research paper on environmental economics

An official website of the United States government

Here’s how you know

Official websites use .gov A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS A lock ( Lock A locked padlock ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

JavaScript appears to be disabled on this computer. Please click here to see any active alerts .

Research in Environmental Economics - NCEE Working Paper Series

EPA's National Center for Environmental Economics (NCEE) publishes a working paper series on research in environmental economics. Paper topics include environmental management, resources and conservation, agriculture, global issues, institutional issues, and other topics. These papers are either authored by NCEE economists or produced with funding from NCEE.

The working papers are distributed for purposes of information and discussion. The opinions and findings expressed in these papers are those of the authors and do not necessarily represent those of EPA or NCEE. NCEE is committed to maintaining a high standard of quality in the NCEE Working Paper Series, and each paper must receive a favorable peer review prior to publication in the series. See Working Paper Series Instructions .

Any questions regarding the NCEE Working Paper Series can be addressed to the Managing Editor, Cynthia Morgan ( [email protected] ).

• Environmental Economics Home
• Overview of Environmental Economics
• Current EPA Guidelines for Preparing Economic Analyses (2016)
• Environmental Economics Reports
• Current Environmental Economic Topics
• EPA Datasets
• EPA Handbook on Land Cleanup and Reuse
• Peer Review of TSD: Social Cost of Greenhouse Gas Estimates (2023)
• Retrospective Study of the Costs of EPA Regulations

Environmental Economics, Regulation, and Innovation

This paper provides a primer on the economics of environmental innovation. Our intention is not to write a pure review paper, but to also provide an up-to-date textbook treatment on the issue. Thus, we start by defining the marginal costs of both emissions and of emissions abatement. We then analyze theoretically how innovation may affect marginal abatement costs. We also cover the different modelling choices with respect to how the innovation process is represented mathematically and how different environmental policy measures could affect environmental innovation. Our theoretical propositions are all illustrated with examples from the empirical literature. A special emphasis is placed on the recent literature on directed technical change and the potential impact of government intervention in the research and development choices of private firms.

A revised version of this paper will appear in the forthcoming Handbook on Innovation and Regulation by Edward Elgar Publishing We thank Pontus Braunerhjelm, editor of the Handbook, for excellent comments and many good suggestions on earlier drafts. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.

MARC RIS BibTeΧ

Download Citation Data

More from NBER

In addition to working papers , the NBER disseminates affiliates’ latest findings through a range of free periodicals — the NBER Reporter , the NBER Digest , the Bulletin on Retirement and Disability , the Bulletin on Health , and the Bulletin on Entrepreneurship  — as well as online conference reports , video lectures , and interviews .

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
• Explore content
• About the journal
• Publish with us
• Sign up for alerts

Environmental economics articles within Nature Climate Change

Policy Brief | 28 March 2024

Global corporate tax competition leads to unintended yet non-negligible climate impacts

The worldwide trend of decreasing corporate tax in recent years has contributed to an increase in global carbon emissions, but implementing a global minimum tax rate of 15% could partially mitigate this impact. Policymakers should coordinate corporate tax policies with climate regulations.

• , Zengkai Zhang
•  &  Yi Lu

Perspective | 28 March 2024

Designing water markets for climate change adaptation

Global climate change will continue to reconfigure water resources and lead to more extreme events. Water markets may provide a low-cost adaptation tool. This Perspective discusses the opportunities and challenges for surface and groundwater markets to manage water resources.

• Ellen M. Bruno
•  &  Katrina Jessoe

Article | 28 March 2024

Global corporate tax competition challenges climate change mitigation

Countries use corporate tax cuts to attract foreign investment, which reshapes patterns of global production. This research shows that such competition will lead to higher carbon emissions and shift them to developing countries, while a global minimum tax could help alleviate these problems.

Article | 03 January 2024

The social costs of hydrofluorocarbons and the benefits from their expedited phase-down

Hydrofluorocarbons are a class of important greenhouse gases, and quantitative estimates of their social cost are still lacking. This research develops a set of direct estimates of their economic costs and shows their rapid phase-down could lead to large climate benefits.

• , Lisa Rennels
•  &  Bryan Parthum

Research Highlight | 03 November 2023

Disaster and capital stock

• Lingxiao Yan

Perspective | 30 October 2023

Realizing the social value of impermanent carbon credits

Nature-based solutions are essential to avoid climate crisis, yet how best to estimate their long-run effects is unclear. Here the authors propose a new dynamic accounting method that captures the impermanence of these carbon impacts, allowing investors to make robust comparisons across projects.

