, Issue 1
• I want security: stylized facts about CCP collateral and their systemic context (D Murphy, H Holden and M Houllier) Journal of Financial Market Infrastructures, Vol. 5, No. 2
• Identifying historical episodes for CCP stress testing (D Murphy and D Macdonald) Journal of Financial Market Infrastructures, Vol. 4, No 3
• Capital stocks and capital services:  integrated and consistent estimates for the United Kingdom, 1950-2013 (N Oulton and G Wallis) Economic Modelling, Vol. 54
• An intellectual property-based approach to the mandatory disclosure among lenders of credit data for small and medium enterprises (P Siciliani) Journal of Banking Regulation, Vol. 17, Issue 4
• Efficient evaluation of collisional energy transfer terms for plasma particle simulations (A E Turrell, M Sherlock and S J Rose) Journal of Plasma Physics, Vol. 82, Issue 1
• Tax incentives and investment in the UK (G Wallis) Oxford Economic Papers, Vol. 68, Issue 2
• The two faces of cross-border banking flows (Dennis Reinhardt and Steven J Riddiough), IMF Economic Review, Vol. 63, Issue 4
• Curbing the credit cycle (D Aikman, A G Haldane and B D Nelson) The Economic Journal, Vol. 125, Issue 585
• Reputation, risk-taking and macroprudential policy (D Aikman, B Nelson and M Tanaka) Journal of Banking and Finance, Vol. 50
• Simple banking: profitability and the yield curve (P Alessandri and B Nelson) Journal of Money, Credit and Banking, Vol. 47, Issue 1
• Risky bank lending and countercyclical capital buffers (J Benes and M Kumhof) Journal of Economic Dynamics and Control, Vol. 58
• Big data and central banks (D Bholat) Big Data and Society, April
• Inflation targeting and term premia estimates for Latin America (A Blake, G Rule and O Rummel) Latin American Economic Review, Vol. 24, Issue 1
• House Prices and Rents: Microevidence form a Matched Data Set in Central London (P Bracke) Real Estate Economics, Vol. 43, Issue 2
• Bayesian mixed frequency VARs (J Chiu, B Eraker, A T Foerster, T B Kim and H D Seoane) Journal of Financial Econometrics, Vol. 13, Issue 3
• An empirical sectoral model of unconventional monetary policy: the impact of QE (J Cloyne, R Thomas, A Tuckett and S Wills) The Manchester School, Vol. 83, Supplement S1
• FDI, debt and capital controls (S Dell'Erba and D Reinhardt) Journal of International Money and Finance, Vol. 58
• DSGE priors for BVAR models (T Filippeli and K Theodoridis) Empirical Economics, Vol. 48, Issue 2
• Estimating the effects of forward guidance in rational expectations models (R Harrison) European Economic Review, Vol. 79
• Generalised density forecast combinations (G Kapetanios, J Mitchell, S Price and N Fawcett) Journal of Econometrics, Vol. 188, Issue 1
• A new approach to multi-step forecasting using dynamic stochastic general equilibrium models (G Kapetanios, S Price and K Theodoridis) Economics Letters, Vol. 136
• Factor adjustment costs:  a structural investigation (H Mumtaz and F Zanetti) Journal of Economic Dynamics and Control, Vol. 51
• Optimal tolerance for failure (C Siegert and P Trepper) Journal of Economic Behavior and Organization, Vol. 109
• Monetary policy and the financial sector (A Singh, S Stone and J Suda) Economics Letters, Vol. 132
• Ultrafast collisional ion heating by electrostatic shocks (A E Turrell, M Sherlock and S J Rose) Nature Communications, Vol. 6

