kuznets hypothesis ppt

Essential Economics Terms: Kuznets Curve

Controversial Trickle-Down Theory of Economic Development

Jason Kerwin/Wikimedia Commons/ CC BY-SA 2.5

• U.S. Economy
• Supply & Demand
• Archaeology
• Ph.D., Business Administration, Richard Ivey School of Business
• M.A., Economics, University of Rochester
• B.A., Economics and Political Science, University of Western Ontario

The Kuznets curve is a hypothetical curve that graphs economic inequality against income per capita over the course of economic development (which was presumed to correlate with time). This curve is meant to illustrate economist Simon Kuznets’ (1901-1985) hypothesis about the behavior and relationship of these two variables as an economy develops from a primarily rural agricultural society to an industrialized urban economy.

Kuznets’ Hypothesis

In the 1950s and 1960s, Simon Kuznets hypothesized that as an economy develops, market forces first increase then decrease the overall economic inequality of the society, which is illustrated by the inverted U-shape of the Kuznets curve. For instance, the hypothesis holds that in the early development of an economy, new investment opportunities increase for those who already have the capital to invest. These new investment opportunities mean that those who already hold the wealth have the opportunity to increase that wealth. Conversely, the influx of inexpensive rural labor to the cities keeps wages down for the working class thus widening the income gap and escalating economic inequality.

The Kuznets curve implies that as a society industrializes, the center of the economy shifts from rural areas to the cities as rural laborers, such as farmers, begin to migrate seeking better-paying jobs. This migration, however, results in a large rural-urban income gap and rural populations decrease as urban populations increase. But according to Kuznets’ hypothesis, that same economic inequality is expected to decrease when a certain level of average income is reached and the processes associated with industrialization, such as democratization and the development of a welfare state, take hold. It is at this point in economic development that society is meant to benefit from trickle-down effect and an increase in per-capita income that effectively decreases economic inequality. 

The inverted U-shape of Kuznets curve illustrates the basic elements of the Kuznets’ hypothesis with income per capita graphed on the horizontal x-axis and economic inequality on the vertical y-axis. The graph shows income inequality following the curve, first increasing before decreasing after hitting a peak as per-capita income increases over the course of economic development.

Kuznets’ curve has not survived without its share of critics. In fact, Kuznets himself emphasized the “fragility of [his] data” among other caveats in his paper. The primary argument of critics of Kuznets’ hypothesis and its resulting graphical representation is based on the countries used in Kuznets’ data set. Critics say that the Kuznets curve does not reflect an average progression of economic development for an individual country, but rather it is a representation of historical differences in economic development and inequality between countries in the dataset. The middle-income countries used in the data set are used as evidence for this claim as Kuznets primarily used countries in Latin America, which have had histories of high levels of economic inequality as compared to their counterparts in terms of similar economic development. The critics hold that when controlling for this variable, the inverted U-shape of the Kuznets curve begins to diminish. Other criticisms have come to light over time as more economists have developed hypotheses with more dimensions and more countries had undergone rapid economic growth that did not necessarily follow Kuznets’ hypothesized pattern.

Today, the environmental Kuznets curve (EKC)—a variation on the Kuznets curve—has become standard in environmental policy and technical literature.

• Visualizing Social Stratification in the U.S.
• Rostow's Stages of Growth Development Model
• The Slope of the Aggregate Demand Curve
• What Is Neoliberalism? Definition and Examples
• What You Need to Know About Economic Inequality
• The Demand Curve Explained
• The Lorenz Curve
• What Is Distributive Justice?
• How Money Supply and Demand Determine Nominal Interest Rates
• Socialism vs. Capitalism: What Is the Difference?
• What is a Cost Function?
• The Critical View on Global Capitalism
• The Beveridge Curve
• Drought Causes, Stages, and Problems
• Economic Geography Overview
• Shifting the Demand Curve

• Search Menu
• Browse content in Arts and Humanities
• Browse content in Archaeology
• Anglo-Saxon and Medieval Archaeology
• Archaeological Methodology and Techniques
• Archaeology by Region
• Archaeology of Religion
• Archaeology of Trade and Exchange
• Biblical Archaeology
• Contemporary and Public Archaeology
• Environmental Archaeology
• Historical Archaeology
• History and Theory of Archaeology
• Industrial Archaeology
• Landscape Archaeology
• Mortuary Archaeology
• Prehistoric Archaeology
• Underwater Archaeology
• Urban Archaeology
• Zooarchaeology
• Browse content in Architecture
• Architectural Structure and Design
• History of Architecture
• Residential and Domestic Buildings
• Theory of Architecture
• Browse content in Art
• Art Subjects and Themes
• History of Art
• Industrial and Commercial Art
• Theory of Art
• Biographical Studies
• Byzantine Studies
• Browse content in Classical Studies
• Classical Literature
• Classical Reception
• Classical History
• Classical Philosophy
• Classical Mythology
• Classical Art and Architecture
• Classical Oratory and Rhetoric
• Greek and Roman Archaeology
• Greek and Roman Epigraphy
• Greek and Roman Law
• Greek and Roman Papyrology
• Late Antiquity
• Religion in the Ancient World
• Digital Humanities
• Browse content in History
• Colonialism and Imperialism
• Diplomatic History
• Environmental History
• Genealogy, Heraldry, Names, and Honours
• Genocide and Ethnic Cleansing
• Historical Geography
• History by Period
• History of Agriculture
• History of Education
• History of Emotions
• History of Gender and Sexuality
• Industrial History
• Intellectual History
• International History
• Labour History
• Legal and Constitutional History
• Local and Family History
• Maritime History
• Military History
• National Liberation and Post-Colonialism
• Oral History
• Political History
• Public History
• Regional and National History
• Revolutions and Rebellions
• Slavery and Abolition of Slavery
• Social and Cultural History
• Theory, Methods, and Historiography
• Urban History
• World History
• Browse content in Language Teaching and Learning
• Language Learning (Specific Skills)
• Language Teaching Theory and Methods
• Browse content in Linguistics
• Applied Linguistics
• Cognitive Linguistics
• Computational Linguistics
• Forensic Linguistics
• Grammar, Syntax and Morphology
• Historical and Diachronic Linguistics
• History of English
• Language Variation
• Language Families
• Language Acquisition
• Language Evolution
• Language Reference
• Lexicography
• Linguistic Theories
• Linguistic Typology
• Linguistic Anthropology
• Phonetics and Phonology
• Psycholinguistics
• Sociolinguistics
• Translation and Interpretation
• Writing Systems
• Browse content in Literature
• Bibliography
• Children's Literature Studies
• Literary Studies (Modernism)
• Literary Studies (Asian)
• Literary Studies (European)
• Literary Studies (Eco-criticism)
• Literary Studies (Romanticism)
• Literary Studies (American)
• Literary Studies - World
• Literary Studies (1500 to 1800)
• Literary Studies (19th Century)
• Literary Studies (20th Century onwards)
• Literary Studies (African American Literature)
• Literary Studies (British and Irish)
• Literary Studies (Early and Medieval)
• Literary Studies (Fiction, Novelists, and Prose Writers)
• Literary Studies (Gender Studies)
• Literary Studies (Graphic Novels)
• Literary Studies (History of the Book)
• Literary Studies (Plays and Playwrights)
• Literary Studies (Poetry and Poets)
• Literary Studies (Postcolonial Literature)
• Literary Studies (Queer Studies)
• Literary Studies (Science Fiction)
• Literary Studies (Travel Literature)
• Literary Studies (War Literature)
• Literary Studies (Women's Writing)
• Literary Theory and Cultural Studies
• Mythology and Folklore
• Shakespeare Studies and Criticism
• Browse content in Media Studies
• Browse content in Music
• Applied Music
• Dance and Music
• Ethics in Music
• Ethnomusicology
• Gender and Sexuality in Music
• Medicine and Music
• Music Cultures
• Music and Culture
• Music and Religion
• Music and Media
• Music Education and Pedagogy
• Music Theory and Analysis
• Musical Scores, Lyrics, and Libretti
• Musical Structures, Styles, and Techniques
• Musicology and Music History
• Performance Practice and Studies
• Race and Ethnicity in Music
• Sound Studies
• Browse content in Performing Arts
• Browse content in Philosophy
• Aesthetics and Philosophy of Art
• Epistemology
• Feminist Philosophy
• History of Western Philosophy
• Metaphysics
• Moral Philosophy
• Non-Western Philosophy
• Philosophy of Action
• Philosophy of Law
• Philosophy of Religion
• Philosophy of Science
• Philosophy of Language
• Philosophy of Mind
• Philosophy of Perception
• Philosophy of Mathematics and Logic
• Practical Ethics
• Social and Political Philosophy
• Browse content in Religion
• Biblical Studies
• Christianity
• East Asian Religions
• History of Religion
• Judaism and Jewish Studies
• Qumran Studies
• Religion and Education
• Religion and Health
• Religion and Politics
• Religion and Science
• Religion and Law
• Religion and Art, Literature, and Music
• Religious Studies
• Browse content in Society and Culture
• Cookery, Food, and Drink
• Cultural Studies
• Customs and Traditions
• Ethical Issues and Debates
• Hobbies, Games, Arts and Crafts
• Lifestyle, Home, and Garden
• Natural world, Country Life, and Pets
• Popular Beliefs and Controversial Knowledge
• Sports and Outdoor Recreation
• Technology and Society
• Travel and Holiday
• Visual Culture
• Browse content in Law
• Arbitration
• Browse content in Company and Commercial Law
• Commercial Law
• Company Law
• Browse content in Comparative Law
• Systems of Law
• Competition Law
• Browse content in Constitutional and Administrative Law
• Government Powers
• Judicial Review
• Local Government Law
• Military and Defence Law
• Parliamentary and Legislative Practice
• Construction Law
• Contract Law
• Browse content in Criminal Law
• Criminal Procedure
• Criminal Evidence Law
• Sentencing and Punishment
• Employment and Labour Law
• Environment and Energy Law
• Browse content in Financial Law
• Banking Law
• Insolvency Law
• History of Law
• Human Rights and Immigration
• Intellectual Property Law
• Browse content in International Law
• Private International Law and Conflict of Laws
• Public International Law
• IT and Communications Law
• Jurisprudence and Philosophy of Law
• Law and Society
• Law and Politics
• Browse content in Legal System and Practice
• Courts and Procedure
• Legal Skills and Practice
• Primary Sources of Law
• Regulation of Legal Profession
• Medical and Healthcare Law
• Browse content in Policing
• Criminal Investigation and Detection
• Police and Security Services
• Police Procedure and Law
• Police Regional Planning
• Browse content in Property Law
• Personal Property Law
• Study and Revision
• Terrorism and National Security Law
• Browse content in Trusts Law
• Wills and Probate or Succession
• Browse content in Medicine and Health
• Browse content in Allied Health Professions
• Arts Therapies
• Clinical Science
• Dietetics and Nutrition
• Occupational Therapy
• Operating Department Practice
• Physiotherapy
• Radiography
• Speech and Language Therapy
• Browse content in Anaesthetics
• General Anaesthesia
• Neuroanaesthesia
• Browse content in Clinical Medicine
• Acute Medicine
• Cardiovascular Medicine
• Clinical Genetics
• Clinical Pharmacology and Therapeutics
• Dermatology
• Endocrinology and Diabetes
• Gastroenterology
• Genito-urinary Medicine
• Geriatric Medicine
• Infectious Diseases
• Medical Oncology
• Medical Toxicology
• Pain Medicine
• Palliative Medicine
• Rehabilitation Medicine
• Respiratory Medicine and Pulmonology
• Rheumatology
• Sleep Medicine
• Sports and Exercise Medicine
• Clinical Neuroscience
• Community Medical Services
• Critical Care
• Emergency Medicine
• Forensic Medicine
• Haematology
• History of Medicine
• Medical Ethics
• Browse content in Medical Dentistry
• Oral and Maxillofacial Surgery
• Paediatric Dentistry
• Restorative Dentistry and Orthodontics
• Surgical Dentistry
• Browse content in Medical Skills
• Clinical Skills
• Communication Skills
• Nursing Skills
• Surgical Skills
• Medical Statistics and Methodology
• Browse content in Neurology
• Clinical Neurophysiology
• Neuropathology
• Nursing Studies
• Browse content in Obstetrics and Gynaecology
• Gynaecology
• Occupational Medicine
• Ophthalmology
• Otolaryngology (ENT)
• Browse content in Paediatrics
• Neonatology
• Browse content in Pathology
• Chemical Pathology
• Clinical Cytogenetics and Molecular Genetics
• Histopathology
• Medical Microbiology and Virology
• Patient Education and Information
• Browse content in Pharmacology
• Psychopharmacology
• Browse content in Popular Health
• Caring for Others
• Complementary and Alternative Medicine
• Self-help and Personal Development
• Browse content in Preclinical Medicine
• Cell Biology
• Molecular Biology and Genetics
• Reproduction, Growth and Development
• Primary Care
• Professional Development in Medicine
• Browse content in Psychiatry
• Addiction Medicine
• Child and Adolescent Psychiatry
• Forensic Psychiatry
• Learning Disabilities
• Old Age Psychiatry
• Psychotherapy
• Browse content in Public Health and Epidemiology
• Epidemiology
• Public Health
• Browse content in Radiology
• Clinical Radiology
• Interventional Radiology
• Nuclear Medicine
• Radiation Oncology
• Reproductive Medicine
• Browse content in Surgery
• Cardiothoracic Surgery
• Gastro-intestinal and Colorectal Surgery
• General Surgery
• Neurosurgery
• Paediatric Surgery
• Peri-operative Care
• Plastic and Reconstructive Surgery
• Surgical Oncology
• Transplant Surgery
• Trauma and Orthopaedic Surgery
• Vascular Surgery
• Browse content in Science and Mathematics
• Browse content in Biological Sciences
• Aquatic Biology
• Biochemistry
• Bioinformatics and Computational Biology
• Developmental Biology
• Ecology and Conservation
• Evolutionary Biology
• Genetics and Genomics
• Microbiology
• Molecular and Cell Biology
• Natural History
• Plant Sciences and Forestry
• Research Methods in Life Sciences
• Structural Biology
• Systems Biology
• Zoology and Animal Sciences
• Browse content in Chemistry
• Analytical Chemistry
• Computational Chemistry
• Crystallography
• Environmental Chemistry
• Industrial Chemistry
• Inorganic Chemistry
• Materials Chemistry
• Medicinal Chemistry
• Mineralogy and Gems
• Organic Chemistry
• Physical Chemistry
• Polymer Chemistry
• Study and Communication Skills in Chemistry
• Theoretical Chemistry
• Browse content in Computer Science
• Artificial Intelligence
• Computer Architecture and Logic Design
• Game Studies
• Human-Computer Interaction
• Mathematical Theory of Computation
• Programming Languages
• Software Engineering
• Systems Analysis and Design
• Virtual Reality
• Browse content in Computing
• Business Applications
• Computer Games
• Computer Security
• Computer Networking and Communications
• Digital Lifestyle
• Graphical and Digital Media Applications
• Operating Systems
• Browse content in Earth Sciences and Geography
• Atmospheric Sciences
• Environmental Geography
• Geology and the Lithosphere
• Maps and Map-making
• Meteorology and Climatology
• Oceanography and Hydrology
• Palaeontology
• Physical Geography and Topography
• Regional Geography
• Soil Science
• Urban Geography
• Browse content in Engineering and Technology
• Agriculture and Farming
• Biological Engineering
• Civil Engineering, Surveying, and Building
• Electronics and Communications Engineering
• Energy Technology
• Engineering (General)
• Environmental Science, Engineering, and Technology
• History of Engineering and Technology
• Mechanical Engineering and Materials
• Technology of Industrial Chemistry
• Transport Technology and Trades
• Browse content in Environmental Science
• Applied Ecology (Environmental Science)
• Conservation of the Environment (Environmental Science)
• Environmental Sustainability
• Environmentalist Thought and Ideology (Environmental Science)
• Management of Land and Natural Resources (Environmental Science)
• Natural Disasters (Environmental Science)
• Nuclear Issues (Environmental Science)
• Pollution and Threats to the Environment (Environmental Science)
• Social Impact of Environmental Issues (Environmental Science)
• History of Science and Technology
• Browse content in Materials Science
• Ceramics and Glasses
• Composite Materials
• Metals, Alloying, and Corrosion
• Nanotechnology
• Browse content in Mathematics
• Applied Mathematics
• Biomathematics and Statistics
• History of Mathematics
• Mathematical Education
• Mathematical Finance
• Mathematical Analysis
• Numerical and Computational Mathematics
• Probability and Statistics
• Pure Mathematics
• Browse content in Neuroscience
• Cognition and Behavioural Neuroscience
• Development of the Nervous System
• Disorders of the Nervous System
• History of Neuroscience
• Invertebrate Neurobiology
• Molecular and Cellular Systems
• Neuroendocrinology and Autonomic Nervous System
• Neuroscientific Techniques
• Sensory and Motor Systems
• Browse content in Physics
• Astronomy and Astrophysics
• Atomic, Molecular, and Optical Physics
• Biological and Medical Physics
• Classical Mechanics
• Computational Physics
• Condensed Matter Physics
• Electromagnetism, Optics, and Acoustics
• History of Physics
• Mathematical and Statistical Physics
• Measurement Science
• Nuclear Physics
• Particles and Fields
• Plasma Physics
• Quantum Physics
• Relativity and Gravitation
• Semiconductor and Mesoscopic Physics
• Browse content in Psychology
• Affective Sciences
• Clinical Psychology
• Cognitive Neuroscience
• Cognitive Psychology
• Criminal and Forensic Psychology
• Developmental Psychology
• Educational Psychology
• Evolutionary Psychology
• Health Psychology
• History and Systems in Psychology
• Music Psychology
• Neuropsychology
• Organizational Psychology
• Psychological Assessment and Testing
• Psychology of Human-Technology Interaction
• Psychology Professional Development and Training
• Research Methods in Psychology
• Social Psychology
• Browse content in Social Sciences
• Browse content in Anthropology
• Anthropology of Religion
• Human Evolution
• Medical Anthropology
• Physical Anthropology
• Regional Anthropology
• Social and Cultural Anthropology
• Theory and Practice of Anthropology
• Browse content in Business and Management
• Business History
• Business Strategy
• Business Ethics
• Business and Government
• Business and Technology
• Business and the Environment
• Comparative Management
• Corporate Governance
• Corporate Social Responsibility
• Entrepreneurship
• Health Management
• Human Resource Management
• Industrial and Employment Relations
• Industry Studies
• Information and Communication Technologies
• International Business
• Knowledge Management
• Management and Management Techniques
• Operations Management
• Organizational Theory and Behaviour
• Pensions and Pension Management
• Public and Nonprofit Management
• Strategic Management
• Supply Chain Management
• Browse content in Criminology and Criminal Justice
• Criminal Justice
• Criminology
• Forms of Crime
• International and Comparative Criminology
• Youth Violence and Juvenile Justice
• Development Studies
• Browse content in Economics
• Agricultural, Environmental, and Natural Resource Economics
• Asian Economics
• Behavioural Finance
• Behavioural Economics and Neuroeconomics
• Econometrics and Mathematical Economics
• Economic Methodology
• Economic Systems
• Economic History
• Economic Development and Growth
• Financial Markets
• Financial Institutions and Services
• General Economics and Teaching
• Health, Education, and Welfare
• History of Economic Thought
• International Economics
• Labour and Demographic Economics
• Law and Economics
• Macroeconomics and Monetary Economics
• Microeconomics
• Public Economics
• Urban, Rural, and Regional Economics
• Welfare Economics
• Browse content in Education
• Adult Education and Continuous Learning
• Care and Counselling of Students
• Early Childhood and Elementary Education
• Educational Equipment and Technology
• Educational Strategies and Policy
• Higher and Further Education
• Organization and Management of Education
• Philosophy and Theory of Education
• Schools Studies
• Secondary Education
• Teaching of a Specific Subject
• Teaching of Specific Groups and Special Educational Needs
• Teaching Skills and Techniques
• Browse content in Environment
• Applied Ecology (Social Science)
• Climate Change
• Conservation of the Environment (Social Science)
• Environmentalist Thought and Ideology (Social Science)
• Natural Disasters (Environment)
• Social Impact of Environmental Issues (Social Science)
• Browse content in Human Geography
• Cultural Geography
• Economic Geography
• Political Geography
• Browse content in Interdisciplinary Studies
• Communication Studies
• Museums, Libraries, and Information Sciences
• Browse content in Politics
• African Politics
• Asian Politics
• Chinese Politics
• Comparative Politics
• Conflict Politics
• Elections and Electoral Studies
• Environmental Politics
• European Union
• Foreign Policy
• Gender and Politics
• Human Rights and Politics
• Indian Politics
• International Relations
• International Organization (Politics)
• International Political Economy
• Irish Politics
• Latin American Politics
• Middle Eastern Politics
• Political Theory
• Political Methodology
• Political Communication
• Political Philosophy
• Political Sociology
• Political Behaviour
• Political Economy
• Political Institutions
• Politics and Law
• Public Administration
• Public Policy
• Quantitative Political Methodology
• Regional Political Studies
• Russian Politics
• Security Studies
• State and Local Government
• UK Politics
• US Politics
• Browse content in Regional and Area Studies
• African Studies
• Asian Studies
• East Asian Studies
• Japanese Studies
• Latin American Studies
• Middle Eastern Studies
• Native American Studies
• Scottish Studies
• Browse content in Research and Information
• Research Methods
• Browse content in Social Work
• Addictions and Substance Misuse
• Adoption and Fostering
• Care of the Elderly
• Child and Adolescent Social Work
• Couple and Family Social Work
• Developmental and Physical Disabilities Social Work
• Direct Practice and Clinical Social Work
• Emergency Services
• Human Behaviour and the Social Environment
• International and Global Issues in Social Work
• Mental and Behavioural Health
• Social Justice and Human Rights
• Social Policy and Advocacy
• Social Work and Crime and Justice
• Social Work Macro Practice
• Social Work Practice Settings
• Social Work Research and Evidence-based Practice
• Welfare and Benefit Systems
• Browse content in Sociology
• Childhood Studies
• Community Development
• Comparative and Historical Sociology
• Economic Sociology
• Gender and Sexuality
• Gerontology and Ageing
• Health, Illness, and Medicine
• Marriage and the Family
• Migration Studies
• Occupations, Professions, and Work
• Organizations
• Population and Demography
• Race and Ethnicity
• Social Theory
• Social Movements and Social Change
• Social Research and Statistics
• Social Stratification, Inequality, and Mobility
• Sociology of Religion
• Sociology of Education
• Sport and Leisure
• Urban and Rural Studies
• Browse content in Warfare and Defence
• Defence Strategy, Planning, and Research
• Land Forces and Warfare
• Military Administration
• Military Life and Institutions
• Naval Forces and Warfare
• Other Warfare and Defence Issues
• Peace Studies and Conflict Resolution
• Weapons and Equipment

• < Previous chapter
• Next chapter >

4 The Kuznets Curve: Yesterday and Tomorrow

• Published: May 2006
• Cite Icon Cite
• Permissions Icon Permissions

The Kuznets curve hypothesis has been one of the most debated issues in development economics since the mid-1950s. In a nutshell, the hypothesis simply states that income inequality should follow an inverse-U shape along the development process: first rising with industrialization and then declining, as more and more workers join the high-productivity sectors of the economy. Today, the Kuznets curve is widely held to have doubled back on itself, especially in the United States, with the period of falling inequality during the first half of the 20th century being followed by a sharp reversal of the trend since the 1970s. This essay looks at this “technical change” view of inequality dynamics, whereby waves of technological innovations generated waves of inverse-U curves. It considers the inequality decline that took place in the West during the first half of the 20th century, arguing that recent historical research is rather damaging for Kuznets’s interpretation: the reasons why inequality declined in rich countries seem to be due to very specific shocks and circumstances that do not have much to do with the migration process described by Kuznets, and that are very unlikely to occur again in today’s poor countries. It takes a broader perspective on the technical change view of inequality dynamics, drawing both from historical experience and from more recent trends. It argues that this view has proven to be excessively naïve and that country-specific institutions often play a role that is at least as important as technological waves.

Signed in as

Institutional accounts.

• GoogleCrawler [DO NOT DELETE]
• Google Scholar Indexing

Personal account

• Sign in with email/username & password
• Get email alerts
• Save searches
• Purchase content
• Activate your purchase/trial code

Institutional access

• Sign in with a library card Sign in with username/password Recommend to your librarian
• Institutional account management
• Get help with access

Access to content on Oxford Academic is often provided through institutional subscriptions and purchases. If you are a member of an institution with an active account, you may be able to access content in one of the following ways:

IP based access

Typically, access is provided across an institutional network to a range of IP addresses. This authentication occurs automatically, and it is not possible to sign out of an IP authenticated account.

Sign in through your institution

Choose this option to get remote access when outside your institution. Shibboleth/Open Athens technology is used to provide single sign-on between your institution’s website and Oxford Academic.

• Click Sign in through your institution.
• Select your institution from the list provided, which will take you to your institution's website to sign in.
• When on the institution site, please use the credentials provided by your institution. Do not use an Oxford Academic personal account.
• Following successful sign in, you will be returned to Oxford Academic.

If your institution is not listed or you cannot sign in to your institution’s website, please contact your librarian or administrator.

Sign in with a library card

Enter your library card number to sign in. If you cannot sign in, please contact your librarian.

Society Members

Society member access to a journal is achieved in one of the following ways:

Sign in through society site

Many societies offer single sign-on between the society website and Oxford Academic. If you see ‘Sign in through society site’ in the sign in pane within a journal:

• Click Sign in through society site.
• When on the society site, please use the credentials provided by that society. Do not use an Oxford Academic personal account.