• Andrew Balmford
• , Srinivasan Keshav
•  &  Tom Swinfield

Research Highlight | 04 August 2023

Economics of adaptation

• Bronwyn Wake

Article | 17 July 2023

Co-firing plants with retrofitted carbon capture and storage for power-sector emissions mitigation

Coal–biomass co-firing power plants with retrofitted carbon capture and storage are seen as a promising decarbonization solution for coal-dominant energy systems. Framework with spatially explicit biomass sources, plants and geological storage sites demonstrate its effectiveness in China.

• Jing-Li Fan
• , Jingying Fu
•  &  Xi Lu

Analysis 06 July 2023 | Open Access

The historical social cost of fossil and industrial CO 2 emissions

Historical CO 2 emissions could lead to future climate damages and harm human inclusive wealth. This analysis proposes the concept of climate wealth borrowing and quantifies the country-specific present value of climate change impacts arising from energy and industrial CO 2 emissions of the period of 1950–2018.

• Wilfried Rickels
• , Felix Meier
•  &  Martin Quaas

Comment | 26 June 2023

Green premiums are a challenge and an opportunity for climate policy design

Adjusting green public support programmes to green premiums can reduce public spending, yet this is challenged by uncertainty. Underfunding green technologies can delay the green transition, and overfunding them can increase transition costs. Both risks of under- and overfunding can be reduced using responsive adjustments.

• Till Köveker
• , Olga Chiappinelli
•  &  Xi Sun

Policy Brief | 26 June 2023

State of global land regulation inadequate to control biofuel land-use-change emissions

Under current land-use regulation, carbon dioxide emissions from biofuel production exceed those from fossil diesel combustion. Therefore, international agreements need to ensure the effective and globally comprehensive protection of natural land before modern bioenergy can effectively contribute to achieving carbon neutrality.

• Leon Merfort
• , Nico Bauer
•  &  Elmar Kriegler

Research Highlight | 09 May 2023

Carbon taxes and inflation

News & Views | 04 May 2023

Principles of decarbonization politics

Climate change mitigation politics is not delivering on climate targets. Recent research suggests that a general formal framework that represents the behaviour of citizens, consumers, firms and parties explains why.

• Linus Mattauch
•  &  Sugandha Srivastav

News & Views | 23 March 2023

Strong climate action is worth it

An immediate and rapid reduction in global emissions is required for many reasons. Integrated research supports the economic case for strong near-term climate action, even before accounting for expected negative impacts on biodiversity, health and tipping points.

• Jarmo S. Kikstra
•  &  Paul Waidelich

Article | 23 March 2023

New damage curves and multimodel analysis suggest lower optimal temperature

Cost-benefit analysis of climate change depends heavily on the damage function used, and it is difficult to get credible information. Multimodel comparison with newly developed bottom-up damage functions indicates the optimal temperature could be much lower than previously estimated.

• Kaj-Ivar van der Wijst
• , Francesco Bosello
•  &  Detlef van Vuuren

Research Highlight | 08 March 2023

Trade barriers

Research Briefing | 23 February 2023

Flood risks are insufficiently priced into housing markets but better pricing would leave some worse off

We estimate that US residential properties exposed to flood risks are overvalued by close to US$200 billion. We discuss where these properties are located, and who stands to lose the most from housing prices that better reflect climate risks.

Article 16 February 2023 | Open Access

Unpriced climate risk and the potential consequences of overvaluation in US housing markets

Natural hazards exacerbated by climate change pose serious risks to property markets in the United States. Ignoring these risks could create instability in housing values. This research shows the magnitude of unpriced flood risk and who stands to lose from housing prices that reflect climate risks.

• Jesse D. Gourevitch
• , Carolyn Kousky
•  &  Joakim A. Weill

Research Highlight | 08 February 2023

India’s financial system

Article | 06 February 2023

Coal-exit alliance must confront freeriding sectors to propel Paris-aligned momentum

Coal phase-out is an irreplaceable part of the overall mitigation effort and bottom-up momentum has emerged to accelerate the process. With a new approach considering political feasibility, this research shows the potential spillover risks that may undermine the sectoral actions.