Other External Publications 2015-2023

• Price discrimination and mortgage choice (Jamie Coen, Anil Kashyap and May Rostom), CEPR Discussion Paper No. 18478
• Price discrimination and mortgage choice (Jamie Coen, Anil Kashyap and May Rostom), NBER Working Paper No. 31652
• The ‘F words’: why surveying businesses about intangibles is so hard (Josh Martin and Cain Baybutt), Economic Statistics Centre of Excellence, Discussion Paper No. 2022-20
• Rights, duties, best interests? Critically assessing the emerging jurisprudence of financial consumer protection (Colin Shields), Financial Regulation International
• Sectoral shocks and monetary policy in the United Kingdom (Huw Dixon, Jeremy Franklin and Steven Millard), Cardiff Economics Working Paper No. E2021/10
• The impact of COVID-19 on artificial intelligence in banking (Julia Anderson, David Bholat, Mohammed Gharbawi and Oliver Thew), Bruegel
• Central bank swap lines during the Covid-19 pandemic (Saleem Bahaj and Ricardo Reis), Covid Economics, Centre for Economic Policy Research
• Monetary policy surprises and their transmission through term premia and expected interest rates (Iryna Kaminska, Haroon Mumtaz and Roman Sustek), Centre for Macroeconomics Discussion Paper, CFM-DP2020-24
• Supervision reforms in the frame (Pau Ling Yap and Paolo Cadoni), The Actuary, June
• Climate change: macroeconomic impact and implications for monetary policy (Sandra Batten, Rhiannon Sowerbutts and Misa Tanaka), Chapter 2 in Walker, T, Gramlich, D, Bitar, M and Fardnia, P (eds), Ecological, Societal, and Technological Risks and the Financial Sector, Palgrave Studies in Sustainable Business In Association with Future Earth, Palgrave Macmillan
• UK Business and Financial Cycles Since 1660 - Volume I: A Narrative Overview (Nicholas Dimsdale and Ryland Thomas) Palgrave Macmillan
• Monetary policy for commodity booms and busts (Thomas Drechsel, Michael McLeay and Silvana Tenreyro), Jackson Hole Economic Policy Symposium - Challenges for Monetary Policy
• In the face of spillovers: prudential policies in emerging economies (Andra Coman and Simon Lloyd), ECB Working Paper Series No. 2339
• Exchange rate risk and business cycles (Simon Lloyd and Emile Marin), Cambridge Working Papers in Economics CWPE1996
• Mortgage cash-flows and employment (Fergus Cumming), Centre for Macroeconomics Discussion Paper, CFM-DP2019-22
• The Brexit vote, productivity growth and macroeconomic adjustments in the United Kingdom (Ben Broadbent, Federico Di Pace, Thomas Drechsel, Richard Harrison and Silvana Tenreyro), Centre for Macroeconomics Discussion Paper, CFM-DP2019-16
• Optimal inflation and the identification of the Phillips curve (Michael McLeay and Silvana Tenreyro), chapter in Eichenbaum, M S, Hurst, E and Parker, J A (eds), NBER Macroeconomics Annual 2019, Vol. 34
• Revisiting the global decline of the (non-housing) labor share (Germán Gutiérrez and Sophie Piton), Centre for Macroeconomics Discussion Paper, CFM-DP2019-13
• Four years of concurrent stress testing at the Bank of England: developing the macroprudential perspective (Rohan Churm and Paul Nahai-Williamson), Chapter 6 in Siddique, A, Hasan, I and Lynch, D (eds), Stress Testing (2nd Edition), Approaches, Methods and Applications, Risk Books
• Do unit labour costs matter? A decomposition exercise on European data (Sophie Piton), Centre for Macroeconomics Discussion Paper, CFM-DP2019-10
• Shapley regressions: A framework for statistical inference on machine learning models (Andreas Joseph), King’s Business School, DAFM Working Paper Series No. 2019/7
• Estimating market-implied value with jump-diffusion models (Somnath Chatterjee and Ken Deeley), MathWorks Technical Articles and Newsletters 
• The BoC-BoE Sovereign Default Database revisited: what’s new in 2018? (David Beers and Jamshid Mavalwalla), Bank of Canada Staff Working Paper 2018-30
• The missing link: monetary policy and the labor share (Cristiano Cantore, Filippo Ferroni and Miguel A. Leon-Ledesma), Centre for Macroeconomics Discussion Paper, CFM-DP2018-29
• History dependence in the housing market (Philippe Bracke and Silvana Tenreyro), CEP Discussion Paper No. 1568
• Macroprudential policy in a globalised world (Dennis Reinhardt and Rhiannon Sowerbutts), Chapter 8 in Mizen, P, Rubio, M and Turner, P (eds), Macroprudential Policy and Practice, Cambridge University Press 
• Monetary policy, corporate finance and investment (James Cloyne, Clodomiro Ferreira, Maren Froemel and Paolo Surico), NBER Working Paper No. 25366
• House price dynamics, optimal LTV limits and the liquidity trap (Andrea Ferrero, Richard Harrison and Benjamin Nelson), CEPR Discussion Paper DP13400
• Bank risk-taking and misconduct (Ieva Sakalauskaite), Bank of Lithuania Working Paper No. 53
• Macroprudential FX regulations: shifting the snowbanks of FX vulnerability? (Toni Ahnert, Kristin Forbes, Christian Friedrich and Dennis Reinhardt), NBER Working Paper No. 25083
• To ask or not to ask? Bank capital requirements and loan collateralization (Hans Degryse, Artashes Karapetyan and Sudipto Karmakar), CEPR Discussion Paper DP13331
• Climate change: what implications for central banks and financial regulators? (Sandra Batten, Rhiannon Sowerbutts and Misa Tanaka), Chapter in Caldecott, B (ed), Stranded Assets and the Environment: Risk, Resilience and Opportunity, Routledge Explorations in Environmental Studies
• Central bank accounting (David Bholat and Robin Darbyshire), Chapter 16 in Conti-Brown, P and Lastra, R M (eds), Research Handbook on Central Banking, Edward Elgar
• Macroprudential FX regulations: shifting the snowbanks of FX vulnerability? (Toni Ahnert, Kristin Forbes, Christian Friedrich and Dennis Reinhardt), MIT Sloan School Working Paper 5293-18
• Solvency II in the UK - evolution rather than revolution (David Humphry), Chapter in Marano, P and Siri, M (eds), Insurance regulation in the European Union - Solvency II and beyond, Palgrave Macmillan
• Business cycles in an oil economy (Drago Bergholt, Vegard H Larsen and Martin Seneca), BIS Working Papers No. 618
• Peer-to-peer lending and financial innovation in the UK (D Bholat and U Atz), Chapter in Aldohni, A K (ed), Law and finance after the financial crisis: the untold stories of the UK financial market, Routledge
• Identifying ‘Uber Risk’: central banks’ analytics aid stability (David Bholat and Chiranjit Chakraborty), Global Public Investor, June
• Threshold-based forward guidance: hedging the zero bound (Lena Boneva, Richard Harrison and Matt Waldron), CEPR Discussion Paper No. 11749
• The consumption response to positive and negative income changes (Philip Bunn, Jeanne LeRoux, Kate Reinold and Paolo Surico), CEPR Discussion Paper No. 11829
• Culture and household saving (Benjamin Guin), ECB Working Paper Series No. 2069
• Labour market adjustment in Europe during the crisis: microeconomic evidence from the Wage Dynamics Network survey (Mario Izquierdo, Juan Francisco Jimeno, Theodora Kosma, Ana Lamo, Stephen Millard, Tairi Rõõm, Eliana Viviano), ECB Occasional Paper Series No. 192
• The leverage ratio, risk-taking and bank stability (Jonathan Acosta Smith, Michael Grill, Jan Hannes Lang), ECB Working Paper Series No. 2079
• UK broad money growth and nominal spending during the Great Recession: an analysis of the money creation process and the role of money demand (R Thomas), Chapter in Congdon, T (ed), Money in the Great Recession:  Did a Crash in Money Growth Cause the Global Slump?, Buckingham Studies in Money, Banking and Central Banking, Edward Elgar
• Capital requirements, risk shifting and the mortgage market (Arzu Uluc and Tomasz Wieladek), ECB Working Paper Series No. 2061
• Monetary versus macroprudential policies: causal impacts of interest rates and credit controls in the era of the UK Radcliffe Report (D Aikman, O Bush and A M Taylor), NBER Working Paper No. 22380; CEPR Discussion Paper No. 11353;  LSE Economic History Working Paper No. 246
• The residential collateral channel (S Bahaj, A Foulis and G Pinter), Centre for Macroeconomics Discussion Paper No. CFM-DP2016-07
• Modelling metadata in central banks (D Bholat) ECB Statistics Paper No. 13
• The analysis of money and credit during the financial crisis:  the approach at the Bank of England (J Bridges, J Cloyne, R Thomas and A Tuckett), Chapter in Cobham, D (ed), Monetary analysis at central banks, Palgrave Macmillan
• Household debt and spending in the United Kingdom (P Bunn and M Rostom), Chapter in Chadha, J, Crystal, A, Pearlman, J, Smith, P and Wright, S (eds), The UK economy in the long expansion and its aftermath, Cambridge University Press
• VAR models with non-Gaussian shocks (J Chiu, H Mumtaz and G Pinter), Centre for Macroeconomics Discussion Paper No. CFM-DP2016-09
• The real effects of capital requirements and monetary policy: evidence from the United Kingdom (F De Marco and T Wieladek), CEPR Discussion Paper No. 11265
• QE:  the story so far (A Haldane, M Roberts-Sklar, T Wieladek and C Young), CEPR Discussion Paper No. 11691
• Monetary policy and the current account:  theory and evidence (I Hjortsoe, M Weale and T Wieladek), CEPR Discussion Paper No. 11204
• UK broad money growth in the long expansion, 1992–2007:  what can it tell us about the role of money? (M McLeay and R Thomas), Chapter in Chadha, J, Crystal, A, Pearlman, J, Smith, P and Wright, S (eds), The UK economy in the long expansion and its aftermath, Cambridge University Press
• Capital requirements, risk shifting and the mortgage market (A Uluc and T Wieladek), CEPR Discussion Paper No. 11214
• Market discipline, public disclosure and financial stability (P Zimmerman and R Sowerbutts), Chapter in The Handbook of post crisis financial modelling, Palgrave Macmillan
• Interest rates, debt and intertemporal allocations:  evidence from notched mortgage contracts in the UK (M Best, J Cloyne, E Ilzetzki and H Kleven), July 2015 NBER conference paper 
• Monetary policy when households have debt (J Cloyne, C Ferreira and P Surico), July 2015 NBER conference paper
• Monetary policy when households have debt:  new evidence on the transmission mechanism (J Cloyne, C Ferreira and P Surico), CEPR Discussion Paper No. 11023
• Notes on the Underground:  monetary policy in resource-rich economies (A Ferrero and M Seneca), OxCarre Research Paper Series No. 158
• Deposit protection and bank resolution (N Kleftouri), Oxford University Press
• Sovereign risk and financial crisis (S Pepino), Palgrave Macmillan