If you do not have a society account or have forgotten your username or password, please contact your society.

Sign in using a personal account

Some societies use Oxford Academic personal accounts to provide access to their members. See below.

A personal account can be used to get email alerts, save searches, purchase content, and activate subscriptions.

Some societies use Oxford Academic personal accounts to provide access to their members.

Viewing your signed in accounts

Click the account icon in the top right to:

• View your signed in personal account and access account management features.
• View the institutional accounts that are providing access.

Signed in but can't access content

Oxford Academic is home to a wide variety of products. The institutional subscription may not cover the content that you are trying to access. If you believe you should have access to that content, please contact your librarian.

For librarians and administrators, your personal account also provides access to institutional account management. Here you will find options to view and activate subscriptions, manage institutional settings and access options, access usage statistics, and more.

Our books are available by subscription or purchase to libraries and institutions.

• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Rights and permissions
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Final dates! Join the tutor2u subject teams in London for a day of exam technique and revision at the cinema. Learn more →

Reference Library

Collections

• See what's new
• All Resources
• Student Resources
• Assessment Resources
• Teaching Resources
• CPD Courses
• Livestreams

Study notes, videos, interactive activities and more!

Economics news, insights and enrichment

Currated collections of free resources

Browse resources by topic

• All Economics Resources

Resource Selections

Currated lists of resources

Study Notes

What is the Kuznets Inequality Curve?

Last updated 11 Oct 2023

• Share on Facebook
• Share on Twitter
• Share by Email

The Kuznets curve, named after economist Simon Kuznets, is a graphical representation of the relationship between economic development and income inequality. It suggests that as an economy develops from a low-income agrarian society to a higher-income industrial and then post-industrial society, income inequality follows a specific pattern. The Kuznets curve is often depicted as an inverted U-shaped curve. Here's a simplified explanation of the Kuznets curve:

• Low-Income Stage (Agrarian Economy): At the initial stage of economic development, when a society is primarily agrarian, income inequality tends to be relatively low. In agrarian economies, most people are engaged in similar occupations, and there are limited opportunities for significant income disparities.
• High-Income Stage (Industrialization): As the economy develops and transitions into an industrial phase, income inequality may increase. Industrialization often leads to the growth of cities and the emergence of new industries. This can result in wage disparities between skilled and unskilled workers, as well as between urban and rural areas. Income inequality rises during this phase.
• Turning Point: The Kuznets curve suggests that there is a "turning point" at which income inequality reaches its peak. This turning point is often associated with a shift from industrialization to a more advanced, post-industrial economy.
• High-Income Stage (Post-Industrial or High-Income Economy): After reaching the turning point, as the economy becomes more post-industrial, income inequality is expected to decline. In post-industrial societies, there may be more emphasis on service industries, education, and technology, which can lead to a more even distribution of income.

The Kuznets curve is often used to describe historical trends in industrialized Western economies during the 20th century. According to this hypothesis, as these countries transitioned from primarily agrarian to industrial economies, income inequality increased, peaking around the mid-20th century, and then began to decrease.

It's important to note that the Kuznets curve is a simplification of the complex relationship between economic development and income inequality. It is not a universal law, and its applicability varies among different countries and regions. The turning point and the exact shape of the curve can differ depending on various factors, including government policies, labor market conditions, and social institutions.

Moreover, some economists argue that in today's globalized world, the Kuznets curve may not be as relevant, as the relationship between economic development and income inequality is influenced by a wide range of factors, including technological change, globalization, and policy choices. Nonetheless, the Kuznets curve remains a useful concept for understanding the historical evolution of income inequality in certain contexts.

• Simon Kuznets
• Kuznets (Inequality) Curve

You might also like

Economics of inequality (revision video).

Topic Videos

Searching for an Environmental Kuznets Curve

6th November 2015

Environmental Kuznets Curve (EKC)

Economic growth - rising inequality and rapid growth, rostow's five stages of economic growth model, the modern kuznets curve - a second wave of inequality.

26th March 2016

Inequality - Incomes of richest 1% in the USA surge

22nd February 2023

2.5.4 Impact of Economic Growth (Edexcel A-Level Economics Teaching PowerPoint)

Teaching PowerPoints

Our subjects

• › Criminology
• › Economics
• › Geography
• › Health & Social Care
• › Psychology
• › Sociology
• › Teaching & learning resources
• › Student revision workshops
• › Online student courses
• › CPD for teachers
• › Livestreams
• › Teaching jobs

Boston House, 214 High Street, Boston Spa, West Yorkshire, LS23 6AD Tel: 01937 848885

• › Contact us
• › Terms of use
• › Privacy & cookies

© 2002-2024 Tutor2u Limited. Company Reg no: 04489574. VAT reg no 816865400.

• Search Search Please fill out this field.
• Macroeconomics

Simon Kuznets: Who Was He and What Is the Kuznets Curve?

Who Was Simon Kuznets?

Simon Kuznets, a Russian-American development economist and statistician, was awarded the 1971 Nobel Memorial Prize in Economics for his research on economic growth. He set the standard for national income accounting, enabling accurate estimates of gross national product to be calculated for the first time.

Key Takeaways

• Simon Kuznets, a Russian-American economist, set the standard for national income accounting that helped advance ideas of Keynesian economics and the study of econometrics.
• Kuznets is also known for the Kuznets curve, which hypothesizes that industrializing nations experience a rise and subsequent decline in income inequality.
• The rise in inequality occurs after rural labor migrates to urban areas and becomes socially mobile. After a certain income level is reached, inequality declines as a welfare state takes hold.
• A modification of the curve, known as environmental Kuznets curve, has become popular to chart the rise and decline of pollution in an industrializing nation's economy.

Understanding Simon Kuznets

Simon Kuznets set the standard for national income accounting —funded by the nonprofit National Bureau of Economic Research . His measures of savings, consumption and investment helped advance Keynesian economics and advanced the study of econometrics. He also helped lay the foundation for the study of trade cycles, known as "Kuznets cycles," and developed ideas about the relationship between economic growth and income inequality.

Kuznets was born in Ukraine in 1901, and moved to the U.S. in 1922. He earned his Ph.D. from Columbia University and was a professor of economics and statistics at the University of Pennsylvania (1930-54), a professor of political economy at Johns Hopkins (1954-60), and a professor of economics at Harvard (1960-71). He died in 1985 in Cambridge, MA.

Kuznets Curve

Kuznets’ work on economic growth and income distribution led him to hypothesize that industrializing nations experience a rise and subsequent decline in economic inequality, characterized as an inverted "U"—the “Kuznets curve."

He thought economic inequality would increase as rural labor migrated to the cities, keeping wages down as workers competed for jobs. But according to Kuznets, social mobility increases again once a certain level of income was reached in “modern” industrialized economies, as the welfare state takes hold.

However, since Kuznets postulated this theory in the 1970s, income inequality has increased in advanced developed countries—although inequality has declined in fast-growing East Asian countries.

Environmental Kuznets Curve

A modification of Kuznets curve has become popular to chart the rise and subsequent decline in pollution levels of developing economies. First developed by Gene Grossman and Alan Krueger in a 1995 NBER paper and later popularized by the World Bank, the environmental Kuznets curve follows the same basic pattern as the original Kuznets curve.

Thus, environmental indicators deteriorate as an economy industrializes until a turning point is reached. The indicators then begin improving again with the aid of new technology and more money that is funneled back to society to improve the environment.

There is mixed empirical evidence to prove validity of the environmental Kuznets curve. For example, carbon emissions have steadily risen for both developed and developing economies. The development of modern carbon trading infrastructure also means that developed economies are not actually reducing pollution but exporting it to developing economies, which are also involved in producing goods for them.

That said, certain types of pollutants declined as an economy industrialized. For example, sulphur dioxide levels decreased in the United States with increased regulation even as the number of cars on its roads held steady or increased.

Evidence and Criticism of Kuznets Curve

Empirical evidence of Kuznets curve has been mixed. The industrialization of English society followed the curve's hypothesis. The Gini coefficient , a measure of inequality in society, in England rose to 0.627 in 1871 from 0.400 in 1823. By 1901, however, it had fallen to 0.443. The rapidly-industrializing societies of France, Germany, and Sweden also followed a similar trajectory of inequality around the same time.

But Netherlands and Norway had a different experience and inequality declined, for the most part, consistently as their societies transitioned from agrarian economies to industrial ones. The East Asian economies - Japan, South Korea, and Taiwan - also witnessed a constant decline in their inequality numbers during their periods of industrialization.

Different theories have been put forward to explain these anomalies. Some ascribe it to cultural quirks. That explanation, however, does not account for the experiences of Netherlands and Norway in contrast to the rest of Europe.

Others have focused on development of political systems that enabled rapid redistribution of wealth. For example, Daron Acemoglu and James Robinson posited that the inequality due to capitalist industrialization contained "seeds of its own destruction" and gave way to political and labor reform in Britain and France, enabling redistribution of wealth.

In East Asian economies land reforms that occurred in the 1940s and 1950s helped pave the way for equitable redistribution even though political reform was delayed. In other words, it was politics, and not economics as Kuznets suggested, that determined inequality levels.

When he defined the concept, Kuznets himself suggested that there was much more work to be done and data to be collected in order to conclusively prove the relationship between economic development and inequality.

• Terms of Service
• Editorial Policy
• Privacy Policy
• Your Privacy Choices

Kuznet’s Inverted U-Hypothesis on Income Inequality | Economics

Simon Kuznets put forward the hypothesis that relationship between per capita national income and the degree of inequality in income distribution may be of the form of inverted-U. Due to limitations of data he used an inequality measure of the ratio of income share of the richest 20 per cent of the population to the bottom 60 per cent of the population known as Kuznets’ ratio.

According to the Kuznets’ Inverted U-hypothesis, as per capita national income of a country increases, in the initial stages of growth, inequality in income distribution rises and after reaching the highest degree in the intermediate level the income inequality falls. This is shown in Fig. 65.4 where as a country develops and its per capita income rises, the degree of income inequality initially rises and after reaching the maximum level, it falls as GDP per capita increases further.

As time series data of the transition of the poor underdeveloped countries from underdeveloped stage to the developed stage was not available, he used the data of cross section of countries including both developed and developing countries. In his 1955 study he calculated the Kuznets’ ratios and found that the developing countries tend to have a higher degree of inequality whereas the rich developed countries tend to have a lower degree of inequality.

Kuznets Curve, Hypothesis, Graph, Criticism, Applications

Kuznets Curve describes the relationship between economic development and income inequality. Read all about Kuznets Curve Hypothesis, Graph, Applications & its Criticism for the UPSC exam.

Table of Contents

Kuznets Curve

The Kuznets curve is an economic concept that describes the relationship between economic development and income inequality. The curve was named after Simon Kuznets, a Nobel Prize-winning economist who first proposed the idea in the 1950s.

According to the Kuznets curve, income inequality tends to increase in the early stages of economic development, as a country moves from an agricultural-based economy to an industrial-based one. However, as economic development continues and a country becomes more prosperous, income inequality begins to decrease.

Kuznets Curve Graph

The Kuznets Curve is often depicted as an inverted U-shape, with income inequality rising during the early stages of development and then declining once a certain level of economic prosperity is reached. However, there is debate over whether this pattern holds true in all cases, and some researchers have suggested that other factors, such as government policies, can influence the relationship between economic development and income inequality

Read about: Phillips Curve

Kuznetz Curve Hypothesis 

The Kuznets curve hypothesis is the idea that economic development is associated with a specific pattern of income inequality over time, as described by the Kuznets curve. The hypothesis suggests that income inequality initially increases as a country undergoes economic development, but then reaches a peak and begins to decline as the country becomes more prosperous.

Kuznets observed that in the early stages of industrialization, income inequality tended to increase as workers moved from low-paying agricultural jobs to higher-paying industrial jobs. However, as a country became more prosperous and developed, income inequality tended to decline as workers gained access to better education, training, and job opportunities.

Read about: Gini Coefficient

Kuznets Curve in Economics Applications

The Kuznets curve has several applications in economics, including:

Understanding Relationship between Economic Development and Income Inequality

The Kuznets curve provides a framework for understanding how income inequality changes over time as a country develops. This can help policymakers to design interventions to reduce income inequality in the early stages of development and promote economic growth in the later stages.

Identifying the Drivers of Income Inequality

The Kuznets curve suggests that income inequality is driven by structural changes in the economy as a country develops. By examining the factors that contribute to income inequality at different stages of development, policymakers can identify the most effective ways to reduce inequality.

Evaluating the Effectiveness of Economic Policies

The Kuznets curve can be used to evaluate the effectiveness of economic policies designed to reduce income inequality. By comparing income inequality before and after the implementation of a policy, policymakers can determine whether the policy has been successful in achieving its goals.

Predicting the Future of Income Inequality

The Kuznets curve can also be used to make predictions about the future of income inequality in a country. By examining the current stage of development and economic growth, policymakers can estimate where a country is on the Kuznets curve and predict how income inequality is likely to change in the future.

Read about: Lorenz Curve

Kuznets Curve Criticism 

While the Kuznets curve hypothesis has been influential in shaping our understanding of the relationship between economic development and income inequality, it has also been subject to criticism. Some researchers argue that the pattern described by the Kuznets curve is not universal and that factors such as government policies, social institutions, and cultural norms can influence the relationship between economic development and income inequality in different ways.

Read about: Purchasing Power Parity

Kuznets Curve UPSC 

The Kuznets Curve is a relevant topic for the UPSC Economics Syllabus and has implications for other subjects such as sociology, public administration, and international relations. Aspirants preparing for the UPSC exam can benefit from UPSC Online Coaching and UPSC Mock Test to understand income inequality patterns, economic policies, and limitations of the Curve.

Read about: Basel Norms

Sharing is caring!

Kuznets Curve FAQs

What is the concept of kuznets curve.

The Kuznets curve is a hypothesis that income inequality initially increases and then declines with economic development.

What is the Lorenz curve and Kuznets curve?

The Lorenz curve measures income distribution, while the Kuznets curve relates economic development to income inequality.

What is the concept of Lorenz curve?

The Lorenz curve is a graph showing the degree of income inequality in a society.

Why is it called a Lorenz curve?

It's called a Lorenz curve after Max O. Lorenz who developed it in 1905.

What are the two variables of Kuznets curve?

The two variables of the Kuznets curve are economic development and income inequality.

I, Sakshi Gupta, am a content writer to empower students aiming for UPSC, PSC, and other competitive exams. My objective is to provide clear, concise, and informative content that caters to your exam preparation needs. I strive to make my content not only informative but also engaging, keeping you motivated throughout your journey!

Leave a comment

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

• UPSC Online Coaching
• UPSC Exam 2024
• UPSC Syllabus 2024
• UPSC Prelims Syllabus 2024
• UPSC Mains Syllabus 2024
• UPSC Exam Pattern 2024
• UPSC Age Limit 2024
• UPSC Calendar 2024
• UPSC Syllabus in Hindi
• UPSC Full Form

Recent Posts

• UPPSC Exam 2024
• UPPSC Calendar
• UPPSC Syllabus 2024
• UPPSC Exam Pattern 2024
• UPPSC Application Form 2024
• UPPSC Eligibility Criteria 2024
• UPPSC Admit card 2024
• UPPSC Salary And Posts
• UPPSC Cut Off
• UPPSC Previous Year Paper

BPSC Exam 2024

• BPSC 70th Notification
• BPSC 69th Exam Analysis
• BPSC Admit Card
• BPSC Syllabus
• BPSC Exam Pattern
• BPSC Cut Off
• BPSC Question Papers

IB ACIO Exam

• IB ACIO Salary
• IB ACIO Syllabus

CSIR SO ASO Exam

• CSIR SO ASO Exam 2024
• CSIR SO ASO Result 2024
• CSIR SO ASO Exam Date
• CSIR SO ASO Question Paper
• CSIR SO ASO Answer key 2024
• CSIR SO ASO Exam Date 2024
• CSIR SO ASO Syllabus 2024

Study Material Categories

• Daily The Hindu Analysis
• Daily Practice Quiz for Prelims
• Daily Answer Writing
• Daily Current Affairs
• Indian Polity
• Environment and Ecology
• Art and Culture
• General Knowledge
• Biographies

IMPORTANT EXAMS

• Terms & Conditions
• Return & Refund Policy
• Privacy Policy

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: 21 February 2024

Rethinking the environmental Kuznets curve hypothesis across 214 countries: the impacts of 12 economic, institutional, technological, resource, and social factors

• Qiang Wang   ORCID: orcid.org/0000-0002-8751-8093 1 , 2 ,
• Yuanfan Li 1 &
• Rongrong Li 1 , 2  

Humanities and Social Sciences Communications volume  11 , Article number:  292 ( 2024 ) Cite this article

1272 Accesses

1 Altmetric

Metrics details

• Development studies
• Environmental studies

Research over the past three decades has provided rich empirical evidence for the inverted U-shaped EKC theory, but current problems facing advancing climate mitigation actions require us to re-examine the shape of global EKC rigorously. This paper examined the N-shaped EKC in a panel of 214 countries with 12 traditional and emerging variables, including institutions and risks, information and communication technology (ICT), artificial intelligence(AI), resource and energy use, and selected social factors. The two-dimensional Tapio decoupling model based on N-shaped EKC to group homogeneous countries is developed to explore the inter-group heterogeneous carbon emission effects of each variable. Global research results show that the linear and cubic terms of GDP per capita are significantly positive, while the quadratic term is significantly negative, regardless of whether additional variables are added. This means the robust existence of an N-shaped EKC. Geopolitical risk, ICT, and food security are confirmed to positively impact per capita carbon emissions, while the impact of composite risk, institutional quality, digital economy, energy transition, and population aging are significantly negative. The impact of AI, natural resource rents, trade openness, and income inequality are insignificant. The inflection points of the N-shaped EKC considering all additional variables are 45.08 and 73.44 thousand US dollars, respectively. Combining the turning points and the calculated decoupling coefficients, all countries are categorized into six groups based on the two-dimensional decoupling model. The subsequent group regression results show heterogeneity in the direction and magnitude of the carbon emission impacts of most variables. Finally, differentiated carbon emission reduction strategies for countries in six two-dimensional decoupling stages are proposed.

Similar content being viewed by others

The economic commitment of climate change

Worldwide divergence of values

Climate damage projections beyond annual temperature

Introduction.

The suitability and stability of the Earth’s climate are crucial for human survival (Carlson et al., 2022 ; Cavicchioli et al., 2019 ) and development (Brown et al., 2023 ; Ray et al., 2015 ; Diaz and Moore, 2017 ). However, as of January 2024, global surface temperatures have accelerated by ~1.5 °C since the pre-industrial period (Berkeley, 2024 ). According to the IPCC’s 7th climate assessment report (AR 7), the global warming is “unequivocally” caused primarily by human activities and anthropogenic greenhouse gas emissions (IPCC, 2023 ). Upon the societal acknowledgment of the causality, the question of how to transform the existing development paradigm to conform with climate mitigation principles emerges as a central focus for the governmental (Duan et al., 2021 ; Dai et al., 2023 ) and scholarly attention (Fawzy et al., 2020 ; Hoegh-Guldberg et al., 2019 ; Fankhauser et al., 2022 ).

The environmental Kuznets curve (EKC) is among the most prevalent economic-environmental simulation models utilized for policy analysis (Zou et al., 2022 ; Koondhar et al., 2021 ; Mao et al., 2022 ). The EKC hypothesis linked economic development to environmental quality in such a paradigm: per capita income and per capita carbon dioxide emissions increase together until a certain inflection point in income is reached, after which the growth of pollutants levels off and then overturns (Grossman and Krueger, 1995 ). Such inverted U-shaped nonlinear patterns have been empirically confirmed in countless number of early studies and explanations are mainly given by inefficient and dirty production techniques in the pre-industrial stages and the strengthened environmental regulations and advanced clean technologies in the developed stages (Balsalobre-Lorente et al., 2018 ). Over the past three decades, such paradigms of “grow first and clean later” have often served as a benchmark for other climate models and even as references for national plans (Koondhar et al., 2021 ; Kaika and Zervas, 2013a ).

However, many of the news and events we hear and observe recently suggest a weakening relationship between advanced economies and improved environment, which seems to contradict the existing explanation. First, some developed countries including the United States are known to be ecologically deficit (Usman et al., 2020 ; Lark et al., 2020 ). Moreover, in November 2020, as the world’s largest economy and second-largest carbon emitter, the U.S. vowed to withdraw from the Paris Agreement to restart its traditional and highly polluting sectors (Schiermeier, 2020 ; Roelfsema et al., 2020 ). Finally, many developed countries, including Germany, the Netherlands, and Austria, are once again turning to coal for electricity and heating as they struggle with a European energy crisis triggered by the Russia-Ukraine conflict beginning in 2022 (Tollefson, 2022 ; Guan et al., 2023 ).

These events highlight the need to revisit the global EKC, specifically discussing the N-shaped curve. On the one hand, the new paradigm of the N-shaped curve is more suitable for the current actual situation than the inverted U-shaped EKC. This is because the carbon emission trajectory may still be bent upward at a high level of economic development if the technological effects brought about by green innovation cannot keep up with the growing scale effects (Lorente and Álvarez-Herranz, 2016 ). On the other hand, once the “development first, clean later” commitment of the previous inverted U-shaped EKC theory is disillusioned, then developing countries can no longer sit back and be excluded from the climate agenda and emission reduction plans (Koondhar et al., 2021 ; Kaika and Zervas, 2013a ). Hence, an accurate understanding of the current shape of global EKC is critical for different categories of countries to identify their unique drivers of carbon emissions and adopt prompt action.

The current situation, on the other side, also demonstrates the complexity of the determining rules of carbon emissions. Although some additional variables including trade openness, institutional quality, and energy consumption have been widely incorporated into the EKC framework (Leal and Marques, 2022 ), recent research has established some new links between external factors and carbon emissions. For example, geopolitical risks and national political, economic, and financial risks are proven to have the potential to change countries’ economic development and energy use patterns (Anser et al., 2021 ; Hassan et al., 2022 ; Qiang Wang et al., 2023a ). Digital technology and artificial intelligence, as key to the third and fourth industrial revolutions, have been found to have carbon-reducing effects in some studies (Jinning Zhang et al., 2022a ; Ding et al., 2023 ). The transition from traditional fossil fuels to new energy sources has been empirically confirmed to effectively mitigate air pollution and accelerate carbon neutrality (Feng Dong et al., 2022b ; Naeem et al., 2023 ). Finally, some social issues, including population aging (Fan et al., 2021 ) and food security (Naseem et al., 2020 ), are also included in the environmental analysis. As (Stern, 2017 ) note, it is necessary to reveal the effects of other variables other than economic growth within the EKC framework. Banking on these current issues of global environment and carbon emissions, we ask the following questions: What is the current shape of global EKC when new additional factors are taken into account? Is there heterogeneity in the impact of various factors on carbon emissions in different types of countries?

This article aims to answer the above questions by empirically examining the current shape of the global EKC and fully considering the impact of multiple carbon emission drivers and the heterogeneity in the panel. Specifically, this paper contributes to the literature in the following three aspects. First, this paper studies N-shaped EKC in 214 countries and incorporates 12 traditional or novel institutional, technological, resource, and social factors as additional variables in the EKC equation. This work provides preliminary but crucial information for mitigation actions by elucidating the relationship between global economic development and carbon emissions and identifying potential drivers of carbon emissions. Second, to handle the heterogeneity of countries, we develop a two-dimensional Tapio decoupling model based on N-type EKC to evaluate the economic development stage and decoupling situation of each country. We believe that starting a classified and detailed discussion can help to more systematically reveal the different situations or problems in economic development and carbon emission reduction in countries around the world. Finally, based on the grouping results, we analyze the inter-group heterogeneity of the carbon emission effects of economic and additional variables, providing a baseline reference for countries at different stages to formulate targeted and effective carbon emission reduction policies.

The remainder of this article is arranged as follows. The section “Literature review and theoretical background” reviews studies related to EKC and additional variables and develops key hypotheses. The section “Method and data” briefly introduces the methodology, model and data. In the section “Empirical results”, we discuss the global and group research results. The section “Concluding remarks” is left for conclusion.

Literature review and theoretical background

Literature review, the origin, evolution, and relevant criticism of ekc.

The nonlinear relationship between per capita income and income inequality proposed by (Kuznets, 1955 ) was re-interpreted as the environmental Kuznets curve (EKC) in the environmental economics field in the 1990s. The EKC concept, originally used to describe the inverted U-shaped relationship between economic growth and environmental degradation, was proposed by (Grossman and Krueger, 1991 ) in their landmark study of the environmental impacts of the North American Free Trade Agreement. Early EKC researchers analyzed the relationship between economic growth and various environmental degradation indicators without any other explanatory variables and obtained much evidence for an inverted U-shaped of first positive and then negative linkages between the two (Grossman and Krueger, 1991 ; Beckerman, 1992 ; Grossman and Krueger, 1995 ; Heerink et al., 2001 ).

Subsequent research mainly made two key improvements to the classic EKC empirical research framework. On the one hand, N-shaped EKCs have been theoretically proposed and they were empirically confirmed in global, cross-country, and single-country studies in the latest empirical studies. For example, Allard et al. ( 2018 ) tested the relationship between CO2 emissions and GDP per capita in four income groups (low, upper-middle, lower-middle, and high-income groups) containing a total of 74 countries and found N-shaped EKC for all income groups except upper-middle-income countries. Using ecological footprint as a proxy variable for environmental pollution, Numan et al. ( 2022 ) obtained very similar conclusions among the country groups of four income levels including 85 countries. In addition, N-shaped curves were also found in the case of nine top nuclear-producing countries (Jahanger et al., 2023 ), OPEC countries (Ullah et al., 2023 ), E-10 countries (Fakher et al., 2023 ), and 14 emerging countries (Rashdan et al., 2021 ). N-shaped linkages between growth and pollution were also individually detected in India (Hossain et al., 2023 ), China (Zhengxia et al., 2023 ), Algeria (Shehzad et al., 2022 ), and other countries.