• Stephen L. Bi
•  &  Jessica Jewell

Research Highlight | 12 January 2023

Science-based targets

Research Highlight | 01 December 2022

Directed climate innovation

Policy Brief | 24 October 2022

Climate-proofing the National Flood Insurance Program

Reforms are required to maintain a healthy and robust flood insurance market under future climate conditions for the United States. Therefore, policymakers should implement premiums that reflect flood risk and incentivize household-level risk reduction, complemented with regional flood adaptation investments.

• Lars T. de Ruig
• , Toon Haer
•  &  Jeroen C. J. H. Aerts

Article | 24 October 2022

How the USA can benefit from risk-based premiums combined with flood protection

The National Flood Insurance Program is a key tool for managing growing flood risk in the USA. This research shows that premiums based on local risk, rather than national averages, will generate large societal benefits, and investments in large-scale adaptation infrastructure will enhance these impacts.

Comment | 05 September 2022

Pragmatic cost–benefit analysis for infrastructure resilience

Infrastructure investment needs to account for climate change globally, yet most day-to-day projects are small and poorly served by economic assessment processes. Four simple adjustments to cost–benefit analysis practices would greatly improve decision making for future infrastructure resilience.

• Russell M. Wise
• , Tim Capon
•  &  Mark Stafford-Smith

Research Highlight | 01 September 2022

Ranchers and dry season

Research Highlight | 02 August 2022

Impacts on tourism demand

• Carmen Dayrell

Research Highlight | 07 July 2022

Prices with quantities

Comment | 27 June 2022

Accounting for ecosystem service values in climate policy

Ecosystem services are often omitted from climate policy owing to difficulties in estimating the economic value of climate-driven ecosystem changes. However, recent advances in data and methods can help us overcome these challenges and move towards a more comprehensive accounting of climate impacts.

• Hannah Druckenmiller

Value wild animals’ carbon services to fill the biodiversity financing gap

Incorporating the carbon services of wild animals into financial markets has the potential to benefit both climate and conservation through the development of carbon offsets that are equitable and nature positive. However, for this paradigm to be successful, many challenges regarding science, finance and law still need to be overcome.

• Fabio Berzaghi
• , Thomas Cosimano
•  &  Ralph Chami

Research Highlight | 08 June 2022

Governance modes

Research Briefing | 26 May 2022

Fossil-fuel stranded asset risks held by individuals in OECD countries and non-OECD governments

A model based on plausible changes in expectations of future oil and gas demand identifies the ultimate financial owners of potential stranded assets to be predominantly OECD-based individual investors (through pension funds and shareholdings) and governments of non-OECD countries.

News & Views | 31 March 2022

Assessing nuclear phase-out

Controversy remains as to whether nuclear power should be part of a sustainable future energy mix. Phasing out nuclear too rapidly could lead to substantial health damages from air pollution.

• Sebastian Rausch

Research Highlight | 09 March 2022

Adaptation funding

Research Highlight | 08 February 2022

Firm emissions reduction

Policy Brief | 29 November 2021

Protecting the poor with a carbon tax and equal per capita dividend

We find that if all countries adopt the necessary uniform global carbon tax and then return the revenues to their citizens on an equal per capita basis, it will be possible to meet a 2 °C target while also increasing wellbeing, reducing inequality and alleviating poverty. These results indicate that it is possible for a society to implement strong climate action without compromising goals for equity and development.

• Mark Budolfson
• , Francis Dennig
•  &  Stéphane Zuber

Analysis | 29 November 2021

Climate action with revenue recycling has benefits for poverty, inequality and well-being

Climate policy analyses often ignore the possibility of progressive redistribution of carbon tax revenues and assume that mitigation cost will burden the poor in the short term. Integrated Assessment Model (IAM) estimation suggests such redistribution could reduce inequality, alleviate poverty and increase well-being globally.

News & Views | 22 November 2021

Risk transfers support adaptation

Finding effective ways to support rural communities in adapting to climate change is critical for building climate-resilient societies. Now research shows the potential of risk-transfer policies for improving adaptation and securing the livelihoods of smallholder farmers.

• Roman Hoffmann

Policy Brief | 28 October 2021

Freely available deforestation alerts can reduce emissions from land-use change

Subscriptions to a free, weekly deforestation alert system available on the simple interface Global Forest Watch reduced deforestation in the protected areas and logging concessions of tropical African forests. This suggests that freely available near-real-time forest monitoring systems can help reduce emissions from deforestation if they are integrated with forest policies.