Research publications and events

Growth-at-risk for macroprudential policy..., growth-at-risk for macroprudential policy stance assessment: a survey, dominant currency pricing transition, lash risk and interest rates, econdat 2024 spring meeting: economics with..., econdat 2024 spring meeting: economics with nontraditional data and analytical tools.

Back to top

Explore millions of high-quality primary sources and images from around the world, including artworks, maps, photographs, and more.

Explore migration issues through a variety of media types

• Part of The Streets are Talking: Public Forms of Creative Expression from Around the World
• Part of The Journal of Economic Perspectives, Vol. 34, No. 1 (Winter 2020)
• Part of Cato Institute (Aug. 3, 2021)
• Part of University of California Press
• Part of Open: Smithsonian National Museum of African American History & Culture
• Part of Indiana Journal of Global Legal Studies, Vol. 19, No. 1 (Winter 2012)
• Part of R Street Institute (Nov. 1, 2020)
• Part of Leuven University Press
• Part of UN Secretary-General Papers: Ban Ki-moon (2007-2016)
• Part of Perspectives on Terrorism, Vol. 12, No. 4 (August 2018)
• Part of Leveraging Lives: Serbia and Illegal Tunisian Migration to Europe, Carnegie Endowment for International Peace (Mar. 1, 2023)
• Part of UCL Press

Harness the power of visual materials—explore more than 3 million images now on JSTOR.

Enhance your scholarly research with underground newspapers, magazines, and journals.

Explore collections in the arts, sciences, and literature from the world’s leading museums, archives, and scholars.