Another major improvement is to add additional variables to the right-hand side of the empirical EKC equations. By stripping out political, social, or technological factors and discussing their impact on carbon emissions separately, scholars have narrowed and clarified the concept of economic development in the traditional EKC equation (Leal and Marques, 2022 ). These studies provide more accurate and richer insights than would be possible without the inclusion of control variables. In this regard, first established by (Cole, 2004 ), trade openness has so far become a key additional variable in the EKC framework. The inclusion of this variable is often relevant for testing the Pollution Haven Hypothesis (PHH). Because under the trade-opening paradigm, polluting industries shifting from developed countries to developing countries will directly affect the level of economic development and carbon emission levels as well as the relationship between the two (Qiang Wang et al., 2023b ). Energy consumption is another well-established additional variable (Qiang Wang et al., 2024a ). In the pioneering literature (Ang, 2008 ), energy consumption was included in the EKC framework and the results confirmed the long-term and positive Granger causality between economic growth, energy use, and pollution in Malaysia. In terms of national institutions, Apergis and Ozturk ( 2015 ) incorporated the four institutional quality variables of political stability, government efficiency, regulatory quality, and corruption control into the research equation and confirmed the EKC hypothesis in a panel of 14 Asian countries. They emphasized that national institutions provide a mechanism within which to share knowledge and diffuse technologies on energy efficiency and emissions control. Lantz and Feng ( 2006 ) and Higón et al. ( 2017 ) discuss the role of technological progress and ICT in the EKC framework. Both results indicate that traditional EKC does not exist and is replaced by an inverted U-shaped techonology-carbon emissions nexus.

Finally, this article reviews some theoretical and empirical criticisms of EKC. They are related to the research ideas of this article and the development directions of future research. The first issue that has been widely criticized relates to the alleged heterogeneity present in the EKC panel study. To provide a broader assessment and reference, a considerable number of EKC studies choose to use cross-sectional or panel data including a group of countries (Kaika and Zervas, 2013b ). Although these studies often argue that the selection of country groups is based on the close links between them (such as regional links, trade links, international agreements, etc.), however, the EKC curves obtained from country groups often prove to be unsuitable for each of them (Leal and Marques, 2022 ). A second thorny criticism questions the practical relevance of the EKC conceptual framework and development model (Gill et al., 2018 ). Even if each country’s EKC curve were accurately assessed, it would only depict a baseline development trajectory for the future. However, the more important issue at present is how to reduce environmental pollution and repair environmental damage according to the EKC insight rather than just “grow now and clean later”(Gill et al., 2018 ). The third issue is proposed in (Stern, 2017 ). After reviewing the 25 years of evolution of EKC research, he worried that the EKC model ignores the impact of other effects except growth and underscored that “they are the opposite force of the scale effect.” In addition, other econometrics issues, including the selection of pollution variables, the sensitivity of results to measurement methods, and the rationality of inflection points, were also pointed out (Leal and Marques, 2022 ).

Research gaps and research design

Through an in-depth discussion of the literature on EKC, we are convinced that our research is based on solid foundations. However, it is necessary to point out the research gap to realize the contribution of this article and promote the development of related literature. First, to the best of the authors’ knowledge, there is still a lack of global panel testing of the existence of N-shaped EKC, especially for a total of 214 countries. Second, existing studies often subjectively select one perspective or a few variables to incorporate into the EKC equation, rather than using a comprehensive variable set to cover various aspects that potentially affect carbon emissions. Third, based on the criticism of the heterogeneity of EKC, scholars often study the shape of EKC in different income groups defined by the World Bank to handle heterogeneity. However, this solution still suffered from two main limitations. On the one hand, it is not effective enough to handle heterogeneity by grouping countries only from the economic perspective, ignoring the different environmental situations faced by countries at the same stage of development. On the other hand, with the second criticism of EKC, previous studies still focus on determining the shape of EKC after grouping, while neglecting the question of how each group of countries can reduce carbon emissions or repair environmental problems, which may lead to a lack of feasible policy implications and action reference.

To address the first issue, we initiated an empirical examination of the existence of N-shaped EKCs in a complete global panel of 214 countries. For the second problem, we selected 12 additional variables to be included in the EKC equation. These variables cover four perspectives: institutions and risks, digital technology, resource and energy utilization, and other social factors, and cover some new variables that have rarely been included in EKC in the past. Regarding the last one, we develop a two-dimensional decoupling model that can fully consider the heterogeneity of countries in terms of economic development and carbon emission decoupling status to achieve country grouping. In the group study, we focused on the heterogeneous impact of various factors rather than the shape of EKC, to propose more targeted and feasible solutions for each group of countries to reduce carbon emissions.

Theoretical background and hypothesis

Based on the research design, we review in this section the theoretical background regarding EKC shape and additional variables to be included in the study. Accordingly, we will propose specific research hypotheses for further empirical research.

Theory about the shape of EKC

The traditional EKC theory believes that there is an inverted-U shape between economic growth and environmental pollution (Grossman and Krueger, 1995 ; 1991 ) and implicitly assumes structural changes in the process of economic growth (Dinda, 2004 ). In other words, as economic development moves from a lower stage to a higher stage, the production sector also changes from agricultural production to industrial production, and finally to the third sector. In addition, under the scheme of EKC, in each of the above processes, scale effects, composition effects, and technical effects will successively dominate. Within this framework, scholars use the relative sizes of technical effects and scale effects to explain changes in the relationship between economic development and environmental pollution (Dinda, 2004 ; Koondhar et al., 2021 ; Andreoni and Levinson, 2001 ): In the first stage of U-shaped EKC, that is, the pre-industrial stage, the development of primary, low-efficiency industries will produce scale effects and pollution. The reason for the emergence of the second stage of EKC is explained as the replacement of dirtier technologies with cleaner technologies in the production process. Therefore technological effects brought about by innovation exceed the scale effects and achieve environmental improvements as in the developed stage (Andreoni and Levinson, 2001 ).

However, scholars supporting N-shaped EKC have proposed a different theory. They believed that even at a higher stage of economic development if innovation cannot keep up with the speed of economic growth and the increase in technological effects is slower than the scale effect, environmental pollution may reappear (Lorente and Álvarez-Herranz, 2016 ). In other words, the shape of the EKC may extend from an inverted U-shape to an N-shape. This N-shaped model assumes that economic growth leads to environmental degradation that reverses with economic progress after reaching a positive peak and that pollution levels approach a negative peak before starting to rise again with further economic progress. Compared with the inverted U-shaped EKC, the most critical gap lies in the rebound of environmental pollution in the third stage. Scholars refer to this pattern as the technological obsolescence effect, representing the re-emergence of scale effects and overcoming the compositional and technological effects that preceded the second turning point (Lorente and Álvarez-Herranz, 2016 ; Balsalobre-Lorente et al., 2018 ). The emergence of this phenomenon is believed to be related to neglected environmental regulations and slow technological innovation. Many studies have shown how technological innovation combined with environmental regulation can weaken technological obsolescence and delay new rising trends in pollution (Álvarez-Herránz et al., 2017 ; Lorente and Álvarez-Herranz, 2016 ). These theoretical analyses indicate that N-shaped EKC may appear when the development of environmental protection awareness, low-carbon energy utilization, and clean production technology is slow. Combined with the large number of real-life cases from developed countries discussed in the introduction, this phenomenon does exist and we proposed the hypothesis below:

H1: The current shape of the global EKC is N-shaped.

H2: The N-shaped EKC remains robust with additional variables.

The relationship between additional variables and carbon emissions

The purpose of this section is to establish the relationship between the involved 12 additional variables and carbon emissions based on previous literature. These factors cover four perspectives: institutions and risks, digital technology, resource and energy utilization, and other social factors.

Sound institutions and a stable national environment are fundamental to a country’s development (Asiedu, 2006 ). They are considered closely linked to carbon emissions through policy, economic, and technical pathways (Karim et al., 2022 ). Institutional quality is a relatively traditional research variable, which represents the effectiveness and stability of a country’s policy formulation, laws, regulations, and governance structures. In a study aimed at validating the EKC in 47 emerging markets and developing economies (Le and Ozturk, 2020 ), institutional quality was found to boost carbon emissions. However, using annual data from 30 sub-Saharan African countries (Karim et al., 2022 ) confirmed that corruption control, regulatory quality, and rule of law in six WGI indices can significantly reduce carbon dioxide emissions. Epidemics and conflicts have recently led to the accumulation of global and domestic risks. National risks and geopolitical risks have attracted the attention of some scholars. However, there is no consensus on the impact of various risks on carbon emissions. Anser et al. ( 2021 ) and Hassan et al. ( 2022 ), respectively, found that geopolitical risk and political risk promoted carbon emissions, while Jun Zhao et al. ( 2021a ) found that financial risks reduce carbon emissions.

Technical effects are the key force in pollution reduction in EKC theory. Among them, digital and communication technologies are considered to have triggered the third industrial revolution. Therefore, the relationship between digital technology and carbon emissions has received much research attention. ICT is a representative digital technology variable, generally represented by the availability of communication equipment or digital functions (Charfeddine and Umlai, 2023 ). Scholars have conducted numerous studies on the ICT-carbon emissions nexus (Qiang Wang et al., 2024b ). As is well summarized by Charfeddine and Umlai ( 2023 ), the research findings cover irrelevant, positive relationships, negative relationships, U-shaped relationships, and inverted U-shaped relationships due to the differences in scope, methods, and proxies in the research. A relatively novel approach to measuring digital technology comes from a so-called digital economy perspective, which pays more attention to the integration among smart technology and the industries than ICT. This stream of research generally constructs multi-index indices to measure the digital economy from dimensions such as ICT infrastructure, economic effects and social benefits. Unlike the fierce controversy in ICT research, most research results support the reduction effect of digital technology on carbon emissions (Feng Dong et al., 2022a ; Yi et al., 2022 ), though opposing views exist (Lu Zhang et al., 2022b ). The latest digital technology variable is related to AI. As the energy consumption of training large language models such as Chatgpt has recently attracted the attention of the public and scholars, the energy and environmental impact of AI has been increasingly studied. In recent empirical studies, most scholars used the industrial robot stock to proxy AI (Zhong et al., 2023 ), and a few scholars constructed a composite index for AI development in specific countries like China (Ding et al., 2023 ). Most research results support the reduction effect of AI or industrial robots on carbon emissions (Zhong et al., 2023 ; Yaya Li et al., 2022 ; Mingfang Dong et al., 2023 ), although (Luan et al., 2022 ) got the opposite result and (Liu et al., 2024 ) hold a nonlinear view.

Human society’s use of natural resources and energy is directly related to anthropogenic carbon emissions. Natural resource rents represent the degree of human utilization of traditional resources. However, the impact of natural resource rents on carbon emissions is subject to intense debate. Empirically Rongrong Li et al. ( 2023 ) found that the exploration of natural resources directly triggers economic growth and subsequently stimulates an increase in ecological footprint. In contrast, combined with clean energy and environmental protection technology, some scholars (Xiaoman et al., 2021 ; Rongrong Li et al., 2024 ) believe that natural resource extraction can also improve the local environment. On the other hand, energy transition represents the result of human beings taking the initiative to eliminate traditional fossil fuels and shift to clean energy. Empirical results support that this systemic transformation of the energy system can effectively reduce carbon emissions in most countries (Dogan and Seker, 2016 ; Inglesi-Lotz and Dogan, 2018 ) though the heterogeneity among country in different income groups is often significant (Nguyen and Kakinaka, 2019 ).

Social issues have recently been increasingly linked to climate change. The first variable to be included in the study is trade openness, which is used to test the existence of the pollution paradise hypothesis in the literature (Qiang Wang et al., 2023b ). The opposite view to PHH is called the pollution halo hypothesis, which holds that developed countries spread advanced clean technologies to host countries in the process of foreign investment (Bashir, 2022 ; Qiang Wang et al., 2023b ). However (Bashir, 2022 ) summarized that no single view currently dominates, and that studies supporting positive and negative correlations between trade openness and carbon emissions are both growing. Income inequality is initially closely related to EKC theory as a key variable. However, the impact of income inequality on carbon emissions is not clear. Scholars explain this as the dual impact of income inequality: severe income inequality weakens the high-carbon consumption capacity of the large number of middle- and low-income groups, while triggering rich people to adopt a high-pollution lifestyle (Rojas-Vallejos and Lastuka, 2020 ). The carbon impact of an aging population is also considered twofold. Although an aging society reduces the production capacity and demand for high-carbon industrial products, it also increases the demand for medical and other services (Zhou et al., 2023 ). The latter is an energy- and pollution-intensive industry. After the pandemic, food safety has become a top priority for many underdeveloped countries. However, judging from a recent review (Cheng et al., 2023 ), empirical studies linking food security and carbon emissions are lacking. The only literature seems to focus on Pakistan, where (Akbar et al., 2019 ) found a negative causal relationship between cereal yields and carbon emissions, while (Naseem et al., 2020 ) revealed that an increase in the food security index may bring additional carbon emissions. Based on the above discussion, we found that the above variables may each have an impact on carbon emissions through various channels. Furthermore, there are controversies in findings in the literature for different research subjects, suggesting that there may be country heterogeneity in the relationship between these variables and carbon emissions. Therefore we propose the following hypothesis:

H3: When included in the EKC model, all of the 12 additional variables have significant impacts on carbon emissions.

H4: The relationship between the 12 additional variables and carbon emissions is heterogeneous among different groups of countries.

Method and data

Methodology, preliminary test methods, multicollinearity test method: variance inflation factor (vif).

Multicollinearity results from one independent variable being accurately linearly predicted by other independent variables, which reduces the stability of parameter estimation for the affected variables. This article aims to analyze the shape of the global EKC curve under the influence of 12 additional variables. Even though we have made the best efforts to select determinants from different lenses to approach the carbon emission curve, the original data used in the calculation of different proxy variables may partially overlap. Therefore, we examine the degree of multicollinearity in the model to ensure balance between comprehensiveness and non-replication of variable combinations.

The variance inflation factor (VIF) tests the severity of the multicollinearity problem by measuring how much multicollinearity increases the variance of the estimated coefficients. For example, in the following multiple regression:

the variance of the estimated value of β 1 is:

where \({{\rm{R}}}_{1}^{2}\) is the determining coefficient of the regression of x1 on x 2 -x 6 . The VIF of x 1 is defined as:

Therefore, VIF 1 is a multiplier of the variance of \({\hat{{\rm{\beta }}}}_{1}\) . A larger VIF means a larger \({\rm{Var}}({\hat{{\rm{\beta }}}}_{{\rm{i}}})\) , and lower accuracy of the estimation of β i . According to a rule of thumb, if the combination of explanatory variables has a VIF greater than 10, it means that serious multicollinearity problems exist and must be solved.

Panel unit root test methods

The basic idea of the unit root test (URT) is to test whether there is an obvious upward or downward trend in variables by proposing the null hypothesis that a unit root exists in the sample sequence. If the null hypothesis is rejected, it means that there is no obvious trend in the sequence and the variable is considered stable. The most commonly used panel URT methods include IPS, LLC, Fisher tests (including Fisher-ADF test and Fisher-PP test), etc (Im et al., 2023 ). Because this paper uses unbalanced data, we choose the Fisher test which applies to our case. Taking fisher-ADF as an example, in the simple autoregressive process of variable y it :

where ρ is the auto-correlation coefficient, and the Fisher-ADF statistic is:

where d i represents the p -value of the ADF test for the i-th group of the cross-section. The significant Fisher-ADF statistic represents the rejection of the null hypothesis that ρ  = 1 in Eq. ( 4 ), which means that the series of variable y it does not have a unit root and is stationary.

Cointegration test methods

Most regression models assume that the disturbance term is a normal random variable with zero mean and constant variance. Once the disturbance term is not stationary, the estimation results will become biased and incredible, which is so-called pseudo-regression. To prevent pseudo-regression problems, the cointegration test is often applied to test whether the station of residual sequence. This paper focuses on the Kao test (Kao, 1999 ), which applies to the unbalanced panels in our case. This method is based on the premise that both the explanatory and dependent variables are first-order stationary. Taking the simple regression form without a trend term as an example, in the following regression equation:

x it and y it should first satisfy the first-order stationary condition:

where i represents the individual and t represents time. u it and v it are random disturbance terms with zero mean and constant variance. Kao ( 1999 ) proposed to use DF or ADF statistics to test the stationarity of ε it . Specifically, the residual autoregressive equation applicable to the DF test is:

The residual autoregressive equation applicable to the ADF test is:

where p represents the selected optimal lag length. v it and v itp are random disturbance terms. The cointegration relationship between x it and y it is tested by the null hypothesis ρ  = 1 and the alternative hypothesis ρ  < 1.

Empirical framework of inverted U-shaped and N-shaped EKC

In the pioneer literature (Grossman and Krueger, 1991 ; 1995 ), the traditional EKC hypothesis was first proposed. This hypothesis holds that there is an inverted U-shaped relationship between environmental pollution levels and per capita GDP. Empirically, scholars use the following econometric models to describe and test the shape of EKC (Dinda, 2004 ):

where EP it represents the environmental pollution level of country i in year t, and pgdp it represents the per capita income level. α 0 theoretically represents other factors influencing environmental pollution levels, while are often simply treated as constants for the convenience of analysis. ε it is the random disturbance term. Researchers judge the shape of the EKC by the sign of α 1 and α 2 : a positive α 1 and a negative α 2 represent a U-shaped EKC, and a negative α 1 and a positive α 2 will yield an inverted U-shape one. If α 1 it is not significant or is 0, it means there is a linear relationship between per capita GDP and environmental pollution.

In the subsequent literature, scholars also considered a more complex N-shaped EKC, which can be expressed by the following equation (Lorente and Álvarez-Herranz, 2016 ):

It can be seen that the cubic term of pgdp is included in this equation. The new equations expand the possible shapes of the EKC. When α1 is nonzero, the curve is cubic; otherwise, it will be reduced to a quadratic curve. Specifically, researchers can follow Table 1 to determine the EKC shape.

Based on the determined shape, the inflection points of the curve can be computed. When GDP per capita passes through these points, the direction of change in environmental pollution will overturn, such as from negative to positive or from positive to negative. Among them, N and inverted N shapes belong to cubic curves, yielding two inflection points, respectively

U and inverted U shapes are quadratic curves with an only inflection point given by:

Two-dimensional Tapio decoupling model based on N-shaped EKC

The EKC hypothesis, whether it is an inverted U-shape or an N-shape, describes a one-to-one relationship between economic development and environmental pollution. However, researchers currently cannot find an EKC that is optimal for all countries (Kaika and Zervas, 2013b ). Therefore, in addition to EKC, other methods are also needed to determine the relationship between economic development and environmental pollution to verify and supplement EKC.

Understanding EKC from the perspective of Tapio decoupling coefficient

The Tapio decoupling coefficient exploits the relative elasticity between economic development and environmental pollution to illustrate the extent to which they change simultaneously (Tapio, 2005 ; Kaifeng Wang et al., 2021 ). The coefficient is computed by:

where E, EP, and Y represent the Tapio decoupling elasticity coefficient, environmental pollution level, and economic development level of country i in period t, respectively. ∆ indicates the change between periods t and t−1.

Different decoupling elasticities correspond to different relationships between economic development and environmental pollution. The larger the elasticity coefficient, the closer the connection between economic development and environmental pollution. Tapio ( 2005 ) divides the decoupling level into eight states according to the direction of change of ΔY and the size of the decoupling elasticity (as shown in Table 2 ). Economic growth (ΔY > 0) and economic recession (ΔY < 0) each correspond to four states, with decoupling elasticities of 0, 0.8, and 1.2 as the critical values.

EKC describes the nonlinear relationship between environmental pollution (EP) and economic development (Y) measured by GDP per capita. EKC can also be explained from the perspective of Tapio decoupling states (Kaifeng Wang et al., 2021 ). Specifically, before the inflection point in the inverted U-shaped EKC shown in Fig. 1 , as Y increases, the elasticity gradually diminishes from more than 1.2 to less than 0.8. It goes through the stages of expansion negative decoupling, expansion connection, and relative decoupling in sequence. However, when Y exceeds the inflection point y1, the elasticity will remain negative and maintain absolute decoupling. There is an additional inflection point y2 in N-shaped EKC as shown in the right graph of Fig. 1 . The situation before y2 is consistent with the inverted U-shaped EKC; however, after y2, economic growth “reconnects” with pollution after absolute decoupling. It can be seen that the decoupling coefficient changes to positive again and gradually increases and finally surpasses 1.2. Therefore, the N-shaped EKC ultimately includes seven decoupling stages, and these stages show a symmetrical change pattern. Certainly, as the economy recesses along the EKC curve, it may go through four other stages of decoupling (States 5–8), which, however, are beyond the scope of the EKC theory.

Tapio decoupling states corresponding to EKCs.

Two-dimensional decoupling model

The effectiveness of the change patterns described in the decoupling states discussed above depends on the effectiveness of EKC. However, EKC fails to function sometimes, especially when considering the possibility that some countries may be ahead or lagging in their development process (Kaifeng Wang et al., 2021 ). For example, although some countries have lower per capita GDP, economic development and environmental pollution can maintain an absolute decoupling for a long time. On the other hand, countries belonging to different development stages also probably have the same Tapio decoupling elasticity, see N-shaped EKC. Therefore, the relationship between economic development and environmental pollution of a country cannot be accurately judged solely by the absolute level of economic development or the state of decoupling.

A two-dimensional (2D) decoupling model is used to address the above issues (Kaifeng Wang et al., 2021 ). Based on the inflection point of EKC and the Tapio decoupling elasticity coefficient, the model constructs multiple quadrants in the two-dimensional plane coordinate system of Y-E (economic development-decoupling elasticity) to accommodate different economic development-environmental pollution relationships. Based on the shape of the EKC, the two-dimensional decoupling model can contain one inflection point (U-shape, inverted U-shape) or two inflection points (N-shape, inverted N-shape). Taking the 2D decoupling model based on N-shaped EKC as an example, two vertical lines y = y1 and y = y2 divide the horizontal axis into three areas. Different regions represent different stages of N-shaped EKC. In each area, according to E = 0 and E = 0.8, the vertical axis is further divided into three areas, corresponding to absolute decoupling, relative decoupling, and non-decoupling (including Expansive coupling and Expanding negative decoupling). Therefore, the entire plane is divided into nine areas, corresponding to nine 2D decoupling states (see Fig. 2 ).

Two-dimensional decoupling model based on N-shaped EKC.

This model relaxes the strict assumption in the EKC model that Y and EP are directly corresponding, and also overcomes the problem of Tapio decoupling that only considers the relative change relationship between the two. Through this division, we can clearly and accurately classify countries with similar economic and environmental development conditions into the same category to conduct systematic and targeted research. Considering that many countries lack data for individual years, to reduce statistical errors and random errors, we use the ten-year average as one period to classify the economic development and carbon emission levels of each country. Specifically, we divide countries into different EKC stages according to the average GDP per capita from 2010 to 2020. We also calculate the decoupling status using the average GDP per capita and carbon emissions per capita from 2000 to 2010 and 2010 to 2020 as data in period t-1 and period t, respectively.

Empirical model settings

To test H1 and investigate the shape of the global EKC curve, we first establish the following regression model based on the classic EKC empirical model corresponding to Eq. 10 :

Among them, PCE and PGDP are dependent variables and independent variables, respectively, representing per capita carbon emissions and per capita GDP.

We further add control variables to the above model to test H2 and H3:

where X represents the vector of control variables, including geopolitical risk, composite risk, institutional quality, ICT, digital economy, AI development, natural resource rents, energy transition, trade openness, population aging, income inequality, and food security. Based on the estimation results of this equation, we can determine the shape of the EKC based on the results of β1, β2, and β3 and discuss the influence of additional variables based on the results of γ .

Finally, to examine H4 and explore the determinants of carbon emissions in country groups at different two-dimensional decoupling stages, we first divided the full sample of 214 countries into several groups according to the two-dfimensional decoupling model, and carried out the following regressions in each group:

with X it consistent with the equation.

Variable definition and data description

According to the requirements of the methods and models discussed previously, this article uses per capita carbon emissions as the explained variable, per capita GDP as the explanatory variable, and 12 variables covering institutional risks, digital technology, resource utilization, and social issues as control variables. Based on data availability, the research scope of this article is determined to be 214 countries from 1960 to 2020. The definitions, measurement units, and data sources of all variables are summarized in Table S1 . It can be seen that the explanatory variables and the explained variables are all derived from the WDI database developed by the World Bank (Worldbank, 2023a ).

Control variables include 12 variables. Institutional risk includes three variables, among which geopolitical risk comes from the calculation work of (Caldara and Iacoviello, 2022 ). This database is updated every day and can comprehensively and effectively reflect the risk level of geopolitical threats and actions of various countries. Composite risk is derived from the ICGR published by PRS Group since 1980 (PRS, 2023 ), which takes into account domestic political, economic, and financial risks. Institutional quality is calculated by the average of six indices of the World Governance Index(WGI) released by the World Bank (Worldbank, 2023b ), which reflects the integrity and effectiveness of each country’s institutions.