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Download 55 million PDFs for free

Explore our top research interests.

Engineering

Anthropology

• Earth Sciences

• Computer Science

• Mathematics

• Health Sciences

Join 261 million academics and researchers

Track your impact.

Share your work with other academics, grow your audience and track your impact on your field with our robust analytics

Discover new research

Get access to millions of research papers and stay informed with the important topics around the world

Publish your work

Publish your research with fast and rigorous service through Academia.edu Publishing. Get instant worldwide dissemination of your work

Unlock the most powerful tools with Academia Premium

Work faster and smarter with advanced research discovery tools

Search the full text and citations of our millions of papers. Download groups of related papers to jumpstart your research. Save time with detailed summaries and search alerts.

• Advanced Search
• PDF Packages of 37 papers
• Summaries and Search Alerts

Share your work, track your impact, and grow your audience

Get notified when other academics mention you or cite your papers. Track your impact with in-depth analytics and network with members of your field.

• Mentions and Citations Tracking
• Advanced Analytics
• Publishing Tools

Real stories from real people

Used by academics at over 15,000 universities

Get started and find the best quality research

• Academia.edu Publishing
•   We're Hiring!
•   Help Center
• Find new research papers in:
• Cognitive Science
• Academia ©2024

Cookies on GOV.UK

We use some essential cookies to make this website work.

We’d like to set additional cookies to understand how you use GOV.UK, remember your settings and improve government services.

We also use cookies set by other sites to help us deliver content from their services.

You have accepted additional cookies. You can change your cookie settings at any time.

You have rejected additional cookies. You can change your cookie settings at any time.

• Society and culture
• Sports and leisure
• 2012 Olympic and Paralympic legacy

The research papers and reports covering our sectors are independent. Please note that, while we have consulted on them, we do not necessarily…

The research papers and reports covering our sectors are independent. Please note that, while we have consulted on them, we do not necessarily endorse any findings or recommendations contained within them.

Many of our public bodies also produce a range of research on our sectors. Links to their websites can be found on our research links page .

02/07/2012 Cost-Benefit Analysis of Radio Switchover: Methodology Report

12/08/2011 International comparisons of public engagement in culture and sport

20/06/2011 DCMS Longitudinal Data Library

24/05/2011 Digital Radio Switchover: Willingness to Pay and Consumer Behaviour Research

24/05/2011 Intertek - Research Study on Energy Consumption of Digital Radios: Phase 2

15/12/2010 Measuring the value of culture: a report to the Department for Culture Media and Sport

05/11/2010 Understanding the relationship between taste and value in culture and sport

Cultural and Sporting Evidence (CASE) Research

22/03/2010 Libraries Omnibus

09/11/2009 Digital Britain: Attitudes to supporting non-BBC regional news from the TV licence fee

06/10/2009 London 2012 olympic and paralympic games impacts and legacy evaluation framework

16/06/2009 To accompany the Final Digital Britain Report, the Department is also published - ‘Digital Britain: Attitudes towards internet content among adults’.

02/01/2009 Commissioned report ‘Capturing the Impact of Libraries’ by BOP Consulting has been published.

23/06/2008 Research report by National Children’s Bureau, Play and Exercise in Early Years:Physically active play in early childhood provision

26/02/2008 Scoping study for a UK Gambling Act 2005, impact assessment framework

17/12/2007 Survey of Live Music in England and Wales in 2007

23/10/2007 DCMS commissioned BMG Research to study Residential care and nursing homes: readiness for digital switchover

26/09/2007 The feasibility of a live music economic impact study

01/08/2007 An Assessment of Productivity Indicators for the Creative Industries

24/07/2007 Evaluation of the Cultural Pathfinder programme

06/07/2007 “Culture on demand”  report, proposing practical ways to engage the broadest possible audience for culture by building on existing demand

16/05/2007 Framework for evaluating Cultural Policy Investment  

10/01/2007 BMRB on Digital switchover - readiness of social housing

07/12/2006 Licensing Act 2003: the experience of smaller establishments in applying for live music authorisation

For a full archive of previous research see the publications page in The National Archive .

DCMS has licensed Qualtrics online survey software enabling us to undertake sophisticated online surveys and in-house consultations.

Is this page useful?

• Yes this page is useful
• No this page is not useful

Help us improve GOV.UK

Don’t include personal or financial information like your National Insurance number or credit card details.

To help us improve GOV.UK, we’d like to know more about your visit today. Please fill in this survey (opens in a new tab) .

• Schools & departments

NMR study of the structure and dynamics of the BRCT domain from the kinetochore protein KKT4

Akiyoshi Lab - Springer

Ludzia, P., Hayashi, H., Robinson, T., Akiyoshi, B., and Redfield, C.

Summary of Paper by Eleanor Casey, Marston Lab

The kinetochore is a large protein complex which physically links the spindle microtubule to the chromosome during mitosis. Mutations in kinetochore proteins prevent accurate segregation of the chromosomes. This causes growth defects in lower eukaryotes, and diseases in higher organisms. Due to their essentiality in ensuring proper genetic inheritance, kinetochores and kinetochore proteins are highly conserved among most eukaryotes. However, an evolutionarily divergent group of organisms called kinetoplastids have a compositionally and structurally unique kinetochore which bears very little similarity to most eukaryotes. The kinetoplastid group includes the parasite Trypanosoma brucei which causes sleeping sickness in infected individuals. A better understanding of the unique kinetochore structure of T. brucei parasites would provide novel therapeutic targets to treat sleeping sickness.

In this paper, Ludzia and colleagues use nuclear magnetic resonance (NMR) spectroscopy to understand the structure the T. brucei kinetochore protein KKT4. The researchers had previously characterized the structure of the KKT4 C terminus by X ray crystallography, although the functional significance of this structure remained unknown. NMR allowed the structure of KKT4 to be observed in solution, and showed that the protein was able to bind phosphate ions through 3 key amino acid residues in a sulphate ion binding site. When these 3 amino acids were mutated in live T. brucei cells to disrupt phosphate ion binding, there was a severe growth defect. This demonstrated the importance of KKT4 binding to phosphorylated proteins in maintaining the T. brucei kinetochore structure, and highlights this binding pocket as a potential target for drug discovery.

The green IT revolution: A blueprint for CIOs to combat climate change

Companies and governments looking to combat climate change are turning to tech for help. AI, new technologies, and some promising tech-driven business models have raised hopes for dramatic progress.

About the authors

This article is a collaborative effort by Gerrit Becker, Luca Bennici, Anamika Bhargava, Andrea Del Miglio , Jeffrey Lewis , and Pankaj Sachdeva, representing views from McKinsey Technology.

While many organizations’ climate goals are lofty, enterprise technology leaders—CIOs, chief digital innovation officers (CDIOs), and chief technology officers (CTOs), among others—have not always succeeded at turning climate ambitions into reality. One of the biggest reasons is that hard facts and clear paths of action are scarce. Misconceptions and misinformation have clouded the picture of what CIOs and tech leaders should do.

We have done extensive analysis of where technology can have the biggest impact on reducing emissions. To start, we divided technology’s role into two primary types of activities:

• offense—the use of technology and analytics to cut emissions by reducing (improving operational efficiency), replacing (shifting emission-generating activities to cleaner alternatives), and reusing (recycling material)
• defense—the actions IT can take to reduce emissions from the enterprise’s technology estate

Scope of the McKinsey analysis

McKinsey’s emissions analysis for this report focuses on enterprise technology emissions, which are the business IT emissions from the hardware, software, IT services, enterprise communications equipment, mobile devices, fixed and mobile network services, and internal technology teams that a company uses for its own operations and that a CIO has control over. These include the emissions related to the full life cycles of the products and services that an enterprise IT function uses, including their development, delivery, usage, and end of life (exhibit). Our internal services emissions' analysis assumes around 40 percent of IT workers are working from home.

The analysis does not include the emissions from the technology products and services that a company is selling (such as data center capacity sold by hyperscalers), operational technology devices (such as sensors and point-of-sale systems), and cryptocurrency mining.

The defense activities are where the CIO, as the head of IT, can act independently and quickly. This article focuses on defense, specifically the IT elements over which a CIO has direct control. We examined emissions from use of electricity for owned enterprise IT operations, such as the running of on-premises data centers and devices (classified as scope 2 by the Greenhouse Gas Protocol 1 Greenhouse Gas Protocol: Technical Guidance for Calculating Scope 3 Emissions: Supplement to the Corporate Value Chain (Scope 3) Accounting & Reporting Standard , World Resources Institute & World Business Council for Sustainable Development, 2013. Scope 1 emissions are direct emissions from the activities of an organization or under their control, including fuel combustion on site such as gas boilers, fleet vehicles, and air-conditioning leaks; scope 2 emissions are from electricity purchased and used by the organization; and scope 3 emissions are all indirect emissions not included in scope 2 that occur in the value chain of the reporting company, including both upstream and downstream emissions. ), and indirect emissions from technology devices that the CIO buys and disposes of (scope 3). 2 These calculations do not include emissions from technology-driven services sold, such as cloud capacity. (See sidebar, “Scope of the McKinsey analysis.”)

What the facts say

Our analysis has uncovered several facts that contravene some commonly held views about enterprise technology emissions. These facts involve the significant amount of tech-related emissions, the share of emissions from end-user devices, the variety of mitigation options available, and the favorable impact of shifting to cloud computing.

Enterprise technology generates significant emissions

Enterprise technology is responsible for emitting about 350 to 400 megatons of carbon dioxide equivalent gases (CO 2 e), accounting for about 1 percent of total global greenhouse gas (GHG) emissions. At first blush, this might not seem like a lot, but it equals about half of the emissions from aviation or shipping and is the equivalent of the total carbon emitted by the United Kingdom.

The industry sector that contributes the largest share of technology-related scope 2 and scope 3 GHG emissions is communications, media, and services (Exhibit 1). Enterprise technology’s contribution to total emissions is especially high for insurance (45 percent of total scope 2 emissions) and for banking and investment services (36 percent).

This amount of carbon dioxide and equivalent gases is a significant prize for companies under increasing pressure to cut emissions. Progress on climate change requires action on many fronts, and enterprise technology offers an important option that CIOs and companies can act on quickly.

You’re invited

To a McKinsey Technology webinar on the critical role of technology in building a sustainable enterprise on October 25, 9:30–10:30am ET.

The biggest carbon culprit is end-user devices, not on-premises data centers

End-user devices—laptops, tablets, smartphones, and printers—generate 1.5 to 2.0 times more carbon globally than data centers (Exhibit 2). 3 On-premises and co-located data centers used by enterprises, not including data center capacity sold by hyperscalers. One reason is that companies have significantly more end-user devices than servers in on-premises data centers. In addition, the devices typically are replaced much more often: smartphones have an average refresh cycle of two years, laptops four years, and printers five years. On average, servers are replaced every five years, though 19 percent of organizations wait longer. 4 Rhona Ascierto and Andy Lawrence, Uptime Institute global data center survey 2020 , Uptime Institute, July 2020.