Technology variables include ICT, digital economy, and artificial intelligence (AI). Following (Shufang Zhao et al., 2022 ), ICT is represented by mobile cellular subscriptions. The digital economy is calculated by the entropy weight method and represents the development level of a country’s digital technology and digital industry. The calculation process uses 12 secondary indicators from the International Telecommunication Union (ITU, 2023 ) and WDI, covering four aspects: infrastructure, social support, social effects, and economic effects, as shown in Table S2 . AI is represented by the number of industrial robots operational stocks which comes from the World Robotics Report released by the International Federation of Robotics (IFR, 2023 ).

Variables related to resource use include total natural resource rents and the level of energy transition represented by the share of renewable energy consumption. Finally, among the social variables, population aging is expressed as the proportion of the population over 65 years old. Food safety is derived from the food production index from WDI database. Trade openness is measured by a country’s trade volume as a share of its GDP. The GINI coefficient represents the level of income inequality. These variables are all from WDI. The descriptive statistics of the variables are shown in Table S3 .

Empirical results

Preliminary test, handling the multicollinearity problem.

First, we show the correlation coefficient matrix and VIF (see Fig. S1 ). The highest correlation coefficient reached 0.87, exceeding the safety line of 0.85, representing the need to analyze VIF to rule out the multicollinearity problem. The results show that the VIF of all variables is less than 10, which means that the multicollinearity problem between variables is not serious. Considering that the VIF of institutional quality is closer to 10 compared to other variables, we choose to add it in the last step of the stepwise regression to avoid potential interference.

Handling the pseudo-regression problem

To avoid pseudo-regression, we need to test the stationarity of the disturbance term, that is, test the cointegration relationship between variables. The cointegration test requires that each variable sequence is first-order integrated, and a unit root test needs to be performed on each variable first. Therefore, fisher-adf and fisher-pp tests are used to test the data stationarity. Both results of all variables in the two test methods are stationary in first order (see Table S4 ). Finally, the Kao test is used to test the cointegration (see Table S4 ). The results of Kao ADF statistics reject the null hypothesis that there is no cointegration relationship between panel variables at the 1% level. This indicates that the long-term stable relationship between variables exists and the pseudo-regression problem does not hold.

Selecting the estimation model

Fixed effects models (FE), two-way fixed effects models, and dynamic (system-/difference-) GMM models are among the most popular models in previous panel studies. The fixed effects model is one of the most commonly used models in research. It eliminates endogeneity problems caused by unobservable factors related to individuals through within-group estimators. Both two-way fixed effects and dynamic GMM improve FE by adding explanatory variables to the model: the former adds a vector of time dummy variables, while the latter includes the lagged term of the explained variable. However, this study has selected additional variables from various perspectives to accurately track the changes in the carbon emissions curve and open the black box of the determinants of carbon emissions, making the addition of more variables to the model redundant and potentially harmful (such as exacerbating multicollinearity problems). Therefore, we finally choose to adopt the basic FE models in our empirical process. Other models are used as robustness check methods.

Empirical results of global panel

Confirm n-shaped ekc without additional variables.

To test H1, we perform estimation and robustness testing on the regression equation without additional variables Eq. ( 16 ). The results are shown in Table 3 . First, the coefficient of the cubic term of per capita GDP (PGDP3) in Model 1 is significantly greater than 0 at the 1% level, indicating a cubic EKC. In addition, the coefficient of the quadratic term is significantly negative and the coefficient of the linear term is significantly positive. It shows that as PGDP gradually increases, per capita carbon emissions conform to the N-shaped trend. Including the above regression coefficients into Eq. ( 13 ), it can be calculated that the EKC inflection points without considering control variables are 34.44 and 77.54 (thousand US dollars/person), respectively. We can think of these as two thresholds. Initially, the increase in per capita GDP leads to an increase in per capita carbon emissions and reaches a peak when per capita GDP reaches 34.44. However, when PGDP gradually increases and exceeds this value, per capita carbon emissions will reverse and change from rising to falling. Finally, when per capita GDP further increased and reached 77.54, the trend of per capita carbon emissions changed upward again.

We also utilize four methods to check the robustness of N-shaped EKC, as shown in Model R1-R4 of Table 3 . First, change the environmental pollution indicator from per capita carbon emissions to total carbon emissions. Second, the fixed effects estimation model is replaced by an individual-time-two-way fixed effects model. Third, a random effects model will be used. Fourth, use the system GMM. First, the Sargan test p -value and AR(2) in Model R4 are greater than 0.1, indicating that the dynamic model estimation results are valid. All four models show that the coefficients of PGDP and its cubic term are significantly greater than 0, while the coeffecients of the quadratic term is significantly less than 0. In addition, the results of y1 and y2 show that there are two inflection points in the EKC corresponding to the four models, and they all fall within the sample interval. In conclusion, the global N-shaped EKC exists when control variables are not considered and H1 is confirmed. Farooq et al. ( 2022 ) took 185 countries around the world as a sample, confirmed the positive linear term and negative quadratic term of PGDP, and obtained the result of the inverted U-shaped EKC. However, our results show that the positive cubic term also exists robustly, so this paper resets the shape of the global EKC to an N-shape. From another perspective, this result generalizes the N-shaped EKC confirmed in Numan et al. ( 2022 ) in 85 countries to 191 countries.

Determine the EKC shape and inflection points under additional variables

Table 4 shows the results of adding control variables for testing H2 and H3. To reduce the impact of differences in sample sizes of different variables and alleviate the problem of multicollinearity, we adopt stepwise regressions Footnote 1 to display the results of adding covariates. Among them, Model 1 only uses PGDP and its quadratic and cubic terms as explanatory variables, and the R square is 0.122. Model 2-Model 6 gradually added 12 control variables, and the R square increased accordingly to 0.718 in Model 6. This shows that the addition of explanatory variables can accurately explain changes in carbon emissions. As control variables are continuously added to Models 2–6, the cubic term of PGDP is always significantly positive at the 1% statistical level. In addition, the sign and statistical level of PGDP and its quadratic term are also completely consistent with Model 1. This means that after adding control variables, the shape of the EKC is still a standard N-shape and H2 is confirmed.

After calculation, the two inflection points of the N-shaped EKC in Model 6 are 45.08 and 73.44 (thousand dollars per person), respectively. This result is different from the EKC inflection point in Model 1. We juxtapose the cubic equation curves corresponding to the estimated coefficients in Fig. 3 to discuss the influence of the control variables on the shape of the EKC. Figure 3 shows that after taking control variables into account, not only did the economic growth space between the two inflection points shrink from 43.10 thousand US dollars to 28.35 thousand US dollars, but also the reduction of per capita carbon emissions between the inflection points is compressed from about 2.70 metric tons to about 1.05 metric tons. This suggests that the EKC appears to be bent upwards when trying to account for other explanatory variables. As a result, the only declining part of the N-shaped carbon emissions curve has shrunk in both “duration” and “descending space,” revealing a more severe global climate mitigation situation.

Note: This graph is for comparison of curve shapes only and the position of the curve along the vertical axis does not represent the actual intercept.

Discuss the role of additional variables on carbon emissions

Institutional, technology, resources, and social factors not only affect the shape of EKC but also have an important impact on carbon emissions. This leads to the discussion of H3. Model 2 adds four control variables based on Model 1. Taking the 5% significance level as the standard, the impact coefficients of ICT and food security are positive, and the impact coefficients of digital economy and population aging are negative. We examine the robustness of the effects of these variables in conjunction with other results of the stepwise regressions. Comparing the estimation results of Model 2 with the subsequent four models (Model 3–6), it shows that the above results can be confirmed by at least two models. Specifically, the results of ICT are significantly positive in both Model 3 and Model 4; the positive coefficient of food security exists in Models 3–5; the negative coefficients of digital economy and population aging hold in all models. This shows that the estimation results of ICT, food security, digital economy, and population aging in Model 2 are reliable as decisions.

The estimators of the three factors added in Model 3 are all significant. Among them, the impact coefficient of natural resource rent is positive, and the impact of trade openness and energy transition is negative. However, results from subsequent models support the effects of energy transition rather than that of natural resource rents and trade openness. All models show that the impact coefficient of energy transition is significantly negative, while the effects of natural resource rents and trade openness are insignificant in two of the three subsequent models. Model 4 adds comprehensive risk. Combined with the estimation results of Model 4–6, the impact of comprehensive risk on carbon emissions is determined to be negative, which indicates that the higher the overall country risk level, the higher the corresponding carbon emissions. By comparing the results of Model 5 and Model 6, we determined that the effects of artificial intelligence and income inequality are not significant. Finally, the results of Model 6 show that geopolitical risks intensify carbon emissions, while improvements in institutional quality can reduce carbon emissions. These results show that H3 is partially confirmed.

We summarize the impact of each variable on carbon emissions in Table 5 to make further discussion. From the perspective of institutional risk factors, the rise in geopolitical risks and national comprehensive risks has significantly promoted carbon emissions. This is consistent with some recent research results on geopolitical risks (Anser et al., 2021 ) and political risks (Hassan et al., 2022 ), but is inconsistent with the research results on financial risks (Jun Zhao et al., 2021a ). Risks may cause panic and short-sightedness among governments and investors, which are often linked to irresponsible production models and misuse of fossil fuels (Vakulchuk et al., 2020 ; Zuoxiang Zhao et al., 2023 ). The reducing effect of institutional quality on global carbon emissions is consistent with the regulatory effect hypothesis and consistent with previous research for 3 Asian countries (Salman et al., 2019 ) and 30 Sub-Saharan African countries (Karim et al., 2022 ). Therefore, a stable international and domestic development environment and effective systems play an irreplaceable role in reducing carbon emissions and achieving global climate mitigation.

The results of technological factors are sobering. The digital economy can reduce carbon emissions, while the single ICT industry promotes carbon emissions. The two variables are conceptually similar, but why could they have completely opposite impacts on global carbon emissions? Despite a lack of discussion and explanation of this phenomenon in the literature, a recent study (Jinning Zhang et al., 2022a ) offers a good perspective. They discussed the impact of different dimensions of the digital economy on low-carbon development and found that industrial digitalization has the most significant impact on low-carbon development, followed by digital industrialization, with the carrier playing the smallest role. In other words, it is the integration of digital technology and the economy, that promotes emission reduction, not just the development of ICT technology and equipment (Jinning Zhang et al., 2022a ). The impact of AI is not significant, possibly because the effects of AI in different countries cancel each other out. Luan et al. ( 2022 ) found that industrial robots promoted air pollution in 74 countries while (Zhong et al., 2023 ) claimed that AI reduced carbon emissions in 66 countries. In summary, inconsistent with the views of technological rationalists, our results suggest that smart technological developments alone may not necessarily improve the environment. Alternatively, the rational use of technology to benefit the economy and society may be an efficient approach to achieving carbon emission reductions.

The results on resource utilization emphasize the urgency of clean energy development and deployment. On the one hand, the results confirm the significant reduction effect of the energy transition on carbon emissions. This shows that replacing existing energy infrastructure with one that relies on renewable energy can indeed significantly reduce carbon emissions per capita, which does not exceed the research conclusions of other scholars (Dogan and Seker, 2016 ; Inglesi-Lotz and Dogan, 2018 ). On the other hand, the impact of natural resource rents is uncertain, consistent with the controversy in the literature. The results also highlight the potential impact of social factors on carbon emissions. Population aging is a negative contributor to carbon emissions. This corresponds with (Zhou et al., 2023 ) while contradicting the findings of (Balsalobre-Lorente et al., 2021 ) and (Fan et al., 2021 ). Food security is a positive contributor to carbon emissions, supported by (Naseem et al., 2020 ). It shows that more attention needs to be paid to deforestation, land use, and high-carbon food consumption issues while ensuring food safety in global agricultural development. The effects of trade openness and income inequality are not significant.

Group 214 countries through two-dimensional decoupling model

Before testing H4, we first classify countries according to the development stage and emission reduction status to solve the heterogeneity problem that exists when testing the EKC hypothesis using the traditional panel method. Therefore, this paper first uses a two-dimensional decoupling model to achieve an accurate classification of countries based on their economic development levels and the carbon emissions decoupling states. The calculation and grouping process can be broken down into three steps: (1) Group countries according to the average per capita GDP in the past decade and the EKC inflection point. (2) Group countries according to Tapio decoupling elasticity between the past two decades and the critical values (0 and 0.8). (3) Match the results of the above two steps with the nine two-dimensional decoupling stages shown in Fig. 2 , and obtain the final grouping based on the two-dimensional decoupling model. The results are shown in Fig. 4 .

a Classification results based on EKC turning points. b Classfication results based on Tapio decoupling coefficients. c Final classification results of two-dimensional decoupling state.

According to Fig. 4a , the 214 sample countries are divided into three groups based on the N-shaped EKC inflection points estimated in the baseline regression results. The first phase of EKC is still the main theme of global development. More than 90% of countries are in this stage, with their average per capita GDP between 2010 and 2020 being less than US$45,080. There are 11 countries in the second stage. Their ten-year average per capita GDP exceeds 45,080 but is less than 73,438. These countries have relatively high levels of economic development, mainly including the United States, Australia, and some European countries. According to the EKC hypothesis, they have the best hope of achieving economic growth while reducing carbon emissions. Only 9 countries have a ten-year average per capita GDP higher than 73,439 and are in the third stage, almost all of which are located in Europe. These countries have reached extremely high levels of economic development, but they may be directly responsible for the second upslope of N-type EKC and the rebound in pollution levels.

Figure 4b shows that the number of countries in different decoupling states is relatively even and that there are regional aggregation characteristics. The number of countries with non-decoupling, weak decoupling, and strong decoupling is 57, 47, and 55, respectively, and 34 countries are suffering an economic recession. The results of non-decoupling countries are concerning. The results show that these countries with decoupling elasticity still higher than 0.8 in the past 20 years are mainly located in sub-Saharan Africa, Latin America, and the Middle East. These countries are characterized by economic development that relies heavily on the extraction or utilization of natural resources (Do, 2021 ). How to help them reduce carbon emissions is an important challenge for global climate mitigation. It can be seen that most of the weak decoupling countries are located in non-Middle East Asia. Many of these countries, such as India and China, have achieved a degree of industrialization and urbanization, thereby gradually decoupling their economies from carbon emissions. Strong decoupling countries are mainly distributed in North America, Europe, and Oceania, and have a high degree of overlap with the countries in the second and third stages of the EKC. Finally, some African countries, including Libya and the Central African Republic, are facing economic recession. Stabilizing the economy is the top priority for these countries.

Finally, we classify countries based on the two-dimensional decoupling model. The number of countries in each stage is shown in Fig. 4c . First, the results show that all countries in the sample whose GDP per capita is higher than the first inflection point of the EKC have achieved strong decoupling. In other words, among the six two-dimensional decoupling stages corresponding to the second and third stages of EKC (Stage 4–9), 9 countries are in Stage 6, and 3 countries are in Stage 7 Footnote 2 with no country in other stages. This means that from the perspective of the past 20 years, those countries with relatively advanced economic development have achieved a relatively ideal decoupling of economic development and carbon emissions. Among the three two-dimensional decoupling states (Stage 1–3) belonging to the first stage of EKC, there are 43, 47, and 57 countries in strong decoupling, weak decoupling, and non-decoupling, respectively, which is also not entirely consistent with the EKC theory. In the EKC theory, the economic development of countries whose per capita GDP is lower than the first turning point will be accompanied by environmental deterioration, but two-dimensional decoupling analysis shows that the relationship between economic development and environmental pollution in countries at this stage also has different conditions. In summary, we successfully classified countries of the same type through the two-dimensional decoupling model, so that we can conduct specific research on the determinants of carbon emissions in each country group.

Heterogeneous effects of all variables on carbon emissions

Based on the six divided country groups, we perform panel unit root tests and cointegration tests for each sample. The results show that at the 10% significance level, in each sample, the variables are first-order stationary and cointegrated (see Table S6 ). To discuss the heterogeneous impact of economic, institutional, technological, resource, and social factors on carbon emissions and examine H4, the regression results of six country groups are shown in Table 6 .

First, there is obvious heterogeneity in the impact of per capita GDP on carbon emissions. In the first stage of the EKC, according to the estimation results of Model 7 and Model 9, economic development in both non-decoupling and weak-decoupling countries is accompanied by much carbon emissions. In non-decoupling and weakly decoupling countries, for every US$1 increase in per capita GDP, per capita carbon emissions increase by an average of 0.2658 metric tons and 0.3806 metric tons, respectively. It may be puzzling at first that the impact coefficient of economic development on carbon emissions is more pronounced in weak decoupling countries than in non-decoupling countries. The reason can be that the economic development models of these countries, including most Asian developing countries, have been optimized to a certain extent in the past two decades, but their average environmental costs throughout the development process are still high (Rongrong Li et al., 2021 ). These countries need to make full use of the results of past economic development and strive to transform from weak decoupling to strong decoupling (Hanif et al., 2019 ). Model 9 shows that in the sample of strong decoupling in the first stage of EKC, the impact of per capita GDP on per capita carbon emissions is not significant. This is consistent with the decoupling theory, suggesting that they are not developing directly at a high ecological cost. Model 10 shows that in countries whose economic development has entered the second stage of EKC, for every US$1 increase in GDP per capita, carbon emissions per capita increase by 0.0855 metric tons. Considering the relatively small size of the coefficient, it can be considered that N-shaped EKC is supported, that is, economic development accompanied by less environmental pollution in the second phase of EKC. However, Model 11 shows that when economic development enters the third stage of EKC, economic growth and carbon emissions are reconnected. Compared to the second phase of EKC, carbon emissions per capita for every $1 increase in GDP per capita will double from 0.0855 metric tons to 0.1721 metric tons. This result confirmed the technical obsolescence effect (Jahanger et al., 2023 ). Finally, the carbon emission costs of economic development in countries that have faced economic recession in the past 20 years are also extremely high. These countries urgently need to stabilize their economies at controllable environmental costs.

From the perspective of institutional risk, the impact of geopolitical risk is not significant in most samples, indicating that geopolitical risk has not yet significantly affected carbon emissions in most countries around the world. However, in the non-decoupling sample in the first phase of the EKC, geopolitical risk significantly reduces carbon emissions. This may be related to conflicts and crises caused by the rich natural resources of these countries (Do, 2021 ). The negative relationship is supported by (Weijun Zhao et al., 2021b ) who asserted that geopolitical risks may inhibit investment, trade, and energy consumption to reduce carbon emissions. Each unit increase in the composite risk index can reduce per capita carbon emissions by 0.0123 metric tons and 0.0067 metric tons, respectively, in countries in Stage 2 and Stage 1 of two-dimensional decoupling, which is consistent with the full sample research result. The results for institutional quality are rich in heterogeneity. Its effect of reducing carbon emissions is only reflected in Stage 3 countries, and these non-decoupling countries need to strengthen the construction of the national institution. However, in Stage 1, Stage 6, and Stage 10, institutional quality increases carbon emissions. As (Le and Ozturk, 2020 ) pointed out, cumbersome regulations and bureaucracy may in turn delay the implementation of environmental protection actions.

It is interesting to analyze the impact of digital technology variables. The conclusion that DE, rather than ICT, can significantly reduce carbon emissions and achieve low-carbon development is true in Stage 1, Stage 6, and Stage 10, indicating that most countries should focus on the digitalization of industry and economy. However, this does not always hold, because in Stage 3 it is exactly the opposite: ICT reduces carbon emissions, while the digital economy promotes carbon emissions. Although these countries are in a weak decoupling state, the impact coefficient of PGDP on carbon emissions is the highest among all samples. The combination of a rough economic development model and digital technology may lead to excessive carbon emissions (Feng Dong et al., 2022a ). Therefore, these countries need to focus on the carbon footprint of the digital economic development model in the process of developing ICT. AI reduces emissions in Stage 3 countries while increasing emissions in Stage 2 countries, which is not significant in other groups, indicating that AI needs to be used in the economy and society with more caution.

In terms of resource use, the energy transition significantly reduced carbon emissions in all six samples according to Table 6 . However, this effect was most pronounced in countries in EKC stage 3, followed by EKC stage 2. Every 1% increase in the proportion of renewable energy will lead to a decrease in per capita carbon emissions of 0.4661 metric tons in the former and 0.1096 metric tons in the latter. The emission reduction effects of countries in the first stage of EKC are relatively weak, diminishing in countries with strong decoupling, weak decoupling, and non-decoupling. The distribution pattern of the influence coefficient shows that increased levels of economic development and decoupling lead to better energy transition outcomes. This result is supported by (Nguyen and Kakinaka, 2019 ), who conclude that in the low-income group, renewable energy development inhibits economic development and promotes carbon emissions; while in the high-income group, the opposite holds. On the contrary, natural resource rent promotes carbon emissions in Stage 2 and Stage 6 countries and has no significant impact on other countries.

Finally, among the four social factors, the impact of population aging is relatively consistent. It mainly affects Stage 2 and Stage 1 countries and brings about reductions in carbon emissions. The result of OPEN does not support the existence of the PHH because there is no obvious evidence that countries with better environmental regulations have reduced carbon emissions through openness. Honestly, trade openness brings more carbon emissions in most samples (Stage 1, 7, 10), and reduces carbon emissions only in Stage 2 samples. This means that frequent exchanges of goods and services may increase the logistics and transportation burdens of many countries (Rongrong Li et al., 2021 ). The effects of food security and income inequality are also heterogeneous and may have opposite effects in different countries. Specifically, food security reduces carbon emissions in Stage 1 countries but increases carbon emissions in Stage 2 and Stage 10 countries. Income inequality is a contributor to carbon emissions for Stage 1 and Stage 6 countries, but the opposite is true for Stage 2 countries. The above results highlight between-group heterogeneity in the direction and magnitude of carbon emissions effects for all variables, which establishes H4.

Differential carbon reduction strategies of countries in different two-dimensional decoupling stages

A preliminary discussion of the results of Table 6 shows that not only does the relationship between per capita GDP and carbon emissions exhibit nonlinear and heterogeneous rules, but the impact of the additional variables on carbon emissions also significantly varies in different types of countries. Therefore, countries in different two-dimensional decoupling states can adopt differentiated institutional, technological, resource, and social development strategies to achieve carbon emission reductions. We summarize the positive and negative drivers of carbon emissions in Fig. 5 according to the two-dimensional decoupling stage.

Note: The light red box on the left means positive drivers and the light blue box on the right means negative drivers. The red, dark blue, orange, and light blue ovals mean institutional risks, digital technology, natural resource utilization, and social factors, respectively.

Stage 1 countries belong to the strong decoupling state of the first stage of the EKC. This is the only group of countries where the impact of economic growth on carbon emissions is not significant, meaning that the past development models of these countries did not come at the expense of environmental quality. However, the results in Fig. 5 indicate that in addition to economic development, many other factors significantly affect environmental quality and therefore require special attention. First, in terms of institutions, institutional quality promotes carbon emissions, while the comprehensive country risk index reduces carbon emissions. The policy implication is that these countries should improve the construction and implementation of environmental regulations while striving to stabilize the domestic economic, political, and financial environment (Karim et al., 2022 ). In terms of digital technology, ICT promotes carbon emissions, while digital technology reduces carbon emissions. It shows that these countries should pay attention to the carbon footprint issue of the ICT industry and encourage the penetration of ICT into other economic sectors to form a broader and deeper digital economic industry. In terms of resource utilization, the emission reduction effect of the energy transition is significant. Considering that the magnitude of the effect is not ideal, countries currently in Stage 1 should prioritize economic development and gradually deploy the production and utilization of clean energy in conjunction with economic development. In terms of social management, income inequality, and trade openness bring more carbon emissions, while population aging and food security reduce carbon emissions. It seems that, on the one hand, these countries should introduce fiscal and agricultural production policies to promote fair distribution of wealth and improve food production security (Akbar et al., 2019 ). On the other hand, they need to introduce more green products and technologies in the process of international trade or investment promotion (Qiang Wang et al., 2023b ).

For Stage 2 countries, they are in a state of relative decoupling, between non-decoupling and complete decoupling. It shows that these countries have certain methods to reduce pollution in development, but there is still a lot of room for emission reduction. These countries need to comprehensively understand the positive and negative factors affecting carbon emissions to achieve a transformation from relative decoupling to absolute decoupling. In particular, the results of this study indicate that AI, natural resource rents, and food security are contributing to per capita carbon emissions. Therefore, strengthening environmental regulations and introducing normative and environmentally friendly technologies in artificial intelligence technology, natural resource extraction, and food production processes may be the key to achieving strong decoupling. Disincentives to carbon emissions suggest that maintaining national stability, energy transition and trade openness are also important. Countries in Stage 3 are non-decoupling. Their economic development is accompanied by high carbon emissions, so the most important thing is to solve the problem of rough development. In addition, the control variables still provide emission reduction options from institutional, technical, and natural resource perspectives. For them, the positive driver of carbon emissions is primarily the digital economy. In comparison, the development of ICT and AI technologies cannot curb carbon emissions. This further emphasizes the need to decouple economic development and carbon emissions. In addition, rising levels of geopolitical risks, institutional quality, and energy transition can also reduce carbon emissions.

Stage 6 countries have a relatively high level of development and have entered the second stage of EKC and achieved strong decoupling. But they can also pursue better environmental outcomes. Based on the negative drivers of carbon emissions, these countries should focus on building environmental institutions, limiting the use of traditional natural resources, and optimizing income structures. In line with the positive drivers of carbon emissions, further development of the digital economy and energy transition has an important role in improving environmental quality.