More worrisome, emissions from end-user devices are on track to increase at a CAGR of 12.8 percent per year. 5 End-user computing market: Growth, trends, COVID-19 impact, and forecasts (2022–2027) , Mordor Intelligence, January 2022. Efforts to address this could target the major causes of emissions from these devices. About three-fourths of the emissions comes from manufacturing, upstream transportation, and disposal. A significant source of these emissions is the semiconductors that power the devices.

Plenty of low-cost/high-impact options exist, starting with improved sourcing

We have found that when it comes to going green, many CIOs think in terms of investments needed to replace items or upgrade facilities. Our analysis, however, finds that CIOs can capture significant carbon benefits without making a significant investment—and in some cases can even save money (Exhibit 3).

Overall, for example, 50 to 60 percent of emissions related to end-user devices can be addressed through sourcing changes, primarily by procuring fewer devices per person and extending the life cycle of each device through recycling. These options will not require any investment and will lower costs, though companies may want to evaluate the impact on employee experience.

In addition, companies can more aggressively recycle their devices; 89 percent of organizations recycle less than 10 percent of their hardware overall. 6 Sustainable IT: Why it’s time for a green revolution for your organization’s IT , Capgemini Research Institute, 2021. CIOs can put pressure on suppliers to use greener devices, especially as companies in the semiconductor sector are already increasing their commitments to emission reduction. Further low-cost, high-impact actions include optimizing business travel and data center computing needs, as well as increasing the use of cloud to manage workloads.

Moving to cloud has more impact than optimizing data centers

Optimizing an on-premises data center’s power usage effectiveness (PUE) 7 PUE describes how efficiently a computer data center uses energy, expressed as the ratio of total facility energy to IT equipment energy. is expensive and results in limited carbon abatement. If a company were to double what it spends on infrastructure and cloud to reduce PUE, it would cut carbon emissions by only 15 to 20 percent. Structural improvements in data centers and optimized layout can help, but the impact is limited, and many companies have already implemented them. More aggressive measures, such as moving data centers to cooler locations or investing in new cooling tech, are prohibitively expensive.

A more effective approach is to migrate workloads to the cloud. Hyperscalers (also known as cloud service providers) and co-locators are investing significantly to become greener through measures such as buying green energy themselves and investing in ultra-efficient data centers with a PUE equal to or less than 1.10, compared with the average PUE of 1.57 for an on-premises data center. 8 “Uptime Institute 11th annual Global Data Center Survey shows sustainability, outage, and efficiency challenges amid capacity growth,” Uptime Institute, September 14, 2021. (We estimate that companies could achieve just a 1.3 PUE score for their data center if they invested nearly 250 percent more, on average, over what they currently spend for their data centers and cloud presence.)

With thoughtful migration to and optimized usage of the cloud, companies could reduce the carbon emissions from their data centers by more than 55 percent—about 40 megatons of CO 2 e worldwide, the equivalent of the total carbon emissions from Switzerland.

Three steps to take now

With companies and governments under intensifying pressure to cut carbon emissions and with technology playing a key role in delivering on those goals, CIOs will find themselves on the front lines. The challenge will be to reduce IT’s carbon footprint while delivering high-quality, low-cost technology services to customers and employees.

On average, completion of the defensive steps might take three to four years. However, CIOs who act decisively and precisely can achieve 15 to 20 percent of carbon reduction potential in the first year with minimal investment.

CIOs can choose from among a wide array responses, particularly in conjunction with the CEO and the board. However, three measures they can take right now will prepare the organization for longer-term efforts. These measures involve sourcing strategies, key metrics, and a performance management system.

The net-zero transition: What it would cost, what it could bring

Move now on sourcing strategies.

Far and away the fastest and most effective defensive measure for reducing IT carbon emissions is to revise policies for technology sourcing. Optimizing the number of devices in line with standards followed by companies in the top quartile 9 Top quartile in terms of the ratio of devices to people is derived from the number of devices per person. Our analysis uses McKinsey Digital’s Ignite solutions and 2020 data. would reduce about 30 percent of end-user-device emissions, the amount of carbon emitted by Hong Kong. For example, top-quartile companies have one printer for every 16 people in the workplace; the overall average is one printer per eight people.

This sourcing shift does not necessarily lead to a degradation in user experience, because the rollout of 5G and increasingly advanced processing and compute power allow the main processing function to happen at the server. Therefore, devices can be less powerful and consume much less energy. Essentially, this is a software-as-a-service (SaaS) model where high-end and user-friendly experiences happen on the server, not the device. The effectiveness of this approach will depend on having stable networks, less resource-intensive coding at the device level, edge computing capabilities, and shifts of offerings to more efficient platforms (for example, cloud).

As part of this effort, the CIO and the business’s head of procurement will need to collaborate on reviewing and adjusting device refresh timelines and device-to-person ratios, as well as adjusting the basis for purchasing decisions. Procurement generally relies on cost/benefit calculations, and rightly so. That approach will need to expand to account for carbon dioxide emissions. The spirit of collaboration should extend to suppliers as well, with the parties working together to formulate plans that provide the greatest benefits for all.