For Stage 7 countries, their economic development level is higher than that of Stage 6, but their economic development and carbon emissions have been reconnected. Due to the limited sample, the results of most factors are not significant. However, our results still point to two emission reduction ideas. On the one hand, the effects of energy transformation in these countries are far better than in other countries, and clean energy should be further developed and deployed in these countries; on the other hand, trade openness is positively related to carbon emissions, and they should more strictly scrutinize imported goods or foreign investment. While Stage 10 countries are facing economic recession, economic development has a high coefficient of impact on carbon emissions. These countries not only need to stabilize their economy but are also supposed to find ways to reduce environmental costs. The results indicate that they need to establish more robust environmental regulations and address carbon emissions issues in ICT, food production, and trade openness. In addition, the digital economy and energy transition can be seen as further developments in green industries.

Concluding remarks

Early research provided a solid theoretical foundation and empirical evidence for the inverted U-shaped EKC. However, with the increasing urgency of mitigating climate change, we have observed that many developed countries are struggling with environmental degradation, high energy consumption, and carbon emissions. This forces us to start considering N-shaped global EKC. In addition, the impact of risks, digital technology, energy transition, and various social factors on social activities, economic development, and energy use are also proven to be significant contributors to carbon emissions. Against this background, we included as many countries as possible in the study, re-examined the shape of the EKC in the global panel, and incorporated 12 traditional and novel institutional, technological, resource, and social factors as additional variables into the EKC equation. In addition, we developed a two-dimensional Tapio decoupling model based on the inflection point in N-shaped EKC to achieve group discussion of sample countries. Finally, based on grouping, we discussed the heterogeneous impact of all variables and the differential emission reduction solutions of each group between groups and obtained a series of conclusions.

First, in the global panel, N-shaped EKC exists robustly regardless of whether additional variables are taken into account. In other words, the cubic and linear terms of GDP per capita are significantly positive, while the quadratic term is significantly negative. Among them, when additional variables are not considered, the N-shape has passed four robustness tests. When considering additional variables, this N-shaped EKC also holds at each step of the stepwise regression. Thus H1 and H2 are completely confirmed. After determining the existence of the N-shape, we finally calculated the inflection points of EKC including all control variables, which were 45.08 and 73.44 (thousand dollars/person), respectively. By comparing the final N-shaped EKC curve with the N-shaped curve without adding control variables, we found that both the “duration” and “dropping space” of the only declining part in the curve have shrunk, which makes global climate mitigation even more severe. Through mutual verification of each step of the stepwise regression, we finally determined the direction of influence of all additional variables. Among them, geopolitical risks, ICT, and food security were confirmed to have a positive impact on per capita carbon emissions. However, comprehensive risks, institutional quality, digital economy, energy transition, and population aging were found to have a robust negative impact. Artificial intelligence, natural resource rents, trade openness, and income inequality have insignificant effects on carbon emissions. These results partially confirmed H3.

In the group study, we first used the per capita GDP of 45.08 and 73.44 (thousand US dollars per person) obtained in the EKC as the basis for classification, and initially divided all countries into three EKC stages. As a result, 194 countries belong to the first stage, 11 countries belong to the second stage, and 9 countries belong to the third stage. Secondly, we calculated the Tapio decoupling elasticity coefficient between the past two decades for all countries and further divided the countries into three states using 0 and 0.8 as thresholds. The results show that there are 57, 47, 55, and 34 countries in non-decoupling, weak decoupling, strong decoupling, and recession, respectively. Based on this, we established two-dimensional decoupling coordinates and combined the three states of each of the above two dimensions with each other to obtain nine two-dimensional decoupling states. We divide the sample into five groups by matching the non-recession sample results to the nine decoupling conditions. Among them, there are 43, 47, 57, 9, and 3 countries in Stages 1, 2, 3, 6, and 7, respectively. Adding the declining countries (Stage 10) we get 6 panels. We also conduct multiple linear regressions in these panels and discuss the heterogeneous effects of economic development and 12 additional variables on carbon emissions. The results completely validate H4 and show that the effects of most variables vary according to country conditions. The most robust variable is the energy transition, which shows a significant carbon reduction effect in all groupings. However, the magnitude of the impact of energy transition is also heterogeneous. In countries with higher levels of economic development and decoupling, the effects of energy transition are stronger. Finally, we discuss the differential emission reduction plans of countries in different two-dimensional decoupling stages based on the direction of influence of variables.

The above findings have valuable policy implications. First of all, global results show that the N-shaped EKC is more severe than the inverted U-shaped EKC. The empirical validation of technological obsolescence underscores the imperative for developed nations worldwide to bolster their focus on internal environmental regulations and elevate levels of innovation in clean technologies. Furthermore, the inclusion of additional variables in our analysis underscores the pivotal role played by a stable international and domestic developmental milieu alongside effective institutional frameworks in curbing carbon emissions and advancing global climate mitigation efforts. The uncritical pursuit of purely digital or smart technological advancements does not inherently translate into environmental amelioration. Conversely, a judicious harnessing of technology for the betterment of both the economy and society presents a promising avenue for achieving reductions in carbon emissions. Moreover, our findings on resource utilization underscore the pressing need for accelerated development and adoption of clean energy sources. Addressing social determinants, particular attention is warranted towards optimizing land usage and curtailing the prevalence of high-carbon footprint foods within agricultural production. Finally, specific emission reduction ideas for each country have also been discussed in depth in the section “Differential carbon reduction strategies of countries in different two-dimensional decoupling stages”.

We need to point out the limitations of this article to provide ideas for improvement. First, due to non-uniform data sources, our empirical process uses an unbalanced panel. In some steps, the effective sample does not cover all 214 countries. This may ignore information from certain countries. Second, we include some important or novel additional variables to comprehensively discuss the determinants of carbon emissions other than growth. However, we cannot consider all factors. Second, we developed Tapio two-dimensional decoupling based on N-shape to divide countries into 9 stages to fully consider the heterogeneity between countries. However, this classification also fails to fully handle heterogeneity. Our overall study is not a substitute for single-country and sub-national case studies. Therefore, subsequent scholars have conducted more in-depth research from the perspective of improving the unity of data sources, providing more novel and critical additional variables, and conducting more refined case studies.

Data availability

The datasets publicly available should be through https://doi.org/10.7910/DVN/0I1EYG .

When determining the order in which variables are added, we consider the data characteristics of the variables, particularly the data volume, so that as many countries as possible are included in each step.

The number of countries belonging to the second and third stages of EKC in two-dimensional decoupling in Fig. 4c is smaller than that in Fig. 4a . This is because some countries only have per capita GDP data but do not have enough carbon emission data in the past two decades.

Akbar A, Ali I, Samiullah NU, Khan SA, Rehman Z, Rehman S (2019) Functional, antioxidant, antimicrobial potential and food safety applications of Curcuma longa and Cuminum cyminum. Pak J Bot 51(3):1129–1135

Article   CAS   Google Scholar  

Allard A, Takman J, Uddin GS, Ahmed A (2018) The N-shaped environmental Kuznets curve: an empirical evaluation using a panel quantile regression approach. Environ Sci Pollut Res 25:5848–5861

Álvarez-Herránz A, Balsalobre D, Cantos JM, Shahbaz M (2017) Energy innovations-GHG emissions nexus: fresh empirical evidence from OECD countries. Energy Policy 101:90–100

Article   ADS   Google Scholar  

Andreoni J, Levinson A (2001) The simple analytics of the environmental Kuznets curve. J Public Econ 80(2):269–286

Article   Google Scholar  

Ang JB (2008) Economic development, pollutant emissions and energy consumption in Malaysia. J Policy Model 30(2):271–278

Anser MK, Syed QR, Apergis N (2021) Does geopolitical risk escalate CO 2 emissions? Evidence from the BRICS countries. Environ Sci Pollut Res 28(35):48011–48021

Apergis N, Ozturk I (2015) Testing environmental Kuznets curve hypothesis in Asian countries. Ecol Indic 52:16–22

Asiedu E (2006) Foreign direct investment in Africa: The role of natural resources, market size, government policy, institutions and political instability. World Econ 29(1):63–77

Balsalobre-Lorente D, Shahbaz M, Roubaud D, Farhani S (2018) How economic growth, renewable electricity and natural resources contribute to CO 2 emissions? Energy policy 113:356–367

Balsalobre-Lorente D, Sinha A, Driha OM, Mubarik MS (2021) Assessing the impacts of ageing and natural resource extraction on carbon emissions: a proposed policy framework for European economies. J Clean Prod 296:126470

Bashir MF (2022) Discovering the evolution of Pollution Haven Hypothesis: a literature review and future research agenda. Environ Sci Pollut Res 29(32):48210–48232

Beckerman W (1992) Economic growth and the environment: Whose growth? Whose environment? World Dev 20(4):481–496

Berkeley (2024) Press Release: 2023 was warmest year since 1850. Berkeley Earth. https://berkeleyearth.org/press-release-2023-was-the-warmest-year-on-recordpress-release/

Brown PT, Hanley H, Mahesh A, Reed C, Strenfel SJ, Davis SJ et al. (2023) Climate warming increases extreme daily wildfire growth risk in California. Nature 621(7980):760–766. https://doi.org/10.1038/s41586-023-06444-3

Article   ADS   CAS   PubMed   Google Scholar  

Caldara D, Iacoviello M (2022) Measuring geopolitical risk. Am Econ Rev 112(4):1194–1225

Carlson CJ, Albery GF, Merow C, Trisos CH, Zipfel CM, Eskew EA et al. (2022) Climate change increases cross-species viral transmission risk. Nature 607(7919):555–562. https://doi.org/10.1038/s41586-022-04788-w

Cavicchioli R, Ripple WJ, Timmis KN, Azam F, Bakken LR, Baylis M et al. (2019) Scientists’ warning to humanity: microorganisms and climate change. Nat Rev Microbiol 17(9):569–586

Article   CAS   PubMed   PubMed Central   Google Scholar  

Charfeddine L, Umlai M (2023) ICT sector, digitization and environmental sustainability: a systematic review of the literature from 2000 to 2022. Renew Sustain Energy Rev 184:113482

Cheng P, Tang H, Lin F, Kong X (2023) Bibliometrics of the nexus between food security and carbon emissions: hotspots and trends. Environ Sci Pollut Res 30(10):25981–25998

Cole MA (2004) Trade, the pollution haven hypothesis and the environmental Kuznets curve: examining the linkages. Ecol Econ 48(1):71–81

Dai M, Sun M, Chen B, Shi L, Jin M, Man Y et al. (2023) Country-specific net-zero strategies of the pulp and paper industry. Nature 626:327–334. https://doi.org/10.1038/s41586-023-06962-0

Article   CAS   PubMed   Google Scholar  

Diaz D, Moore F (2017) Quantifying the economic risks of climate change. Nat Clim Change 7(11):774–782

Dinda S (2004) Environmental Kuznets curve hypothesis: a survey. Ecol Econ 49(4):431–455

Ding T, Li J, Shi X, Li X, Chen Y (2023) Is artificial intelligence associated with carbon emissions reduction? Case of China. Resour Policy 85:103892

Do TK (2021) Resource curse or rentier peace? The impact of natural resource rents on military expenditure. Resour Policy 71:101989

Dogan E, Seker F (2016) Determinants of CO 2 emissions in the European Union: the role of renewable and non-renewable energy. Renew Energy 94:429–439

Dong F, Hu M, Gao Y, Liu Y, Zhu J, Pan Y (2022a) How does digital economy affect carbon emissions? Evidence from global 60 countries. Sci Total Environ 852:158401

Dong F, Li Y, Gao Y, Zhu J, Qin C, Zhang X (2022b) Energy transition and carbon neutrality: exploring the non-linear impact of renewable energy development on carbon emission efficiency in developed countries. Resour Conserv Recycl 177:106002

Dong M, Wang G, Han X (2023) Artificial intelligence, industrial structure optimization, and CO 2 emissions. Environ Sci Pollut Res 30(50):108757–108773

Duan H, Zhou S, Jiang K, Bertram C, Harmsen M, Kriegler E et al. (2021) Assessing China’s efforts to pursue the 1.5 C warming limit. Science 372(6540):378–385

Fakher HA, Ahmed Z, Acheampong AO, Nathaniel SP (2023) Renewable energy, nonrenewable energy, and environmental quality nexus: an investigation of the N-shaped Environmental Kuznets Curve based on six environmental indicators. Energy 263:125660

Fan J, Zhou L, Zhang Y, Shao S, Ma M (2021) How does population aging affect household carbon emissions? Evidence from Chinese urban and rural areas. Energy Econ 100:105356

Fankhauser S, Smith SM, Allen M, Axelsson K, Hale T, Hepburn C et al. (2022) The meaning of net zero and how to get it right. Nat Clim Change 12(1):15–21

Farooq S, Ozturk I, Majeed MT, Akram R (2022) Globalization and CO 2 emissions in the presence of EKC: a global panel data analysis. Gondwana Res 106:367–378

Article   ADS   CAS   Google Scholar  

Fawzy S, Osman AI, Doran J, Rooney DW (2020) Strategies for mitigation of climate change: a review. Environ Chem Lett 18:2069–2094

Gill AR, Viswanathan KK, Hassan S (2018) The Environmental Kuznets Curve (EKC) and the environmental problem of the day. Renew Sustain Energy Rev 81:1636–1642

Grossman GM, Krueger AB (1991) Environmental impacts of a North American free trade agreement. National Bureau of Economic Research Cambridge, MA, USA

Grossman GM, Krueger AB (1995) Economic growth and the environment. Q J Econ 110(2):353–377

Guan Y, Yan J, Shan Y, Zhou Y, Hang Y, Li R et al. (2023) Burden of the global energy price crisis on households. Nat Energy 8(3):304–316

Hanif I, Raza SMF, Gago-de-Santos P, Abbas Q (2019) Fossil fuels, foreign direct investment, and economic growth have triggered CO 2 emissions in emerging Asian economies: some empirical evidence. Energy 171:493–501

Hassan T, Khan Y, He C, Chen J, Alsagr N, Song H (2022) Environmental regulations, political risk and consumption-based carbon emissions: evidence from OECD economies. J Environ Manag 320:115893

Heerink N, Mulatu A, Bulte E (2001) Income inequality and the environment: aggregation bias in environmental Kuznets curves. Ecol Econ 38(3):359–367

Higón DA, Gholami R, Shirazi F (2017) ICT and environmental sustainability: a global perspective. Telemat Inform 34(4):85–95

Hoegh-Guldberg O, Jacob D, Taylor M, Guillén Bolaños T, Bindi M, Brown S et al. (2019) The human imperative of stabilizing global climate change at 1.5 C. Science 365(6459):eaaw6974

Hossain MR, Rej S, Awan A, Bandyopadhyay A, Islam MS, Das N et al. (2023) Natural resource dependency and environmental sustainability under N-shaped EKC: the curious case of India. Resour Policy 80:103150

IFR (2023) World Robotics 2023 Report. International Federation of Robotics. https://ifr.org/ifr-press-releases/news/world-robotics-2023-report-asia-ahead-of-europe-and-the-americas

Im KS, Pesaran MH, Shin Y (2023) Reflections on “Testing for unit roots in heterogeneous panels”. J Econ 234:111–114

Article   MathSciNet   Google Scholar  

Inglesi-Lotz R, Dogan E (2018) The role of renewable versus non-renewable energy to the level of CO 2 emissions a panel analysis of sub-Saharan Africa’s Βig 10 electricity generators. Renew Energy 123:36–43

IPCC (2023) AR 6. intergovernmental panel on climate change. https://www.ipcc.ch/report/ar6/syr/downloads/report/IPCC_AR6_SYR_SPM.pdf

ITU (2023) ICT statistics. International Telecommunication Union. https://www.itu.int/en/ITU-D/Statistics/Pages/stat/default.aspx

Jahanger A, Hossain MR, Onwe JC, Ogwu SO, Awan A, Balsalobre-Lorente D (2023) Analyzing the N-shaped EKC among top nuclear energy generating nations: a novel dynamic common correlated effects approach. Gondwana Res 116:73–88

Kaika D, Zervas E (2013a) The environmental Kuznets curve (EKC) theory—Part A: concept, causes and the CO 2 emissions case. Energy Policy 62:1392–1402

Kaika D, Zervas E (2013b) The environmental Kuznets curve (EKC) theory. Part B: critical issues. Energy Policy 62:1403–1411

Kao C (1999) Spurious regression and residual-based tests for cointegration in panel data. J Econ 90(1):1–44

Karim S, Appiah M, Naeem MA, Lucey BM, Li M (2022) Modelling the role of institutional quality on carbon emissions in Sub-Saharan African countries. Renew Energy 198:213–221

Koondhar MA, Shahbaz M, Memon KA, Ozturk I, Kong R (2021) A visualization review analysis of the last two decades for environmental Kuznets curve “EKC” based on co-citation analysis theory and pathfinder network scaling algorithms. Environ Sci Pollut Res 28:16690–16706

Kuznets S (1955) Economic growth and income inequality. Am Econ Rev 45(1):1–28

Google Scholar  

Lantz V, Feng Q (2006) Assessing income, population, and technology impacts on CO 2 emissions in Canada: where’s the EKC? Ecol Econ 57(2):229–238

Lark TJ, Spawn SA, Bougie M, Gibbs HK (2020) Cropland expansion in the United States produces marginal yields at high costs to wildlife. Nat Commun 11(1):4295

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Le HP, Ozturk I (2020) The impacts of globalization, financial development, government expenditures, and institutional quality on CO 2 emissions in the presence of environmental Kuznets curve. Environ Sci Pollut Res 27:22680–22697

Leal PH, Marques AC (2022) The evolution of the environmental Kuznets curve hypothesis assessment: a literature review under a critical analysis perspective. Heliyon

Li R, Wang Q, Guo J (2024) Revisiting the environmental Kuznets curve (EKC) hypothesis of carbon emissions: exploring the impact of geopolitical risks, natural resource rents, corrupt governance, and energy intensity. J Environ Manag 351:119663. https://doi.org/10.1016/j.jenvman.2023.119663

Li R, Wang Q, Li L, Hu S (2023) Do natural resource rent and corruption governance reshape the environmental Kuznets curve for ecological footprint? Evidence from 158 countries. Resour Policy 85:103890

Li R, Wang Q, Liu Y, Jiang R (2021) Per-capita carbon emissions in 147 countries: the effect of economic, energy, social, and trade structural changes. Sustain Prod Consum 27:1149–1164

Li Y, Zhang Y, Pan A, Han M, Veglianti E (2022) Carbon emission reduction effects of industrial robot applications: Heterogeneity characteristics and influencing mechanisms. Technol Soc 70:102034

Liu B, Yang X, Zhang J (2024) Nonlinear effect of industrial robot applications on carbon emissions: evidence from China. Environ Impact Assess Rev 104:107297

Lorente DB, Álvarez-Herranz A (2016) Economic growth and energy regulation in the environmental Kuznets curve. Environ Sci Pollut Res 23:16478–16494

Luan F, Yang X, Chen Y, Regis PJ (2022) Industrial robots and air environment: a moderated mediation model of population density and energy consumption. Sustain Prod Consum 30:870–888

Mao F, Miller JD, Young SL, Krause S, Hannah DM (2022) Inequality of household water security follows a Development Kuznets Curve. Nat Commun 13(1):4525

Naeem MA, Appiah M, Taden J, Amoasi R, Gyamfi BA (2023) Transitioning to clean energy: assessing the impact of renewable energy, bio-capacity and access to clean fuel on carbon emissions in OECD economies. Energy Econ 127:107091. https://doi.org/10.1016/j.eneco.2023.107091

Naseem S, Guang Ji T, Kashif U (2020) Asymmetrical ARDL correlation between fossil fuel energy, food security, and carbon emission: providing fresh information from Pakistan. Environ Sci Pollut Res 27:31369–31382

Nguyen KH, Kakinaka M (2019) Renewable energy consumption, carbon emissions, and development stages: some evidence from panel cointegration analysis. Renew Energy 132:1049–1057

Numan U, Ma B, Meo MS, Bedru HD (2022) Revisiting the N-shaped environmental Kuznets curve for economic complexity and ecological footprint. J Clean Prod 365:132642

PRS (2023) ICRG. The International Country Risk Guide. https://www.prsgroup.com/explore-our-products/icrg/

Rashdan MOJ, Faisal F, Tursoy T, Pervaiz R (2021) Investigating the N-shape EKC using capture fisheries as a biodiversity indicator: empirical evidence from selected 14 emerging countries. Environ Sci Pollut Res 28:36344–36353

Ray DK, Gerber JS, MacDonald GK, West PC (2015) Climate variation explains a third of global crop yield variability. Nat Commun 6(1):5989

Roelfsema M, van Soest HL, Harmsen M, van Vuuren DP, Bertram C, den Elzen M et al. (2020) Taking stock of national climate policies to evaluate implementation of the Paris Agreement. Nat Commun 11(1):2096

Rojas-Vallejos J, Lastuka A (2020) The income inequality and carbon emissions trade-off revisited. Energy Policy 139:111302

Salman M, Long X, Dauda L, Mensah CN (2019) The impact of institutional quality on economic growth and carbon emissions: Evidence from Indonesia, South Korea and Thailand. J Clean Prod 241:118331

Schiermeier Q (2020) The US has left the Paris climate deal-what’s next? Nature 2022. https://doi.org/10.1038/d41586-020-03066-x

Shehzad K, Zeraibi A, Zaman U (2022) Testing the N-shaped environmental Kuznets Curve in Algeria: an imperious role of natural resources and economic globalization. Resour Policy 77:102700

Stern DI (2017) The environmental Kuznets curve after 25 years. J Bioeconomics 19:7–28

Tapio P (2005) Towards a theory of decoupling: degrees of decoupling in the EU and the case of road traffic in Finland between 1970 and 2001. Transp Policy 12(2):137–151

Tollefson J (2022) What the war in Ukraine means for energy, climate and food. Nature 604(7905):232–233

Ullah A, Khan S, Khamjalas K, Ahmad M, Hassan A, Uddin I (2023) Environmental regulation, renewable electricity, industrialization, economic complexity, technological innovation, and sustainable environment: testing the N-shaped EKC hypothesis for the G-10 economies. Environ Sci Pollut Res 30(44):99713–99734

Usman O, Alola AA, Sarkodie SA (2020) Assessment of the role of renewable energy consumption and trade policy on environmental degradation using innovation accounting: evidence from the US. Renew Energy 150:266–277

Vakulchuk R, Overland I, Scholten D (2020) Renewable energy and geopolitics: a review. Renew Sustain Energy Rev 122:109547

Wang K, Zhu Y, Zhang J (2021) Decoupling economic development from municipal solid waste generation in China’s cities: Assessment and prediction based on Tapio method and EKC models. Waste Manag 133:37–48

Article   PubMed   Google Scholar  

Wang Q, Ge Y, Li R (2024a) Does improving economic efficiency reduce ecological footprint? The role of financial development, renewable energy, and industrialization. Energy Environ 0958305X231183914. https://doi.org/10.1177/0958305X231183914

Wang Q, Hu S, Li R (2024b) Could information and communication technology (ICT) reduce carbon emissions? The role of trade openness and financial development. Telecommunications Policy, 102699. https://doi.org/10.1016/j.telpol.2023.102699

Wang Q, Ren F, Li R (2023a) Exploring the impact of geopolitics on the environmental Kuznets curve research. Sustain Dev 2023. https://doi.org/10.1002/sd.2743

Wang Q, Wang L, Li R (2023b) Trade openness helps move towards carbon neutrality—insight from 114 countries. Sustain Dev 2023. https://doi.org/10.1002/sd.2720

Worldbank (2023a) WDI. World Development Index. https://datatopics.worldbank.org/world-development-indicators/

Worldbank (2023b) WGI. World Governance Index. https://www.govindicators.org/

Xiaoman W, Majeed A, Vasbieva DG, Yameogo CEW, Hussain N (2021) Natural resources abundance, economic globalization, and carbon emissions: advancing sustainable development agenda. Sustain Dev 29(5):1037–1048

Yi M, Liu Y, Sheng MS, Wen L (2022) Effects of digital economy on carbon emission reduction: new evidence from China. Energy Policy 171:113271

Zhang J, Lyu Y, Li Y, Geng Y (2022a) Digital economy: an innovation driving factor for low-carbon development. Environ Impact Assess Rev 96:106821

Zhang L, Mu R, Zhan Y, Yu J, Liu L, Yu Y et al. (2022b) Digital economy, energy efficiency, and carbon emissions: evidence from provincial panel data in China. Sci Total Environ 852:158403

Zhao J, Shahbaz M, Dong X, Dong K (2021a) How does financial risk affect global CO 2 emissions? The role of technological innovation. Technol Forecast Soc Change 168:120751

Zhao S, Hafeez M, Faisal CMN (2022) Does ICT diffusion lead to energy efficiency and environmental sustainability in emerging Asian economies? Environ Sci Pollut Res 29:12198–12207

Zhao W, Zhong R, Sohail S, Majeed MT, Ullah S (2021b) Geopolitical risks, energy consumption, and CO 2 emissions in BRICS: an asymmetric analysis. Environ Sci Pollut Res 28:39668–39679

Zhao Z, Gozgor G, Lau MCK, Mahalik MK, Patel G, Khalfaoui R (2023) The impact of geopolitical risks on renewable energy demand in OECD countries. Energy Econ 122:106700

Zhengxia T, Haseeb M, Usman M, Shuaib M, Kamal M, Khan MF (2023) The role of monetary and fiscal policies in determining environmental pollution: revisiting the N-shaped EKC hypothesis for China. Environ Sci Pollut Res 30(38):89756–89769

Zhong J, Zhong Y, Han M, Yang T, Zhang Q (2023) The impact of AI on carbon emissions: evidence from 66 countries. Appl Econ 2023:1–15. https://doi.org/10.1080/00036846.2023.2203461

Zhou Y, Wang H, Qiu H (2023) Population aging reduces carbon emissions: evidence from China’s latest three censuses. Appl Energy 351:121799

Zou T, Zhang X, Davidson E (2022) Global trends of cropland phosphorus use and sustainability challenges. Nature 611(7934):81–87

Download references

Acknowledgements

This work is supported by the National Natural Science Foundation of China (Grant Nos. 72104246, 71874203).