A more thoughtful sourcing strategy extends beyond end-user devices. CIOs, for example, should look for green sources of the electricity IT uses. When these sources are unavailable, CIOs can direct procurement to power purchase agreements to offset carbon use. CIOs can also set green standards for their vendors and suppliers, requiring GHG emissions disclosures and incorporating them into their criteria for purchase decisions.

Establish a green ROI metric for technology costs

Any real progress on green technology can happen only when companies measure their “green returns.” But today, most green metrics omit cost and savings, which ultimately makes them impractical. A better metric focuses on cost per ton of carbon saved (accounting for costs saved as well). Sophisticated models calculate emissions throughout the full life cycle, including production, transportation, and disposal.

CIOs can further assess suppliers, manufacturers, and service providers based on how advanced they are in recycling and refurbishing electronics; designing circular components; extending product life cycles with better design, higher-quality manufacturing, and more robust materials; offering repair services; and reselling to consumers.

Decisions about IT spending need to consider a range of factors, including technical debt abatement and business strategy. Along with these factors, companies should institutionalize a green ROI metric that is transparent to everybody in the business as an element in IT decision making, including in requests for proposals (RFPs). Doing so will enable companies to better understand the true impact their technology is having on carbon emissions.

Put in place green measurement systems

Establishing a green ROI metric is only a start. CIOs need to establish a baseline of performance, measure progress against the baseline, and track impact in near real time, much as companies track real-time computer and network usage for applications in the cloud. This kind of measuring system ensures that CIOs know what’s working and what isn’t, so they can adjust quickly.

In practice, implementing green measurement can be challenging. Some companies have spent a year measuring their carbon footprint, ending up with an outdated analysis. This tends to happen when companies are determined to measure every bit of carbon emitted, a praiseworthy but time-consuming effort. CIOs can make substantial progress by instead prioritizing measurement where the impact is highest, such as tracking the number of end-user devices purchased and in use, the current duration of use for each device, and the ratio of devices per user. Another way CIOs can make quick progress is to embed emissions- and power-monitoring capabilities into large technology assets and work with external providers, such as electricity companies, to track usage in real time.

Effectively combating climate change won’t happen through one or two big wins; those don’t exist yet. To have real impact, companies and governments will need to act in many areas. Technology has a huge role to play in many of these areas, but CIOs and tech leaders need to act quickly and decisively.

This article is the first in a series about how CIOs can reduce emissions. The next article will explore how CIOs can drive the business’s sustainability agenda by playing offense and implementing reduce, replace, and reuse levers to decarbonize.

Gerrit Becker is an associate partner in McKinsey’s Frankfurt office, Luca Bennici is an associate partner in the Dubai office, Anamika Bhargava is a consultant in the Toronto office, Andrea Del Miglio is a senior partner in the Milan office, Jeffrey Lewis is a senior partner in the New Jersey office, and Pankaj Sachdeva is a partner in the Philadelphia office.

The authors wish to thank Bernardo Betley, Arjita Bhan, Raghuvar Choppakatla, Sebastian Hoffmann, Abdelrahman Mahfouz, Tom Pütz, Jürgen Sailer, Tim Vroman, Alice Yu, and Gisella Zapata for their contributions to this article.

Explore a career with us

Related articles.

It’s not easy buying green: How to win at sustainable sourcing

Sustainability in semiconductor operations: Toward net-zero production

Delivering the climate technologies needed for net zero

• University of Kentucky

• University of Kentucky ®

UK Libraries Recognizes Two Seniors for Excellence in Undergraduate Scholarship

UK Libraries is proud to recognize seniors Zachary Owen and Megan Quish with the 2024  Dean’s Award for Excellence in Undergraduate Scholarship . The annual prize celebrates exceptional and original scholarship in a traditional paper and a media or digital project, and is awarded to students whose research makes substantive and creative use of UK Libraries’ collections, services, and resources.

Zachary Owen, Paper

Zachary Owen, a senior Public Policy major, was awarded for his paper, “Homeownership and Mortgage Loans: How Important is a Conventional Loan?”

Conducted as his senior capstone project, Owen’s paper explores the relationship between mortgage loans and homeownership. During an internship last summer for the US Senate, Owen was able to attend committee hearings, and was immediately struck by a hearing on predatory mortgage lending. “I knew right away: this is my capstone,” he said.

In the fall, Owen began to explore the data, and was particularly struck by racial disparities in homeownership and mortgage denial rates. “In 2021, there was a 29% disparity between Black and white homeownership,” said Owen. “This is the same disparity as during the 1960s, when banks were actively practicing redlining policies, and worse than in the 1940s. I was astonished to see that not only has there not been progress in this area, but we’ve actually moved backwards.”

Owen found that Black applicants are twice as likely to be denied a mortgage than white applicants. “A lot of this is explained by racial disparities in credit scores,” said Owen. His paper proposes introducing rental history as a new metric in credit scoring, which could boost the credit scores of renters by an average of 40 points. “When renters miss a payment, their landlords typically report this to credit bureaus, but making payments on time isn’t reported. Mandating this reporting could improve the credit scores of renters across the country, and help renters secure mortgages so they can become homeowners.”

Owen relied on UK Libraries resources to review the historical and policy backgrounds of the problem – especially papers from the Journal of Public Administration & Management and the National Bureau of Economic Research – and to use statistics from the Federal Housing Administration and other government agencies to track racial disparities in homeownership and mortgage denial rates over time. 