Author information

Authors and affiliations.

School of Economics and Management, China University of Petroleum (East China), Qingdao, 266580, People’s Republic of China

Qiang Wang, Yuanfan Li & Rongrong Li

School of Economics and Management, Xinjiang University, Wulumuqi, 830046, People’s Republic of China

Qiang Wang & Rongrong Li

You can also search for this author in PubMed   Google Scholar

Contributions

QW: Conceptualization, methodology, software, data curation, writing—original draft preparation, supervision, writing—reviewing and editing. YL: Methodology, software, data curation, investigation writing—original draft, writing—reviewing and editing. RL: Conceptualization, methodology, data curation, investigation writing—original draft, writing—reviewing.

Corresponding authors

Correspondence to Qiang Wang or Rongrong Li .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Ethical approval

This article does not contain any studies with human participants performed by any of the authors.

Informed consent

Additional information.

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

Supplementary information

Supplementary information, rights and permissions.

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

Reprints and permissions

About this article

Cite this article.

Wang, Q., Li, Y. & Li, R. Rethinking the environmental Kuznets curve hypothesis across 214 countries: the impacts of 12 economic, institutional, technological, resource, and social factors. Humanit Soc Sci Commun 11 , 292 (2024). https://doi.org/10.1057/s41599-024-02736-9

Download citation

Received : 16 November 2023

Accepted : 23 January 2024

Published : 21 February 2024

DOI : https://doi.org/10.1057/s41599-024-02736-9

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

Home PowerPoint Templates Curves & Lines U-Curve for PowerPoint

U-Curve for PowerPoint

U-Curve for PowerPoint is a simple presentation design with useful slides that you can use to model a U shaped curve in PowerPoint. This curve can be used in business presentations but also for education purposes. This presentation can be helpful if you need an U shaped curve or inverted U shaped curve.

Inside this template you can find a representation of the  Kuznets curve with an inverted U shape. The Kuznets curve is the representation of Simon Kuznets’ hypothesis that states that as a country develops, there is a natural cycle of economic inequality driven by market forces which at first increases inequality. Then it decreases after a certain average income is attained.

You must be logged in to download this file.

Favorite Add to Collection

Details ( slide)

Subscribe today and get immediate access to download our PowerPoint templates.

Related PowerPoint Templates

Dark Blue Curved Wave PowerPoint Background

Whale Curve Customer Profitability PowerPoint Template

Forgetting Curve PowerPoint Template

Gartner Hype Cycle Curve Template for PowerPoint

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• v.8(11); 2022 Nov

The evolution of the environmental Kuznets curve hypothesis assessment: A literature review under a critical analysis perspective

Patrícia hipólito leal.

a University of Beira Interior, Management and Economics Department, Portugal

b NECE-UBI, University of Beira Interior, Portugal

António Cardoso Marques

Associated data.

Data will be made available on request.

Environmental changes based on factors like urbanization, population, economic growth, increase in energy consumption, and agricultural intensification are never far from the top of any agenda. The topics of environmental degradation and climate change cannot be confined to a single country or region but need to be addressed on a global scale. If the focus is on the relationship between environmental degradation and economic growth, then one hypothesis that is comprehensively used as an empirically model is the widely known Environmental Kuznets Curve. A substantial amount of research has been published about the Environmental Kuznets Curve, and this present study provides a detailed and extensive literature review of more than 200 articles from 1998 to 2022 to explain and assess its evolution. This literature review provides in detail the Environmental Kuznets Curve relationship under analysis, the additional variables included, the type of analysis and methods performed, the relationships obtained, and if the turning point is calculated. Furthermore, this comprehensive literature points out critical issues and gaps in the Environmental Kuznets Curve analysis. It is important to note that there are components that are not considered in the Environmental Kuznets Curve analysis. The Environmental Kuznets Curve only focuses on production and overlooks the impact of the consumption of imported goods on the environment. Consequently, environmental improvements from technological progress will be offset, and economic growth will result in more environmental degradation. This goes against the change in consumer behaviour which occurs with a rise in income, which is one basic assumption of the Environmental Kuznets Curve. The relocation of pollutant industries and consequent relocation of emissions could distort the emissions trajectory over the economic growth path and is also not considered in the Environmental Kuznets Curve analysis. On the other hand, the growth path traced by the inverted U-shaped is not efficient, and the environmental damage provoked in the first phases of the EKC might not be repairable. Therefore, technological progress, climate finance, and energy transition could improve the Environmental Kuznets Curve assessment.

• • A EKC literature survey of more than 200 articles from 1998 to 2022.
• • Comprehensive description of the EKC evolution and its functional specification.
• • Three dilemmas of the EKC are explained by the Green Solow Model.
• • EKC estimation is sensitive to functional specification.
• • Climate finance and technological progress could influence EKC assessment.

Environmental kuznets curve; Green solow model; Environmental degradation; Economic growth; EKC growth path.

1. Introduction

In the pre-industrial period, the earth's carbon circle was considered balanced. However, once the industrial revolution was underway, the burning of fossil fuels provoked a substantial increase in greenhouse gas (GHG) emissions. Society's extreme dependence on fossil fuels came from the necessity to meet rising energy demand. In light of this, the creation of wealth and energy consumption, that is, the income per capita of a country and the amount of energy used became indissociable. Since economic growth relies on increasing energy consumption, it goes hand in hand with rising GHG emissions. Therefore, over decades, economic growth has been achieved to the detriment of the environment, leading to global climate change. The current pandemic situation provides further evidence of this relationship. With economic activity severely affected, global emissions in 2020 were lower than the previous year [ 1 ].

Global warming and climate change are primarily a consequence of anthropogenic behaviours. The production of goods, generation of energy, agricultural activity, transport, and the heating and cooling of buildings are responsible for the release of, on average, 51 billion tons of GHG emissions into the atmosphere each year. The planet's biocapacity has been exceeded, and society is living in a state of ecological transcendence [ 2 ]. The rising risk of undesirable effects for human life, the economy and the environment come from increasing global warming. GHG emissions are the primary driver of and are responsible for rising global average temperature. GHG emissions have increased because of the growth of production, consumption, and population. Obsolete technology plays its part as well. The energy sector is strongly linked to the economy, policy, geopolitical demographics, financial market, and the environment [ 3 ]. Carbon dioxide emissions (CO 2 ), the primary greenhouse gas, are closely related to economic growth, human well-being, financial development, industrialization and urbanization [ 4 ].

Throughout the years, there have been many discussions about the climate change path and the future of the environment. The Brundtland Commission (also known as the World Commission on Environment and Development (WCED)), in the Brundtland report of 1987, raised concerns about the capacity of the environment to satisfy the present and future needs of humanity [ 5 ]. In such a way, a conflict between traditional economic development and environmental well-being arose [ 6 , 7 ]. Sustainable development includes appropriate care of the environment. Since the 1990s, mitigation strategies have been the focus of discussion in both developed and developing countries. To discuss these strategies, summits and agreements were established, such as the Earth Summit conference in 1992 and the Kyoto Summit in 1997. After these, the Conference of Parts (COP), particularly COP21 (Paris in 2015), became one of the most relevant conferences, where a limit on the increase of the global temperature of less than 1.5 °C above pre-industrial levels was established, giving rise to the Paris Agreement. This agreement, which is an international treaty on climate change and is considered a valuable landmark in the climate change mitigation process, defined the necessity to meet every five years to re-evaluate the current state of climate change. The 26th United Nations Climate Change Conference of the Parties (COP26), five years apart from the Paris agreement, was the time for countries to strengthen climate action and define ambitious goals.

The Environmental Kuznets Curve (EKC) is one of the most prevalent methods to analyse environmental performance. The EKC is based on an inverted U-shaped curve created by Kuznets in 1955 [ 8 ]. It was initially designed to study the relationship between income per capita and income inequality. The EKC became more popular when the inverted U-shape started to be adopted in environmental studies. Since then, it has been widely and intensely used as a theoretical framework to study the relationship between yield and environmental degradation [ 9 ]. The emergence of the EKC provoked a change in environmental discussion focus. Before the EKC, concerns were focused on the limited capacity of the planet to absorb urban and industrial waste. With the EKC, the environmental concerns changed from environmental resource scarcity to the inevitable necessity of income growth to deal with pollution [ 10 ].

The EKC defines the trajectory of pollution over time and the income resulting from the economic development of an economy [ 11 ]. Therefore, the EKC is commonly divided into three phases: the early stages of economic development, the turning point, and the later stages of economic development. Briefly (a detailed definition is provided in the following section), considering economic growth over time, the first phase is characterized by an intensive use of resources and a rapid increase in environmental degradation. The second phase, the turning point, is reached when a certain level of income is achieved, and a change in the pollution trajectory occurs, which leads to the third phase, characterized by environmental degradation mitigation. Bringing into mind the indissociable relationship described at the beginning of this section, the early stages of economic development represent that. However, when the turning point is reached, income starts to be dissociable from emissions and environmental degradation, leading to the later stages of economic development, where there is the dissemination of clean technology and innovation.

The EKC has been widely applied in the environment-energy-economics literature, and innumerable researchers have attempted to validate the inverted U-shaped between environmental degradation and income. Therefore, the EKC has been assessed for the most diverse contexts (country/ies, time span, variables, and methods) yet there is still no consensus on the results. With this in mind, this review article aims to answer the following research questions: (i) Is the EKC keeping up with the increasing complexity of environmental issues?; (ii) What has been influencing the inverted U-shaped curve?; and (iii) How can the fit of the EKC be improved to meet the complexity of the economic growth and environmental degradation relationship? To answer these questions, an extensive survey of the EKC literature is provided with the objective to (i) describe the evolution of the EKC assessment and provide an integrated overview of the current state of EKC knowledge; (ii) identify the factors that influence the EKC validation; and (iii) describe research insights, existing gaps, and provide improvement needs.

Overall, this research intends to be a valuable tool for EKC researchers and is differentiated from the existing review articles by providing a detailed description of the EKC background, which includes the origins and conceptual framework, an explanation of the EKC shape, and the distinct phases of development, issues, and challenges of the EKC analysis, and the factors that most affect the EKC shape. Besides that, this paper also describes the close relationship between the EKC and the macroeconomic Green Solow Model. Furthermore, this literature review provides an embracing description of the evolution of the EKC analysis through an extensive literature survey and specifies each detail of the analysis of more than 200 papers from 1998 to 2022. The analysis of the EKC literature for this extended period allows us to understand what is currently analysed, in addition to the evolution of the EKC assessment over the years. This literature survey is being conducted so as to be an intuitive tool for researchers to efficiently find specific information about the procedures used in the literature focused on the EKC study, namely: (i) country (ies) and time period; (ii) variables analysed on EKC validity; (iii) additional variables included in the EKC analysis; (iv) types of analysis and method(s) employed; (v) relationships obtained, and (vi) turning point. This literature survey conducts a critical analysis of the EKC approach, identifying critical issues, proposing improvements, and future lines of research.

This paper is divided into five sections. Section 2 presents the origins, conceptual framework and shape of the EKC. Section 3 follows, where the details of the evolution of the EKC analysis can be found. Section 4 describes the gaps in the EKC assessment, and lastly, in Section 5, the conclusions of the research are given.

2. Origins, conceptual framework, and shape of the EKC

The EKC was preceded by the Research of Kuznets [ 8 ], on which the EKC is based. Simon Kuznets won a Nobel prize for his framework based on the economic and social structure of national development procedures [ 12 ]. The results of the research of Kuznets [ 8 ] disclosed an inverted U-shaped relationship between income per capita and income inequality. According to Kuznets [ 8 ], the inverted U-shaped relationship revealed an unequal income distribution in the early stages of income growth that moves towards equal income distribution with increasing economic productivity in the later stages of economic growth. Therefore, Kuznets [ 8 ] specified that the transition from a pre-industrial to an industrial development firstly led to income inequality. This is followed by a rising income per capita together with superior income equality. The EKC attracted a lot of attention from policymakers, theorists and empirical researchers and started to be widely used in environmental studies [ 13 , 14 ] through the seminal research of Grossman and Krueger [ 9 ], carried out in 1991. They revealed that the relationship between income per capita and environmental degradation, like the income per capita and income inequality of Kuznets [ 8 ], also follows an inverted U-shaped curve.

In the early 1990s, the main idea in economics was “ too poor to be green ” [ 15 ]. According to Beckerman's [ 15 ] point of view regarding the effect of economic growth on environmental degradation, the author argues that there is: « clear evidence that, although economic growth usually leads to environmental deterioration in the early stages of the process, in the end, the best and probably the only way to attain a decent environment in most countries is to become rich ». This view reflects the basic philosophy of the EKC theory. The World Development Report in 1992 argues that some environmental problems are aggravated by the growth of economic activity, and it suggests that accelerated equitable income growth will make it possible to achieve higher world output and improved environmental conditions [ 16 , 17 ]. This proposal lays the foundation of the EKC literature. A robust foundation for the EKC is provided by Dinda [ 18 ], Stern [ 19 ], and Kaika and Zervas [ 20 ], and it is presented throughout this paper. Kwabena et al. [ 21 ] and Olale et al. [ 22 ] provide a survey of theoretical research related to the EKC.

2.1. Conceptual framework of the EKC

The EKC is commonly interpreted in two ways. One is through the division into two phases, namely the early and later stages of economic development. The early stages are defined, on the one hand, by a decreasing capacity of ecosystem regeneration as a consequence of intensive use of resources that lead to a rising ecological footprint and pollution [ 13 , 23 ]. On the other hand, the early stages are linked with lax environmental regulations associated with a low capacity to pay for environmental conservation [ 24 ]. The later stages are characterized by mitigation of environmental degradation resulting from the dissemination of clean technology and innovation, society environmental awareness, and effectiveness and institutional quality associated with an increase in the level of income [ 13 , 23 ]. In addition, these stages are also characterized by two effects, i.e., policy effect and income effect. The policy effect consists of greater public concern about the environment, which leads to rigorous regulatory requirements. At the same time, the income effect consists of the increase in income that leads to an increase in the willingness to pay for environmentally-friendly features [ 24 ]The other way that the inverted U-shaped curve is commonly interpreted is when economic development is divided into three phases of [ 13 , 20 , 25 ], namely: (i) the pre-industrial economy, mainly characterised by primary sector and low levels of income; (ii) the industrial economy, constituted by the secondary sector and associated with middle-income levels; and (iii) the post-industrial economy, formed by the tertiary sector and services, and associated with higher levels of income. In the pre-industrial economy, economic activity is limited and results in a natural resource abundance and reduced formation of waste [ 20 , 26 ]. In this phase, the use of pollutant technology, the lack of environmental awareness, and the prioritisation of economic growth result in rising environmental degradation [ 27 ]. The industrial economy is characterised by natural resources that are starting to run out and increasing waste accumulation because of industrialisation. In this phase, a positive relationship between economic growth and environmental deterioration is verified, and it occurs before the turning point is achieved. The third phase of economic development is characterised by a structural change in the economy, changing to information- and technology-intensive industries and a services-directed economy. This change is linked with the reinforcement of environmental regulations, the use of cleaner and efficient technology, and a strengthening of environmental awareness, resulting in a mitigation of environmental degradation [ 20 , 26 ]. In this phase, a negative relationship between economic growth and environmental deterioration is verified, and it occurs after the turning point has been reached.

2.2. Shape of the EKC

The EKC consists of an inverted U-shaped curve between income and environmental degradation; that is, the EKC defines the pollution trajectory over time and income resulting from economic development [ 11 ]. The EKC is a long-run concept [ 28 ]. In light of this, the EKC reflects a dynamic environment–economy relationship concentrating on long-run processes of change [ 29 ]. The EKC is assessed through the nature of the effect of the income and its square (to ensure the concavity of the curve) on environmental degradation. The inverted U-shaped curve is validated through the significant and positive coefficient and elasticity of income simultaneously with the significant and negative coefficient and elasticity of income squared. Therefore, considering β 1 as the coefficient of income and β 2 as the coefficient of income squared, both in the longrun, the EKC is verified according to the condition β 1 > 0 Λ β 2 < 0 (in which this paper is focused).

The assessment of the EKC could lead to the validity of the following conditions (see Figure 1 ):

• 1. β 1 = β 2 = 0 . No relationship between x and y.
• 2. β 1 > 0 Λ β 2 = 0 . Linear relationship between x and y.
• 3. β 1 < 0 Λ β 2 = 0 . Decreasing relationship between x and y.
• 4. β 1 < 0 Λ β 2 > 0 . U-shaped relationship.
• 5. β 1 > 0 Λ β 2 < 0 . Inverted U-shaped relationship—EKC.

Graphical representation of the relationship between an Environmental Indicator and an Economic Indicator. Legenda : (i) No relationship between x and y; (ii) Positive relationship between x and y; (iii) Decreasing relationship between x and y; (iv) U-shaped relationship between x and y; (v) Inverted U-shaped relationship between x and y; (vi) Inverted N-shaped relationship between x and y; and (vii) N-shaped relationship between x and y.

where, y is the environmental indicator and x is the income.

Besides these ones, two more conditions might be obtained in the EKC assessment. These two imply the inclusion of the third polynomial, income cubed ( β 3 ).

• 6. β 1 < 0 , β 2 > 0 Λ β 3 < 0 . Opposed to the N-shaped curve.
• 7. β 1 > 0 , β 2 < 0 Λ β 3 > 0 . Cubic polynomial or N-shaped curve.

Throughout the years, several authors have highlighted factors that affect the shape of the inverted U-shaped curve. Panayotou [ 13 ], one of the first authors assessing the EKC hypothesis, disclosed that policy distortions, such as market breakdowns, under-pricing of natural resources, and subsidies on economic structures intensive in carbon and energy affect the slope of the inverted U-shaped curve. In turn, Kaika and Zervas [ 20 ] identified the following factors: institutional framework and governance, consumers’ preferences, and equity of income distribution. The willingness of governance to implement environmental regulation is considered crucial to mitigate environmental degradation [ 30 ]. When governance institutions are weak, less effective or corrupted, this could affect the shape of a possible EKC and change the turning point to higher income levels [ 31 ].

Many researchers have assessed whether the equity of income distribution affects the EKC pattern [ 32 , 33 , 34 ]. To assess this, the crucial question is whether economic growth leads to equitable income distribution or increases income inequality. The automatic thought could be that economic growth leads to a more equitable income distribution that consequently leads to an improvement in public awareness of environmental degradation and the imposition of suitable environmental regulations. Income distribution is the distribution of power, and pollution decreases or increases depending on the gap of power between the citizens who suffer due to pollution and the ones that benefit from pollution [ 32 ]. Therefore, if income inequality worsens, this will lead to continuing environmental deterioration due to the fact that the ones who suffer from environmental degradation will not have the economic conditions to impose environmental regulations on the ones that would benefit from it [ 34 ].

In the recent literature, the factors most considered to affect the shape of the EKC are the scale, composition, and technique effect; income elasticity of environmental quality; and international trade [ 25 ]. The scale, composition, and technique effect are the three stages used to characterise the relationship between environmental degradation and economic development [ 9 ]. The scale effect denotes environmental degradation as a consequence of economic development, that is, a negative impact of economic growth on the environment. The negative impact is a consequence of intensive use of natural resources to supply an increasing demand and consequent increasing production output. This intensive energy consumption comes mainly from fossil fuels that are a cheap, abundant, and easy-to-transport energy source. The composition effect is characterised by structural changes in the economy, which could provoke both negative and positive impacts of economic development on the environment. A change from an economy directed to the primary sector to energy- and carbon-intensive industries results in a negative impact. In contrast, a shift from pollution-intensive industries to an economy directed to the services sector results in a positive impact [ 23 ]. The technique effect denotes a mitigation effect of economic development on environmental degradation. This is explained by a higher level of income that leads to investing in research and development, replacement of dirty and outdated technologies, and strengthened environmental regulations.

The income elasticity of environmental quality demand consists of the ratio between the variation in the environmental quality demand and the variation in income level. The role of this factor on environmental degradation mitigation is highlighted [ 18 , 23 , 35 ]. The income elasticity denotes that with rising economic development, society intuitively lives in a higher standard and yearns for quality instead of quantity. Therefore, there is greater environmental awareness and availability of money to pay for a cleaner environment [ 36 ], which leads to an adjustment in consumer behaviour, for instance, opting for energy-efficient and environmentally friendly products and services [ 37 , 38 ] and donating to environmental protection organizations [ 18 ].

International trade is considered one of the most crucial factors explaining and affecting the shape of the EKC, and the EKC pattern may appear as a result of it [ 39 , 40 , 41 ]. In light of this, trade policies are crucial to explain the EKC. Trade openness leads to economic expansion through the request to increase the production of goods to satisfy its exports. Broadly speaking, countries tend to become specialized in sectors in which they have a competitive advantage as a consequence of trade liberalization. However, on the one hand, if these sectors derive from weak environmental regulation, trade liberalization induces environmental damage, which consequently results in an industrial process with high pollution abatement costs [ 9 ]. On the other hand, when income and environmental degradation significantly increase, stringent environmental regulations are imposed and implemented, which consequently lead to the shift of pollution-intensive goods production to other countries. These countries are usually low-income countries with weak and lax environmental legislation [ 18 , 20 ]. This is defined as the Pollution Haven Hypothesis (PHH). The lax environmental regulation in the developing countries provides a comparative advantage for the developed economies, which leads to the reduction of environmental degradation in the developed economies while increasing it in the developing ones [ 42 ]. An inverted U-shaped curve is obtained through two phases. Firstly, the export of goods in a developed country causes the upwards slope of the curve (early stages of economic development). After that, the import of goods from developing countries causes the downward slope of the curve (later stages of economic development). The downward slope of the curve is reported as the PHH [ 18 ].

2.3. The EKC and the green solow model

The assessment of the EKC occurs through the analysis of environmental degradation over an increasing income. Economic growth is a macroeconomic indicator, and the Solow Model is considered the main model of modern macroeconomics. Brock and Taylor [ 43 ] developed a theoretical framework to explain the EKC. By incorporating environmental pollution into the Solow Model [ 44 ], the authors created the Green Solow Model. According to Brock and Taylor's [ 43 ] pollution data and their empirical work on the EKC, three dilemmas were revealed, and the Green Solow Model provides an explanation for each one. The dilemmas are namely: (i) the ongoing huge decline in emissions intensity simultaneously with almost stagnant pollution abatement costs; (ii) what feature gives the humped-shape profile to the pollution levels when it is graphically represented against income per capita or time; and (iii) the fragile empirical results of cross-country analysis indicate that the EKC is not validated, or the problem is applying empirical approaches that are subject to extensive variance.

To assess the first dilemma, the authors [ 43 ] analysed two concrete cases, the United States (US) and Europe. In both cases, while a huge variation in emissions occurred, an insignificant variation in the pollution abatement expenditures/costs was observed. However, in both cases, the EKC pattern is graphically visible when emissions are graphed against time and an income increase over the same period is considered [ 43 ]. In the US, a huge variation in emissions has taken place over the last 20 years while simultaneously, pollution abatement costs have remained at less than one-half of 1% of Gross Domestic Product (GDP) for the same time period [ 45 ]. In Europe, an emissions reduction of 4–5% per year has been observed alongside a pollution control cost with an average of only 1–2% of GDP. To provide answers to this dilemma, the authors considered exogenous technological progress in abatement and a fixed intensity of abatement.

Theories based on strict environmental policies expect growing costs to mitigate environmental degradation. In a scenario of a world that does not have technological progress for this, a huge investment in pollution control is needed [ 46 ]. Technological progress in goods production and abatement leads to continual growth alongside increasing environmental quality. Through the formulation and development of the model, the authors conclude that technological progress in goods production is required to produce income growth. Besides that, technological progress in abatement must go above growth in aggregate output for pollution to decrease and, consequently environmental quality to increase. These two conditions make sustainable growth guaranteed. In light of this, technological progress in abatement increases the effectiveness of the share of output applied to reducing environmental damage. Output growth results in an increase in emissions; however, then emissions decrease as technology is applied to offset environmental damage.

The second dilemma refers to the feature that gives the humped-shaped profile to emissions when graphed against time or income. This dilemma consists of the analysis of the existence of the turning point, which allows the humped-shaped through inverting the emissions’ trajectory. Through the Green Solow Model, and as mentioned in the first dilemma, sustainable growth is guaranteed when technological progress exists in production (it is essential to produce income growth), and when technological progress in abatement goes further than the growth of aggregate output (this mitigates pollution). In addition, by recurring to the Cobb-Douglas function, Brock and Taylor [ 43 ] conclude that if an economy has small initial capital stock, then emissions firstly increase and then start to decrease as development continues. Therefore, the emissions humped-shaped EKC profile is obtained if growth is sustainable, and simultaneously the stock of capital at the turning point is higher than the initial stock of capital. This leads to an initial positive growth rate of aggregate emissions that become negative in finite time. The answer to the first dilemma also helps in understanding the second dilemma. Through technological progress in abatement, a time profile of increasing and then decreasing emissions, with income per capita growing along a path of sustainable growth, is generated.