But it was research skills imparted by Business, Economics, & Government Information Librarian  Jennifer Horne that proved to be the most useful library resource. “Jennifer has come to every class I’ve had in the Martin School,” said Owen. “She taught us about where to start our research, all of the resources available to us, and most importantly, how to determine the credibility of resources by using the ‘CRAAP Test.’ For my project, I read a lot of studies from think tanks. You have to do a lot of horizontal reading – uncovering who founded these organizations, how they’re funded, and what their aims are – to see how objective or how partisan they might be.”

“I saw Jennifer at a Martin School event and we recognized one another,” said Owen. “It’s so helpful to have a librarian that’s personally involved with the program. Her dedication to the program is obvious, and it really makes a difference for students when doing their research. I always knew I could reach out to her.”

Owen worked under the mentorship of Dr. Cory Curl, Lecturer and Director of Undergraduate Studies at the Martin School of Public Policy & Administration, while conducting research for his project. “I’m thrilled that Zach took on one of the nation’s most urgent issues – housing – for his capstone policy analysis project,” said Curl. “As instructor for the fall capstone course, I was thrilled that he drew on his internship experience to design an analysis that will be useful in informing decisions facing policy leaders right now. His use of resources from UK Libraries helped him ground the work in data, evidence, and a solid understanding of the history and context that has shaped the current landscape. We are all very thankful to UK Libraries for providing ongoing, specialized support for UK students as they pursue policy research.”

Owen will return to UK in the fall to begin his Master’s in Public Policy at the Martin School.

Megan Quish, Media Project 

Megan Quish, a senior in the School of Architecture, was awarded for her media project, “A House for Aging and Home for Community.”

Quish’s project envisions a multigenerational living community in Lexington’s Davis Bottom, a historically diverse and working class community that was largely demolished in the early 2000s with the extension of the Newtown Pike. Quish’s project engages closely with the rich history and culture of the area in an attempt to rehabilitate what was lost. “When I first began looking into Davis Bottom, I felt this incredible spark in me,” said Quish. “The more I learned, the more I wanted the research to go on.”

Quish sought out a holistic approach to her project, giving space to ideas that go beyond the typical concerns of architecture. She studied the area’s cultural history, zoning, planning, and transportation policy, and situated her findings within the broader contexts and social currents of the last century, especially the dislocating effects of suburban sprawl on families, marginalized communities, and older generations. “Our urban planning and policy choices mean that older people are increasingly getting left behind,” said Quish. 

Quish also read a large body of secondary literature that considered the historical treatment of aging populations and global approaches to housing. “This is where the library is so amazing,” said Quish. “Every new facet of the problem is another door, and every trail you follow leads somewhere else. The library allows you to follow all of these strands. This literature review was the heart of the project. This is a part that architecture education usually ignores.” 

Following her literature review, Quish explored historical images of Lexington and Davis Bottom through several UK Libraries resources, including  ExploreUK and the  John C. Wyatt Lexington Herald-Leader Photographs . “I would sit down on a Friday night, listen to the radio, and just scroll through these amazing images,” said Quish. “There is so much to discover.”

Quish created a series of 36” x 72” collages that combined historical images with original textures and design elements, blending past, present, and future into a vision of a place-based architectural project that will provide older adults with a high quality of life. Each collage includes a written component that draws on her research.

In many ways, Quish’s project is a demonstration of a way of researching. “After compiling all of this research, I found that the best way to immerse in these ideas and think of them more deeply was to express them artistically. By working with visual aspects of the problem, and especially engaging with historic images, I was able to really process the historical development of this area and try to envision a better future.”

“In our society full of new, shiny things and planned obsolescence, we aren’t influenced to value aging in our elders or environment,” said Quish. “Older generations are stigmatized, isolated, and forgotten. Part of this project is about creating a multi-generational community that integrates our elders back into society so that others can benefit from their knowledge, experience, and wisdom.” 

Over the course of her project, Quish worked closely with Dr. Daniel Vivian, Associate Professor and Director of the Undergraduate Certificate in Historic Preservation in the UK College of Design. “I was thrilled to learn that Megan had received the award,” said Vivian. “It is a testament to the quality of her work, its creativity, and the variety of library resources she used to create it. Megan is a serious and dedicated student who combines creative and artistic skill with strong critical-thinking ability and sound research sensibilities. She is excellent at expressing ideas visually and also has the ability to dig into a subject deeply and think about it in ways that are more typical of the social sciences and humanities. This is a somewhat unusual combination of abilities, and Megan uses her strengths to good effect.”

Quish will graduate in December 2024 and plans to pursue graduate school in a field related to the takeaways of the Davis Bottom project. 

“This became so much more than a research project,” she said. “It has catapulted my interest and started this massive growth of where I want to go, in continuing my education and professionally.”  

As Dean’s Award recipients, Owen and Quish will each receive a $1,000 cash award, and their work will be published through  UKnowledge , UK Libraries’ open access digital collection of scholarship created by University of Kentucky faculty, staff, students, departments, research centers, and administration.

Owen and Quish were honored at the annual Spring Celebration on May 16 alongside this year’s recipients of the  Medallion for Intellectual Achievement , the  Paul A. Willis Award for Outstanding Libraries Faculty , and the  Dean’s Award for Outstanding Staff Performance .

Recent News

• UK Libraries Recognizes Two Seniors for Excellence in Undergraduate Scholarship - May 20, 2024
• Ruth Bryan Receives Paul A. Willis Award for Outstanding Libraries Faculty - May 14, 2024
• Dr. Paul Evans Holbrook, Jr. Honored with UK Libraries Medallion for Intellectual Achievement - May 7, 2024