The third dilemma is related to the variation that samples and the estimation procedure provoke on the EKC empirical regressions. The answer for this dilemma starts in the second dilemma by recurring to the Cobb-Douglas function to assess the initial conditions, which are the initial technological progress in goods production, labour and units of pollution. Different profiles of income per capita and emission over time are obtained as a consequence of economies with different initial conditions. Considering this, heterogeneity could explain the sensitivity of the EKC results to the sample. Therefore, this explains the absence of a consensus on the EKC results in country-level data and the possible difference between the EKC empirical results in cross-country analysis and the country-level analysis of the same countries. The cross-country analysis that includes developed and developing countries is a plausible example to demonstrate the effect of heterogeneity. Clearly, these countries differ in more than the initial condition. The heterogeneity in this analysis may further confound the estimation. According to the EKC literature, the time period, the countries sampled, and even the environmental indicator chosen could provoke a change in the shape of the estimated EKC. Even for similar countries, the EKC profiles are not unique due to the differences in the initial conditions.

3. Getting inside the evolution of EKC analysis

The first literature on the theory of the EKC was focused on developing models that replicated the inverted U-shaped curve relationship. Considering the increasing complexity of reality (such as technological development, the introduction of renewable energy sources in the energy sector, increasing industrialization and globalization) the EKC analysis has had to be continuously improved. Throughout the EKC literature, diverse literature surveys were developed. However, on the one hand, most of the articles that provide a comprehensive contextualization of the EKC literature are not focused on the EKC theoretical background or critical analysis, instead, they are focused on an empirical analysis [ 47 , 48 , 49 , 50 ]. On the other hand, articles focused on the evolution of the EKC literature use specific approaches to assess the EKC literature, such as meta- and bibliometric analysis [ 25 , 51 , 52 ], which provide the research areas on the subject, the author's contribution and most cited authors, journals that are publishing on the subject, and keywords used. Differentiated from these, the present review article is focused on providing an extensive and comprehensive contextualization of the EKC framework, the evolution of the literature, current analysis, and critical analysis that addresses gaps, issues, and improvement needs. Besides that, and as a prominent contribution to the literature, this article provides a useful and intuitive tool for EKC researchers where they can find detailed information about the EKC analysis in more than 200 articles since the country (ies) and time period under analysis, and the variables for each EKC relationship is analysed, until each additional variables included, the types of analysis and method(s) employed, the relationship obtained for each sample analysed, and if the turning point is calculated.

The procedure for an analysis of the EKC focuses on two key areas: the selection of the EKC relationship variables and the selection of the EKC functional specification. The latter includes the method, additional variables, temporal period, and cross-country or individual analysis. This section demonstrates how the selection of each element has progressed in the EKC literature. Some examples are displayed in tables (just a few examples from the substantial number of articles presented in the tables in the supplementary data), providing an organised and intuitive literature review of the EKC literature. The tables are organized into the EKC relationship analysed, approaches used, additional variables included, countries analysed (individual or cross-country), and the relationship obtained. This schematisation allows not only an observation of the evolution of the EKC analysis but also an identification of gaps in the analysis. The third dilemma identified by Brock and Taylor [ 43 ] regarding the variation that the sample and the estimation procedure provoke on the EKC empirical regressions is also explored.

3.1. The EKC relationship: from environmental to other types of indicators

The variables selected to assess the EKC, that is, the variables for which a relationship that follows the EKC is assessed, are originally an environmental indicator and an economic indicator. However, over the years, in place of the environmental indicator, several other types of indicators have been used to assess the EKC. Considering the EKC literature collected in this paper, the relationship between CO 2 emissions and GDP is the most frequently analysed [ 4 , 53 , 54 , 55 , 56 , 57 ] (see supplementary data), which makes CO 2 emissions the environmental indicator most often used (about 100 articles out of 200 collected in this paper). Notwithstanding, throughout the years, innumerable environmental indicators have been used to assess the EKC, such as air pollution, ecological footprint, waste, afforestation, water consumption, and others. However, not only are environmental indicators assessed. Besides these indicators, also energy consumption, pollution abatement costs, environmental crimes, and health indicators have been analysed. In Table 1 , the diversity of EKC relationships analysed are displayed.

Table 1

Variables of EKC relationship.

Notes: CO 2 denotes Carbon Dioxide; FDI denotes Foreign Direct Investment; GDP denotes Gross Domestic Product; GHG denotes Greenhouse Gases GRP denotes Gross Regional Product; GVA denotes Gross Value Added; HDI denotes Human Development Index; PM 2.5 denotes Particulate Matter (2.5 μm).

In respect of the environmental indicators, choosing among diverse indicators of environmental degradation is challenging considering the complexity and multiple dimensions of environmental problems. Therefore, the selection of an indicator takes place between numerous types; however, atmospheric indicators have been the most abundant. This type of indicator includes emissions of CO 2 , GHG, Nitrous Oxide (N 2 O or NO 2 ), and others. The pollutant under analysis could be local or global. Some studies use local pollutants, such as Sulphur Dioxide (SO 2 ), water pollution and deforestation, while others use global pollutants, such as CO 2 emissions. Environmental degradation indicators are the most often chosen to assess the EKC. However, the indicator does not have to be of degradation; it could be of environment recovery, concern, or protection. Besides the atmospheric indicators, (i) land and forests; (ii) oceans, seas, coasts, and biodiversity; and (iii) freshwater indicators have also been analysed [ 25 ].

The various indicators used in assessing the EKC relationship throughout the literature have given rise to different forms of this model. Consequently, the EKC concept is often converted depending on the type of indicator used in the relationship. Energy indicators (e.g. renewable energy consumption, non-renewable energy consumption, energy consumption, energy intensity, and others) are frequently used to assess the EKC relationship, from which emerge adaptations of the EKC depending on the energy indicator used, such as the Renewable Kuznets Curve. Therefore, it is common in the literature to see adaptations from the EKC linked to the specific indicator used instead of the usual environmental indicators. Throughout this review article, several EKC forms are addressed (see Table 1 and supplementary data).

The selection of the EKC relationship to analyse is one of the first steps in EKC studies, and this choice influences the validation of the EKC. The empirical results of the EKC are not unique, and they are sensitive to the variables under analysis, as is the type of pollutant. One example of that is the study developed by Shafik and Bandyopadhyay [ 11 ], which analysed ten indicators of environmental pressure, and from these ten, only two followed the EKC. In light of this, it is notable to mention the degree of sensitivity of the EKC regarding the environmental degradation indicator under analysis. Another example is the study developed by Altıntaş and Kassouri [ 94 ], which analysed the EKC relationship between CO 2 emissions and GDP, and ecological footprint and GDP, for the same samples of countries and time period, and with the same approaches. They obtained different results for each environmental indicator. Ecological footprint validated the EKC, while CO 2 emissions revealed a U-shaped curve. The selection of the environmental indicator or other indicator used in the EKC analysis gives rise to a gap in the literature, which confirms that the inverted U-shaped curve (EKC) is only demonstrated for some environmental indicators. This is, according to Liu [ 95 ], due to the lack of consistent data, assessing the EKC for industrial pollution and human health had not been possible.

Besides the immense variety of environmental or other types of indicators applied in the hypothesis, variables chosen for the economic indicator also have been diverse, although not to the same extent. GDP is the most frequent economic indicator used in the EKC relationship. However, over the years, other indicators have been used, such as Gross Regional Product (GRP), Foreign Direct Investment (FDI), Gross Value Added (GVA), Gross State Product, income inequality, economic complexity index [ 96 ], air transport passenger [ 97 ], manufacturing sub-sector output [ 98 ], Oil Rents [ 99 ], and others. In the selection of the economic indicator, not only does the indicator used in the EKC relationship influence its validation, but also the data analysed. Kacprzyk and Kuchta [ 49 ] developed an analysis using different GDP data for analysing the EKC between GDP and CO 2 emissions. The use of three different measures of GDP revealed ixed results.

3.2. EKC functional specification: approach, additional variables, time period and countries sample

After selecting the variables to assess the EKC relationship, follows the adoption of the functional specifications, which consists of choosing the method/approach and the structure of the model. The structure of the model includes the additional variables beyond the variables of the EKC relationship, the time period, and the sample of countries to analyse.

3.2.1. Additional variables

The additional variables are those included in the estimation beyond the variables for which the EKC relationship is assessed. With the increasing complexity of reality, the additional variables included in the estimations are innumerable and of several types. According to Kaufmann et al. [ 100 ] and Itkonen [ 101 ], the inclusion of additional controls influences the EKC assessment and the results of the EKC estimation. Table 2 displays a summary of the additional variables.

Table 2

Additional variables.

Energy consumption quickly became the most common variable added to the EKC estimations [ 102 , 103 , 104 , 105 , 106 ]. Considered as one of the main drivers of environmental degradation and climate change, energy consumption has been analysed with most of the environmental indicators. The analysis of energy consumption has improved over the years. It started with the analysis of energy consumption in its aggregate form, as a whole, and evolved by analysing energy consumption by types of technology, renewable and non-renewable, and after that by energy sources, such as coal, oil, gas, nuclear, solar, wind, and others. The inclusion of energy consumption in the EKC assessment keeps up with any improvement in energy consumption analysis [ 107 , 108 , 109 , 110 , 111 , 112 ]. However, the inclusion of energy consumption as one of the CO 2 emissions determinates could cause an underestimation of both the sensitivity of CO 2 emissions to income growth and the turning point of the EKC [ 101 , 113 ]. This occurs because the two data series are related by construction. Consequently, any other variable included in the model can only explain the carbon intensity of energy consumption and not the CO 2 emissions level [ 101 ].

3.2.2. Approach or method

At the beginning of the EKC literature, countless studies focused on the proximate aspects of the theory, which consequently took to reduced-form models. These models connect income and pollution directly through estimations and tests of correlations between indices of environmental condition and development [ 114 ]. The reduced-form models are simpler and are of limited utility [ 12 ]. Therefore, the need arose to improve the EKC analysis, and the studies started to employ structural equation models and included intervenient variables, which in turn, connected development processes with environmental outcomes. In light of this, and until the present day, the EKC has been assessed through innumerable approaches/methods and econometric procedures. Table 3 displays some examples of the methodologies applied (more examples are given in the supplementary data).

Table 3

Approaches performed.

Notes: ARDL denotes Autoregressive Distributed Lag; CCEMG denotes Common Correlated Effects Mean Group; CCR denotes Canonical Cointegrating Regression; CS denotes Cross-Sectional; DFE denotes Dynamic Fixed Effect; DOLS denotes Dynamic Ordinary Least Square; FGLS denotes Feasible General Least Squares; FMOLS denotes Fully Modified Ordinary Least Square; GARCH denotes Generalized Autoregressive Conditional Heteroskedasticity; GLS denotes Generalized Least Squares; GMM denotes Generalized Method of Moments; LSDV denotes Least Square Dummy Variable; MG denotes Mean Group; MMQR denotes Method of Moments of Quantile Regression; OLS denotes Ordinary Least Square; NARDL denotes Nonlinear Autoregressive Distributed Lag; PCSE denotes Panel Corrected Standard Errors; PMG denotes Pooled Mean Group; SDM denotes Spatial Durbin Model; SEM denotes Spatial Error Model; SFE denotes Static Fixed Effect; SGVAR denotes Semi-Parametric Global Vector Autoregressive Model; SLM denotes Spatial Lag Model; STSM denotes Structural Time Series Model; SUR denotes Seemingly Unrelated Regression; SYS-GMM denotes System Generalized Method of Moments; Sys2Step denotes Two-Step Dynamic System Generalized Method of Moments; VAR denotes Vector Autoregressive Model; VECM denotes Vector Error Correction Model; 2SGMM denotes Two-step Dynamic Generalized Method of Moments.

The econometric issues are one of the main topics criticised in the EKC estimation. Therefore, the methods employed in the EKC analysis give rise to significant criticism. The EKC is commonly estimated through reduced-from regressions, which is frequently disparaged by several researchers [ 141 , 142 , 143 ]. Furthermore, empirical EKC research commonly uses standard cointegration techniques that are often considered unsuitable [ 19 , 144 , 145 ]. Kacprzyk and Kuchta [ 49 ] provide further explanations for this inadequacy. According to Gill, Viswanathan and Hassan [ 10 ], the EKC literature is not econometrically demanding, and the empirical results of the EKC analysis are very sensitive regarding the functional form of the model. Considering this, it is fair to say that the EKC assessment is sensitive and influenced by the model or econometric procedure used (please see Table 4 , Table 5 , Table 6 displays several examples of this).

Table 4

Individual analysis.

Notes: AMG denotes Augmented Mean Group; ARDL denotes Autoregressive Distributed Lag; CCR denotes Canonical Cointegrating Regression; CO 2 denotes Carbon Dioxide; DOLS denotes Dynamic Ordinary Least Square; EC denotes Energy Consumption; EF denotes Ecological Footprint; FMOLS denotes Fully Modified Ordinary Least Square; GDP Gross Domestic Product; NARDL denotes Nonlinear Autoregressive Distributed Lag; OLS denotes Ordinary Least Square; TO denotes Trade Openness; UK denotes United Kingdom; US denotes United States; VECM denotes Vector Error Correction Model.

Table 5

Cross-country analysis.

Notes: ARDL denotes Autoregressive Distributed Lag; BRICS denotes Brazil, Russia, India, China and South Africa; CCE denotes Common Correlated Effects; CO 2 denotes Carbon Dioxide; DOLS denotes Dynamic Ordinary Least Square; EC denotes Energy Consumption; EF denotes Ecological Footprint; FD denotes Financial Development; FE denotes Fixed Effects; FDI denotes Foreign Direct Investment; FMOLS denotes Fully Modified Ordinary Least Square; GDP Gross Domestic Product; GLS denotes Generalized Least Squares; GMM denotes Generalized Method of Moments; HDI denotes Human Development Index; OLS denotes Ordinary Least Square; PMG denotes Pooled Mean Group; SUR denotes Seemingly Unrelated Regression; TO denotes Trade Openness; VECM denotes Vector Error Correction Model.

Table 6

Individual vs cross-country analysis.

Notes: AMG denotes Augmented Mean Group; CCE denotes Common Correlated Effects; CO 2 denotes Carbon Dioxide; DOLS denotes Dynamic Ordinary Least Square; EF denotes Ecological Footprint; FDI denotes Foreign Direct Investment; FMOLS denotes Fully Modified Ordinary Least Square; GDP Gross Domestic Product; HDI denotes Human Development Index; MG denotes Mean Group; OECD denotes Organisation for Economic Co-operation and Development; SO 2 denotes Sulphur dioxide; UK denotes United Kingdom; US denotes United States; USA denotes United States of America; VECM denotes Vector Error Correction Model.

Underlined in Table 6 are examples of studies that use more than one methodology and obtain different results for the same country, depending on the methodology applied. In the study developed by Bilgili et al. [ 146 ] two methodologies are used, Dynamic Ordinary Least Square (DOLS) and Fully Modified Ordinary Least Square (FMOLS) and 17 Organisation for Economic Co-operation and Development (OECD) countries are analysed. In the individual countries analysis, different results are obtained for Turkey, France and Netherlands. With the FMOLS, a U-shaped relationship was obtained for Turkey, while an inverted U-shaped curve (EKC) was obtained for France and Netherlands. However, with the DOLS, an inverted U-shaped curve (EKC) was obtained for Turkey. In contrast, a U-shaped relationship was obtained for France and the Netherlands.

Another example is the study developed by Destek and Sinha [ 147 ]. Two methodologies were used to assess the EKC in the individual countries' analysis, namely FMOLS and Common Correlated Effects (CCE). Through both approaches, FMOLS and CCE, a U-shaped curve relationship was obtained for Austria, Canada, Greece, Italy, Japan, S. Korea, Spain and the US, while an inverted U-shaped curve relationship was revealed for Germany and Turkey. However, for Belgium, Switzerland, Denmark, and the Netherlands, only one of the two methodologies obtained a U-shaped curve relationship. The same for Chile, France, Mexico, New Zealand, Portugal, and the United Kingdom, where only one of the two methodologies obtained an inverted U-shaped curve relationship.

3.2.3. Countries sample

The EKC has been assessed for several individual countries (see Table 4 ) or groups of countries (see Table 5 ).

However, there is no consensus in the results. The selection of the country (ies), and consequently the cross-sectional or individual analysis performed, directly influences the relationship obtained. According to the literature (see some examples in Table 6 ), and as identified by Brock and Taylor [ 43 ] as the third dilemma of the EKC, heterogeneity makes the EKC results sensitive to the sample. With this in mind, heterogeneity could be one of the main reasons for the difference between the EKC empirical results in cross-country analysis and country-level analysis. Therefore, studies that perform both cross-sectional and individual analyses obtain mixed results. This means when validating the EKC for a group of countries and when analysing each country individually, some countries follow the EKC trajectory, other countries follow a U-shaped relationship, and other countries follow neither a U-shaped nor inverted U-shaped relationship. Table 6 displays various examples of EKC studies that performed both cross-sectional and individual analysis, obtaining different results. Cross-sectional analysis results are in bold in Table 6 .

The individual country data analysis assesses the EKC for the environmental condition of a nation, a single economy, throughout time, with increasing income as it develops. Instead of that, cross-country analysis assesses the EKC for the environmental and economic conditions of a group of countries, with distinct stages of development, at a certain moment in time or within a limited time period. Therefore, considering fundamental disparities in national backgrounds and differences in the development paths, a cross-country analysis that reveals an inverted U-shaped pattern does not reveal that each country individually follows the EKC trajectory [ 158 ]. In line with this, through the development of the Green Solow Model, it was concluded that different profiles of income and emissions over time are obtained as a consequence of economies with different initial conditions. EKC profiles are not unique due to the differences in the initial conditions [ 43 ]. The EKC estimation is vulnerable to the selection of scale, sample, and range, as well as the spatial and temporal sample range. Therefore, changes could occur in the estimated coefficients, significance levels and variables specification as a consequence of the country or countries under analysis [ 11 ].

4. Which are the gaps in the EKC assessment?

With an extensive range of literature, the EKC is a method massively employed to analyse economies’ environmental performance. Through the fast growth and development of economies and technology and the increasing complexity of environmental degradation issues, the EKC has started to be employed to analyse not only environmental indicators and not only through simple models. However, an absence of consensus is noted, as well as the degree of sensitivity of the EKC estimation to all the elements that are incorporated into the analysis process. The selection of the indicators, country or countries, time period, methodology and additional variables produce a unique result, and when the EKC is validated, the estimated curve is unlike any other. In light of this, a change of a single component of the functional structure produces changes in the results, validating or not the EKC or changing its shape.

The sensitivity of the EKC function and the consequent change in the results is criticised in some cases. A critical issue raised about the relationship obtained in a panel analysis is that it does not imply that individual countries follow the same pattern (as shown in Table 6 ). The cross-sectional analysis in EKC studies, mainly performed due to the lack of reliable long-term data [ 12 ], analyses a set of economies with different conditions and backgrounds. Taking this into account, the results obtained during a panel analysis should not be directly compared with individual analysis results. Studies that perform a cross-sectional analysis provide results and measures for the specific group of countries under analysis, which may not be the most appropriate for each country in particular. Therefore, to create and apply measures adequate for the characteristics of each economy, the analysis of each country individually or by sector could provide and reveal more specific and beneficial results to the policymakers.

Besides that, EKC estimations are found to be sensitive to the method performed to assess the hypothesis (see Subsection 3.2.2 and Table 6 ). In other words, the shape obtained is linked to the econometric method chosen and functional specification. Several econometric issues in the EKC modelling have been identified throughout the EKC literature. EKC function and the use of income and squared income variables raised some criticism due to the production of econometric issues. Model emissions as a function of income augmented by income squared and income-cubed raise econometric issues, such as collinearity or multicollinearity [ 159 ]. In order to avoid multicollinearity in the estimation, the use of non-parametric or semi-parametric methods should be explored. Besides that, the validity of the EKC hypothesis could be based on the assessment of short- and long-run income elasticities, as stated by Narayan and Narayan [ 159 ]. Income elasticities should be interpreted as if the long-run elasticity is less than the short-run elasticity suggesting that the country has reduced its emissions with income growth and consequently further proving the existence of the EKC [ 119 , 159 ].

Besides multicollinearity, there are more econometric issues associated with the modelling of the EKC. Hasanov et al. [ 160 ] extensively identify the major econometric issues in the EKC literature and provide a full mathematically and empirically explanation for each one. The authors mainly focused on the following issues: functional specifications used, and the econometric techniques employed; the use of a trend in the specification and level versus logarithmic variables; and the monotonic, quadratic, cubic, and quartic potential relationship between income and environmental degradation. In light of this, the authors developed a modelling strategy that should ensure a consistent approach when assessing the EKC. The Green Solow Model might be useful to overcome the sensitivity of the EKC to the econometric models employed.

Besides the critical issues identified in the EKC assessment and strategies to overcome them, from the extensive EKC literature arises the challenge: what are the EKC assessment gaps? Throughout the EKC literature, it has been noted that essential components for environmental degradation may not have been taken into consideration, such as consumption instead of production and technological progress. A criticism made of the EKC arises because it does not take into account the evolution of consumption coinciding with economic growth. That is, the EKC only explains how the process of production is converted into something environmentally friendly as a consequence of economic growth [ 10 ]. Besides, according to Kaika and Zervas [ 161 ], the EKC only focuses on domestic production and overlooks the impact of the consumption of imported goods on the environment. In turn, the income elasticity of demand for dirty goods has been disregarded by the EKC [ 40 ]. If, with high income levels, the demand for dirty goods persists, then this situation will lead to developed countries importing these goods from developing economies to satisfy demand. Consequently, any environmental improvement resulting from technological progress will be offset, and economic growth will result in more environmental degradation. This is a critical issue that goes against one of the basic assumptions of the EKC, that is, that there is a change in consumer behaviour with a rise in income.

Technological progress is a crucial tool to help with climate change and global warming mitigation. EKC supporters believe that environmental mitigation depends on technological progress and improvement. Therefore, they believe that only if technology and investment in the environment persist stagnantly, then the enlargement of economic activities harms the environment. However, the fact that technology enhances environmental quality is an ambition in dynamic economies, making economic growth a tool to accomplish environmental quality instead of being a threat [ 10 ]. The EKC hypothesis assumes that rising income induces technological improvements and environmental awareness, and consequently, it should safeguard the improvement of environmental performance in the later stages of economic growth. At this point surfaces the doubt if technological advancements can outrun the worrying pace of environmental damage. Gill, Viswanathan and Hassan [ 10 ] present a summary of examples of recent technological improvements in diverse sectors. Therefore it is crucial to include technological progress in the EKC assessment. Currently, technological progress is focused on environmental research, and technological progress indexes have been developed [ 162 , 163 ]. Also, the Green Solow Model could be a valuable tool to analyse technological improvements as it incorporates technological progress and resorts to enunciating the two conditions needed to guarantee sustainable growth.

Technological improvement encourages the replacement of obsolete technology and drives intensive pollutant economies to efficient ones. These changes mean that economies that reduce their energy consumption per unit of economic output consequently reduce emissions. The level of technology is vastly different in developing economies compared with developed, and therefore the developing economies are more carbon-intensive. According to Beckerman [ 15 ], only prosperous economies have the means to access environmentally friendly technologies in order to mitigate environmental degradation, while the poor economies, according to the author, are “too poor to be green”. However, when a developed or prosperous economy appears to be able to achieve economic growth without significant environmental impact, then it is necessary to look deeper into the location of that economy's pollutant industries. Developing economies' lax environmental regulations attract the relocation of pollutant industries from developed economies where there is pressure to accomplish environmental targets.

The relocation of pollutant industries separates production from consumption, which allows an increase in GDP with a reduction in emissions as the emissions are being emitted in another country. Therefore, the relocation of emissions-intensive industries could create an illusion that economies are becoming efficient and that they perform the EKC trajectory, and it deserves to be taken into consideration in the analysis of environmental performance. The analysis of technological progress (or energy efficiency index) could be useful in order to observe if any emissions reduction actually comes from the replacement with efficient technology or if their emissions are reducing through the relocation of their pollutant industries. A sectoral analysis or a joint analysis of the home with the host country (the country that receives the industry) could overcome the illusion of emissions reduction when they are just being relocated. A sectoral analysis could reveal from which sector the industries were relocated. In turn, an analysis of the home and host country could reveal the reduction of emissions in the home country that consequently represents an increase in the emissions of the host country.

Nevertheless, in order to achieve a significant reduction in environmental problems and produce more economic growth with less environmental damage, the technology improvements have to be huge. Considering that energy consumption is the main source of emissions, energy efficiency through efficient technology may not be enough to achieve meaningful emissions reduction. In light of this, the studies should focus on how economic growth can promote energy transition from fossil energy use to renewable energy use. Energy transition is currently a hot topic due to the global environmental agenda, and throughout the literature, energy transition indicators have been rising in importance [ 164 , 165 ]. It is critical to be concerned about the shortcoming of the storage of renewable energy sources. Scenarios with nuclear may be analysed as an intermediate process, but they must take into account the risks associated with this energy source. Producing more economic growth with less environmental damage is crucial to accomplishing the Sustainable Development Goals, also known as Agenda 2030. Global economic growth patterns are considered responsible for the issue of rising climatic disasters across the globe [ 166 ], and reducing global emissions and moving toward decarbonization is urgent. However, sustainable development goals accomplishment requires large investment needs. The COP26 was focused on tools to reduce climate change issues, such as green and climate finance. Green and climate finance could be powerful tools to bring about adaptation and mitigation of climate change issues mainly in countries with capital in short supply and could have a meaningful influence on the assessment of the EKC path. Green finance and climate finance have recently started to be addressed in the literature about the environment [ 167 , 168 ].

As the doubt if technological progress can slow down and reverse the pace of environmental damage grows, there is also the increasing uncertainty of what will happen to the environmental degradation already in place. Is it repairable? According to the EKC relationship, after achieving a certain level of income, the turning point, environmental degradation starts to decrease with growing income. However, can the later stages of economic development really repair the environmental damage of the first stages? The EKC hypothesis assumes that in the later stages of economic growth, the environmental damage as a consequence of economic growth can be reversed. However, this assumption is an object of criticism by various researchers. The ability to reverse environmental damage might be effective for specific air and water pollutants but might not work with things like carcinogenic chemicals, as they are considered irreparable [ 24 ]. Furthermore, environmental damage because of industrialization is also extraordinarily complex to overturn.

Global warming is considered the most critical environmental problem humanity has ever faced. Therefore, it is crucial to understand if the environmental damage provoked by economic growth can be repaired through more economic growth. At the moment, it is not enough to only analyse if the country or a group of countries perform the EKC trajectory. It is now necessary to start exploring how the environmental damage provoked in the first phase of the EKC can be repaired. In line with this, also arises the doubt if the growth path traced by the inverted U-shaped is efficient, a Pareto efficient. The Pareto efficient, or optimal, defines the optimum resource allocation at which it is not possible to reallocate in order to benefit or improve a specific resource allocation without harming the allocation of others. The EKC hypothesis transmits the message of ‘grow now and clean later’ [ 10 ]. This growth strategy is highly intensive in resources, which makes it incompatible with being Pareto efficient.

Considering that the EKC is not Pareto efficient and that the growth path is highly resource-intensive, it is highly likely that the environmental damage provoked in the first phases of the EKC might not be repairable. A growth path that takes care of both economic development and the environment simultaneously could avoid substantial losses, avoid the huge environmental impact of economic development, and the percentage of economic growth that in the future will be necessary to repair the environmental damage provoked before. On the one hand, a growth path that takes care of the environment in the early stages of economic growth could represent a global GDP loss of 1%. On the other hand, a loss in order of 5–20% of global GDP is a consequence of the absence of environmental care [ 169 ]. Therefore, environmental protection throughout all the stages of economic development could diminish environmental and GDP losses.

5. Conclusion

Motivated by the will to develop a useful EKC's research tool/guide that allows EKC researchers to learn from the origins and framework of the EKC until the evolution of the literature, gaps, econometric issues, and improvements needs, this present paper fulfils a gap in the EKC literature by providing a detailed and comprehensive description of the EKC framework, an extensive contextualization of the EKC evolution and literature, and a critical analysis. With various novelty aspects, this research strives to enlarge the knowledge of the EKC field. The main contribution of this review article consists of providing an extremely detailed description of the literature and evolution of the EKC analysis. Through the analysis of more than 200 articles from 1998 to 2022, a considerable number of EKC relationships analysed in the literature are provided, along with additional variables included in EKC estimations and methodologies used. Furthermore, each detail of the EKC assessment for each one of the more than 200 articles supplied allows researchers to find specific information to support their analysis.

The knowledge and assessment of the EKC has developed noticeably since its inception. However, despite being broadly assessed within the vast literature, the EKC hypothesis possesses some gaps and econometric issues, and improvement needs are verified. The absence of consensus throughout the EKC literature on the existence and shape of the curve has given rise to doubts about econometric issues in EKC modelling. The same geographic region, country or countries can generate opposing arguments regarding the existence and shape of the EKC. Throughout the literature review, evidence has been provided regarding the sensitivity of the EKC estimation resulting from the data set used, the indicators, the type of analysis (time series or cross-sectional), the methodology applied, and additional variables included. In light of this, the econometric issues are mainly associated with the functional specifications and econometric techniques used and the use of income quadratic, cubic and quartic variables. The use of econometric methods which deal with collinearity and multicollinearity is crucial. Non- or semi-parametric methods assessing short- and long-run elasticities could avoid these phenomena. Also, non-econometric methods instead of econometric could be helpful to avoid EKC sensitivity to the approach used.

Besides improvements needed for the econometric procedure, further improvements should be made to fill gaps in the EKC analysis. Knowledge of the EKC needs to be improved by integrating insights from other disciplines and research areas. These could include the inclusion of certain socio-political indicators that can influence efforts to improve environmental quality, such as research and development of alternative energy sources; economic complexity; economic uncertainty; economic, cultural, and political shocks; corruption; and political cooperation. All of these would be beneficial for economic analysis and policy recommendations. The complexity of environmental degradation issues is increasing, and scenarios such as relocated pollution, delocalized production, energy and production goods countries' dependence, lax environmental regulation, and comparative advantages, among others, can influence the environmental performance of countries. The relocation of pollutant industries could produce a result that does not fit with reality. The relocation that comes from lax environmental regulation from the host country, comparative advantages from home countries over the host ones that result in relocated pollution, delocalized production, and a countries’ dependence increases environmental degradation in the host countries while the goods produced are consumed in the home country. At this point, it is crucial to consider these scenarios during the EKC assessment in order to allow policymakers to develop and implement fair and effective policies conducive to the achievement of the sustainable development goals. Furthermore, policies should be developed in order to dissolve the strong disparity in environmental regulation between developed and developing countries, allowing an EKC analysis to be more reliable.

COP 26, five years apart from the Paris agreement, was the time for countries to strengthen their environmental commitment and goals. Achieving decarbonisation is urgent and requires the collaboration of all countries all over the world. In light of this, as widely implemented to assess environmental performance, the EKC assessment should be demanding. Analysing environmental degradation indicators or other indicators over the economic growth is not enough at this point; it is crucial to look further into environmental pollution indicators. The standard approach to global warming, which mainly consists of alleviating the restrictions on economic growth while supporting continuous technological development thought suitable to compensate for environmental damage, may be critical. A set of strategies and tools have been developed, and they should be included in the assessment of the path of environmental pollution over economic development. Technological progress, energy efficiency, energy transition, potential clean energy sources (such as nuclear), environmental regulation, and green and climate finance can influence this path and provide a realistic route to a cleaner environment. In light of this, the inclusion of these tools and drivers of environmental quality in the EKC analysis may allow policymakers to develop particular policies and measures in order to encourage their progress and improvement towards sustainability. A major concern of environmental mitigation is the economic growth path; however, achieving sustainable and low-carbon development may, under reasonable conditions, operate as an explicit contributing component to growth.

The eminent consequences of global warming place policymakers as central players in the current global discussion of climate change challenges. The EKC is extensively used to evaluate the environmental performance of economies, and several policy recommendations have already been proposed based on its analysis. However, policymakers must be aware of the high volatility and sensitivity of the EKC outcomes. Panel data analysis could produce strong limitations for the development of policies. Specific policies should not be designed for a particular country based on a panel data analysis where the country is inserted when it is not mandatory that the country follows the trajectory that the group follows as a whole. Therefore, the outcomes of the panel data analysis should only be considered as a reference for how those countries behave together under the same conditions and in a specific scenario. Considering the sensitivity of the EKC outcomes, the EKC might be considered as an environmental performance indicator on policy design and implementation, but as a reference and not as a decisive indicator by itself.

The present study provides a comprehensive and detailed picture of the EKC field. The development of this study faced some limitations, which are necessary to be highlighted to improve future research. Each EKC shape obtained is unique, considering things such as the functional specification, econometric methods used, and sample. Consequently, with a very extensive literature on the field, a huge limitation was faced in identifying all the issues related to the EKC assessment, mainly econometric issues. Moreover, developing a review article, mainly a particularly detailed one, is unending research, considering the tremendous amount of research on the field and that it is constantly growing. At this point, the length of the article could be a limitation. For the future direction of research, it could be relevant to explore methods beyond econometric methods that could fulfil the EKC assessment and overcome the identified econometric issues. Also, it could be useful to further investigate the influence of the additional variables on the EKC assessment. Additionally, providing individual reports about each strand of the Kuznets curve, beyond the Environmental, could be a relevant contribution to the literature.

Declarations

Author contribution statement.

All authors listed have significantly contributed to the development and the writing of this article.

Funding statement

Professor António Cardoso Marques and Patrícia Hipólito Leal were supported by Fundação para a Ciência e a Tecnologia [UIBD/04630/2020 & DFA/BD/6026/2020].

Data availability statement

Declaration of interests statement.

The authors declare the following conflict of interests: The author António Cardoso Marques is Associate Editor of Heliyon - Energy Section.

Additional information

No additional information is available for this paper.

Appendix A. Supplementary data

The following is the supplementary data related to this article:

Re-examining the Environmental Kuznets Curve in MENA Countries: Is There Any Difference Using Ecological Footprint and CO 2 Emissions?

• Published: 27 April 2024

Cite this article

• Hicham Ayad   ORCID: orcid.org/0000-0003-1624-3456 1 ,
• Mohd Shuaib 2 ,
• Md. Emran Hossain 3 ,
• Mohammad Haseeb 4 ,
• Mustafa Kamal 5 &
• Masood ur Rehman 6  

Despite the significant research on environmental issues, there has not been considerable investigation on the environmental Kuznets curve (EKC) hypothesis in the MENA nations using both CO 2 emissions (CO 2 e) and ecological footprint (EF) environmental indicators in the same setting. Therefore, the primary goal of this research is to re-examine the EKC hypothesis in 18 MENA nations from 1990 to 2018 using a panel data model labeled the pooled mean group autoregressive distributed lags (PMG-ARDL). We employ CO 2 e and EF as environmental quality proxies to deal with all types of pollutants, not only air pollution, as well as the demand side of environmental assets. The outcomes of the econometric study revealed the absence of the EKC hypothesis using CO 2 e in contrast to EF where the hypothesis is held in the entire group and both oil-exporting and importing countries. Moreover, population and non-renewable energy significantly harm the environment in the three sub-sample groups of countries in this study. This is what requires the governments of these countries to strive more toward achieving environmental sustainability by preserving the environmental capacity of the region and reducing emissions.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Availability of Data and Materials

Data associated with this submission can be available from the corresponding author upon reasonable request.

Huang, Y., Villanthenkodath, M. A., & Haseeb, M. (2023). The nexus between eco‐friendly technology and environmental degradation in India: Does the N or inverted N‐shape load capacity curve (LCC) hypothesis hold? Natural Resources Forum. Oxford, UK: Blackwell Publishing Ltd.

Google Scholar  

Struckmann, C. (2018). A postcolonial feminist critique of the 2030 Agenda for sustainable development: A South African application. Agenda, 32 (1), 12–24.

Article   Google Scholar  

Alnour, M., Awan, A., & Hossain, M. E. (2024). Towards a green transportation system in Mexico: The role of renewable energy and transport public private partnership to curb the emissions. Journal of Cleaner Production,  140984.

Kibria, M. G., Aspy, N. N., Ullah, E., Dewan, M. F., Hasan, M. A., Hossain, M. A., & Hossain, M. E. (2023). Quantifying the effect of agricultural greenhouse gas emissions, food production index, and land use on cereal production in South Asia. Journal of Cleaner Production, 432 , 139764.

Article   CAS   Google Scholar  

Ayad, H., Abbas, S., Nakhli, M. S., Jibir, A., & Shahzad, U. (2023a). Industrial growth, health care policy uncertainty and carbon emissions: Do trade and tax policy uncertainties matter for sustainable development in the USA? Structural Change and Economic Dynamics, 66 , 151–160.

Caglar, A. E. (2023). Can nuclear energy technology budgets pave the way for a transition toward low-carbon economy: Insights from the United Kingdom. Sustainable Development, 31 (1), 198–210.

Caglar, A. E., & Ulug, M. (2022). The role of government spending on energy efficiency R&D budgets in the green transformation process: Insight from the top-five countries. Environmental Science and Pollution Research, 29 (50), 76472–76484.

Caglar, A. E., & Yavuz, E. (2023). The role of environmental protection expenditures and renewable energy consumption in the context of ecological challenges: Insights from the European Union with the novel panel econometric approach. Journal of Environmental Management, 331 , 117317.

Caglar, A. E., Guloglu, B., & Gedikli, A. (2022a). Moving towards sustainable environmental development for BRICS: Investigating the asymmetric effect of natural resources on CO2. Sustainable Development, 30 (5), 1313–1325.

Joof, F., Samour, A., Ali, M., Tursoy, T., Haseeb, M., Hossain, M. E., & Kamal, M. (2023). Symmetric and asymmetric effects of gold, and oil price on environment: The role of clean energy in China. Resources Policy, 81 , 103443.

Al-Otaibi, G. (2015). By the numbers: Facts about water crisis in the Arab world.  The World Bank .

Alharthi, M., Dogan, E., & Taskin, D. (2021). Analysis of CO2 emissions and energy consumption by sources in MENA countries: Evidence from quantile regressions. Environmental Science and Pollution Research, 28 (29), 38901–38908.

Caglar, A. E., & Mert, M. (2022). Carbon hysteresis hypothesis as a new approach to emission behavior: A case of top five emitters. Gondwana Research, 109 , 171–182.

Shi, L., Udemba, E. N., Emir, F., Khan, N. U., Hussain, S., & Boukhris, I. (2023). Mediating role of finance amidst resource and energy policies in carbon control: A sustainable development study of Saudi Arabia. Resources Policy, 82 , 103521.

Rees, W. E. (2018). Ecological footprints and appropriated carrying capacity: What urban economics leaves out. In  The Earthscan Reader in Rural–Urban Linkages,  285–297. Routledge.

Shahbaz, M., Nuta, A. C., Mishra, P., & Ayad, H. (2023). The impact of informality and institutional quality on environmental footprint: The case of emerging economies in a comparative approach. Journal of Environmental Management, 348 , 119325.

Caglar, A. E., Yavuz, E., Mert, M., & Kilic, E. (2022b). The ecological footprint facing asymmetric natural resources challenges: Evidence from the USA. Environmental Science and Pollution Research, 1–14.

Hossain, M. R., Rana, M. J., Saha, S. M., Haseeb, M., Islam, M. S., Amin, M. R., & Hossain, M. E. (2023). Role of energy mix and eco-innovation in achieving environmental sustainability in the USA using the dynamic ARDL approach: Accounting the supply side of the ecosystem. Renewable Energy, 118925.

Panayotou, T. (1993). Empirical tests and policy analysis of development.  ILO Technology and Employment Programme Working Paper, WP238 .

Dai, J., Ahmed, Z., Alvarado, R., & Ahmad, M. (2023). Assessing the nexus between human capital, green energy, and load capacity factor: Policymaking for achieving sustainable development goals. Gondwana Research .

Wackernagel, M., & Rees, W. (1998).  Our ecological footprint: Reducing human impact on the earth, 9 . New society publishers.

Grossman, G., Krueger, A. (1991). Environmental impacts of a North American free trade agreement. https://www.nber.org/papers/w3914 . Accessed 2 February 2023.

Frodyma, K., Papież, M., & Śmiech, S. (2022). Revisiting the environmental Kuznets curve in the European Union countries. Energy, 241 , 122899.

Dinda, S. (2004). Environmental Kuznets curve hypothesis: A survey. Ecological economics, 49 (4), 431–455.

Murshed, M., Haseeb, M., & Alam, M. (2022). The environmental Kuznets curve hypothesis for carbon and ecological footprints in South Asia: The role of renewable energy. GeoJournal, 87 (3), 2345–2372.

Fakher, H. A., & Inglesi-Lotz, R. (2022). Revisiting environmental Kuznets curve: An investigation of renewable and non-renewable energy consumption role. Environmental Science and Pollution Research, 29 (58), 87583–87601.

Borghesi, S., & Ticci, E. (2019). Climate change in the MENA region: Environmental risks, socioeconomic effects and policy challenges for the future. MED,  289–292.

Udemba, E. N., & Alola, A. A. (2022). Asymmetric inference of carbon neutrality and energy transition policy in Australia: The (de) merit of foreign direct investment. Journal of Cleaner Production, 343 , 131023.

Dasgupta, S., Laplante, B., Wang, H., & Wheeler, D. (2002). Confronting the environmental Kuznets curve. Journal of Economic Perspectives, 16 (1), 147–168.

Grossman, G. M., & Krueger, A. B. (1995). Economic growth and the environment. The Quarterly Journal of Economics, 110 (2), 353–377.

Arouri, M. E. H., Youssef, A. B., M’henni, H., & Rault, C. (2012). Energy consumption, economic growth and CO2 emissions in Middle East and North African countries. Energy Policy, 45 , 342–349.

Zaied, Y. B., Cheikh, N. B., & Nguyen, P. (2017). Long-run analysis of environmental Kuznets curve in the Middle East and North Africa. Environmental Economics, 8 (4), 72.

Ozcan, B. (2013). The nexus between carbon emissions, energy consumption and economic growth in Middle East countries: A panel data analysis. Energy Policy, 62 , 1138–1147.

Farhani, S., & Shahbaz, M. (2014). What role of renewable and non-renewable electricity consumption and output is needed to initially mitigate CO2 emissions in MENA region? Renewable and Sustainable Energy Reviews, 40 , 80–90.

Bibi, F., & Jamil, M. (2021). Testing environment Kuznets curve (EKC) hypothesis in different regions. Environmental Science and Pollution Research, 28 (11), 13581–13594.

Ayad, H., Sari-Hassoun, S. E., Usman, M., & Ahmad, P. (2023). The impact of economic uncertainty, economic growth and energy consumption on environmental degradation in MENA countries: Fresh insights from multiple thresholds NARDL approach. Environmental Science and Pollution Research, 30 (1), 1806–1824.

Gorus, M. S., & Aslan, M. (2019). Impacts of economic indicators on environmental degradation: Evidence from MENA countries. Renewable and Sustainable Energy Reviews, 103 , 259–268.

Charfeddine, L., & Mrabet, Z. (2017). The impact of economic development and social-political factors on ecological footprint: A panel data analysis for 15 MENA countries. Renewable and Sustainable Energy Reviews, 76 , 138–154.

Youssef, A. B., Boubaker, S., & Omri, A. (2020). Financial development and macroeconomic sustainability: Modeling based on a modified environmental Kuznets curve. Climatic Change, 163 (2), 767–785.

Usman, O., Rafindadi, A. A., & Sarkodie, S. A. (2021). Conflicts and ecological footprint in MENA countries: Implications for sustainable terrestrial ecosystem. Environmental Science and Pollution Research, 28 (42), 59988–59999.

Djedaiet, A., Ayad, H., & Ben-Salha, O. (2024). Oil prices and the load capacity factor in African oil-producing OPEC members: Modeling the symmetric and asymmetric effects. Resources Policy, 89 , 104598.

The World Bank. (2021). Middle East and North Africa human capital plan. Available at https://thedocs.worldbank.org/en/doc/907071571420642349-0280022019/original/HCPMiddleEastPlanOct19.pdf . Accessed on 03/07/2023.

Pesaran, M. H., & Yamagata, T. (2008). Testing slope homogeneity in large panels. Journal of econometrics, 142 (1), 50–93.

Pesaran, M. H. (2004). General diagnostic tests for cross section dependence in panels. Cambridge Working Papers Econ, 69 (7), 1240.

Baltagi, B. H., Feng, Q., & Kao, C. (2012). A Lagrange multiplier test for cross-sectional dependence in a fixed effects panel data model. Journal of Econometrics, 170 (1), 164–177.

Pesaran, M. H. (2007). A simple panel unit root test in the presence of cross-section dependence. Journal of applied econometrics, 22 (2), 265–312.

Westerlund, J. (2007). Testing for error correction in panel data. Oxford Bulletin of Economics and Statistics, 69 (6), 709–748.

Shahbaz, M., Balsalobre-Lorente, D., & Sinha, A. (2019). Foreign direct Investment–CO2 emissions nexus in Middle East and North African countries: Importance of biomass energy consumption. Journal of Cleaner Production, 217 , 603–614.

Bosah, C. P., Li, S., Ampofo, G. K. M., & Liu, K. (2021). Dynamic nexus between energy consumption, economic growth, and urbanization with carbon emission: Evidence from panel PMG-ARDL estimation. Environmental Science and Pollution Research, 28 (43), 61201–61212.

Pesaran, M. H., Schuermann, T., & Weiner, S. M. (2004). Modeling regional interdependencies using a global error-correcting macroeconometric model. Journal of Business & Economic Statistics, 22 (2), 129–162.

Baltagi, B. H., & Pesaran, H. M. (2007). Heterogeneity and cross section dependence in panel data models: Theory and applications introduction. Journal of Applied Econometrics, 22 (2), 229–232.

Caglar, A. E., & Askin, B. E. (2023). A path towards green revolution: How do competitive industrial performance and renewable energy consumption influence environmental quality indicators? Renewable Energy, 205 , 273–280.

Li, W., Qiao, Y., Li, X., & Wang, Y. (2022). Energy consumption, pollution haven hypothesis, and environmental Kuznets curve: Examining the environment–economy link in belt and road initiative countries. Energy, 239 , 122559.

Jebli, M. B., & Youssef, S. B. (2015). The environmental Kuznets curve, economic growth, renewable and non-renewable energy, and trade in Tunisia. Renewable and Sustainable Energy Reviews, 47 , 173–185.

Ben Jebli, M., Ben Youssef, S., & Ozturk, I. (2015). The role of renewable energy consumption and trade: Environmental Kuznets curve analysis for Sub-Saharan Africa countries. African Development Review, 27 (3), 288–300.

Udemba, E. N., Emir, F., & Philip, L. D. (2022). Mitigating poor environmental quality with technology, renewable and entrepreneur policies: A symmetric and asymmetric approaches. Renewable Energy, 189 , 997–1006.

Koondhar, M. A., Udemba, E. N., Cheng, Y., Khan, Z. A., Koondhar, M. A., Batool, M., & Kong, R. (2021). Asymmetric causality among carbon emission from agriculture, energy consumption, fertilizer, and cereal food production–a nonlinear analysis for Pakistan. Sustainable Energy Technologies and Assessments, 45 , 101099.

Ayad, H., Lefilef, A., & Ben-Salha, O. (2023). A revisit of the EKC hypothesis in top polluted African countries via combining the ARMEY curve into the Kuznets curve: a Fourier ARDL approach. Environmental Science and Pollution Research, 30 (33), 81151–81163.

Dong, K., Dong, X., & Dong, C. (2019). Determinants of the global and regional CO2 emissions: What causes what and where? Applied Economics, 51 (46), 5031–5044.

Ansari, M. A. (2022). Re-visiting the environmental Kuznets curve for ASEAN: A comparison between ecological footprint and carbon dioxide emissions. Renewable and Sustainable Energy Reviews, 168 , 112867.

Rashid, A., Irum, A., Malik, I. A., Ashraf, A., Rongqiong, L., Liu, G., ... & Yousaf, B. (2018). Ecological footprint of Rawalpindi; Pakistan’s first footprint analysis from urbanization perspective. Journal of Cleaner Production, 170 , 362–368.

Usman, A., Ozturk, I., Naqvi, S. M. M. A., Ullah, S., & Javed, M. I. (2022). Revealing the nexus between nuclear energy and ecological footprint in STIRPAT model of advanced economies: Fresh evidence from novel CS-ARDL model. Progress in Nuclear Energy, 148 , 104220.

Ulucak, R., & Bilgili, F. (2018). A reinvestigation of EKC model by ecological footprint measurement for high, middle- and low-income countries. Journal of Cleaner Production, 188 , 144–157.

Al-Mulali, U., & Ozturk, I. (2015). The effect of energy consumption, urbanization, trade openness, industrial output, and the political stability on the environmental degradation in the MENA (Middle East and North African) region. Energy, 84 , 382–389.

Nathaniel, S., Anyanwu, O., & Shah, M. (2020). Renewable energy, urbanization, and ecological footprint in the Middle East and North Africa region. Environmental Science and Pollution Research, 27 (13), 14601–14613.

Ayad, H., Lefilef, A., Jahanger, A., & Balsalobre-Lorente, D. (2023d). Reinvestigate the effect of foreign direct investment on environmental quality in the MENA region: Is there any difference using the load capacity factor? Journal of the Knowledge Economy, 1–31.

Download references

Author information

Authors and affiliations.

LEPPESE Laboratory, University Centre of Maghnia, Maghnia, Algeria

Hicham Ayad

School of Economics and Management, and Center for Industrial Economics, Wuhan University, Wuhan, 430072, China

Mohd Shuaib

Department of Agricultural Sciences, Texas State University, San Marcos, TX, 78666, USA

Md. Emran Hossain

Department of Management Studies, Graphic Era Deemed to be University, Dehradun, Uttarakhand, 248002, India

Mohammad Haseeb

Department of Basic Sciences, College of Science and Theoretical Studies, Saudi Electronic University, Dammam, 32256, Saudi Arabia

Mustafa Kamal

Department of Information Technology, College of Computing and Informatics, Saudi Electronic University, Dammam, 32256, Saudi Arabia

Masood ur Rehman

You can also search for this author in PubMed   Google Scholar

Contributions

Hicham Ayad and Mohd Shuaib conceptualized and wrote the manuscript. Md. Emran Hossain performs data curation and formal analysis. Mohammad Haseeb, Mustafa Kamal, and Masood ur Rehman wrote the original manuscript and reviewed and edited the manuscript. All authors reviewed the manuscript.

Corresponding author

Correspondence to Md. Emran Hossain .

Ethics declarations

Ethical approval.

Not applicable.

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

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Ayad, H., Shuaib, M., Hossain, M. et al. Re-examining the Environmental Kuznets Curve in MENA Countries: Is There Any Difference Using Ecological Footprint and CO 2 Emissions?. Environ Model Assess (2024). https://doi.org/10.1007/s10666-024-09977-7

Download citation

Received : 17 May 2023

Accepted : 18 April 2024

Published : 27 April 2024

DOI : https://doi.org/10.1007/s10666-024-09977-7

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

• Environmental Kuznets curve
• MENA region
• Ecological footprint
• Oil exporting nations
• Find a journal
• Publish with us
• Track your research