• Fanny Moffette
• , Jennifer Alix-Garcia
•  &  Amy H. Pickens

To achieve deep cuts in US emissions, state-driven policy is only slightly more expensive than nationally uniform policy

Using a multi-sector model of human and natural systems, we find that the nationwide cost from state-varying climate policy in the United States is only one-tenth higher than that of nationally uniform policy. The benefits of state-led action — leadership, experimentation and the practical reality that states implement policy more reliably than the federal government — do not necessarily come with a high economic cost.

• , Gokul Iyer
•  &  David G. Victor

News & Views | 25 October 2021

The hidden benefits of high-speed rail

Improvements in public transport are often regarded as essential to combat climate change. A study investigating the Chinese high-speed rail system suggests that these benefits could operate through channels other than those that one might expect.

• Armin Schmutzler

Article | 25 October 2021

Impact of high-speed rail on road traffic and greenhouse gas emissions

Intercity high-speed rail (HSR) can have large climate benefits with its high energy efficiency. This study explores the substitution effects of HSR on road traffic in China, which can be translated to an annual reduction of 14.76 million tons of CO 2 -equivalent emissions.

•  &  Mandi Xu

Article | 07 October 2021

Trade-linked shipping CO 2 emissions

More than 80% of trade by volume occurs via maritime shipping, with growing pressure to reduce associated GHG emissions. The top 10 single-direction trade pairs account for nearly 20% of emissions; optimizing trade patterns could reduce emissions by 38% of current totals.

• Xiao-Tong Wang
•  &  Ke-Bin He

News & Views | 30 September 2021

Do climate dynamics matter for economics?

Economic models of climate change are the basis for climate policy design. However, incorrect representation of physical dynamics in these models could lead to biased advice.

• Richard S. J. Tol

Research Highlight | 01 September 2021

Information Services

Article | 23 August 2021

The surprisingly inexpensive cost of state-driven emission control strategies

The nationwide cost of cutting emissions can be affected by local policies. This study considers the differences across the US states, with integrated assessment model results showing that varying state policies only increases nationwide costs by about 10%.

Research Highlight | 05 August 2021

Climate amenities

Research Highlight | 06 July 2021

Indonesian palm oil

Article | 05 July 2021

The climate consistency goal and the transformation of global finance

Finance flows consistent with low-emission pathways and resilient development are central to the Paris Agreement, yet confusion remains about Article 2.1(c). Clarity is needed, particularly of its relationship with climate finance, to differentiate responsibility and enable suitable monitoring.

• Luis H. Zamarioli
• , Pieter Pauw
•  &  Hugues Chenet

Perspective | 17 June 2021

Integrated perspective on translating biophysical to economic impacts of climate change

Uncertainty in estimates of the economic impacts of climate change makes it difficult to evaluate the benefits of mitigation. This Perspective reviews methods for determining economic damages from biophysical impacts, highlights critical gaps and suggests priorities moving forward.

• Franziska Piontek
• , Laurent Drouet
•  &  Massimo Tavoni

Browse broader subjects

• Environmental social sciences

Quick links

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

• Faculty of Arts and Sciences
• FAS Theses and Dissertations
• Communities & Collections
• By Issue Date
• FAS Department
• Quick submit
• Waiver Generator
• DASH Stories
• Accessibility
• COVID-related Research

Terms of Use

• Privacy Policy
• By Collections
• By Departments

Essays in Environmental Economics

Citable link to this page

Collections.

• FAS Theses and Dissertations [6097]

Contact administrator regarding this item (to report mistakes or request changes)

Academia.edu no longer supports Internet Explorer.

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

•  We're Hiring!
•  Help Center

Environmental Economics

• Most Cited Papers
• Most Downloaded Papers
• Newest Papers
• Save to Library
• Last »
• Ecological Economics Follow Following
• Environmental Policy and Governance Follow Following
• Environmental Studies Follow Following
• Climate Change Follow Following
• Environmental Sustainability Follow Following
• Environmental Science Follow Following
• Economics Follow Following
• Energy Economics Follow Following
• Environmental Law Follow Following
• Resource Economics Follow Following

Enter the email address you signed up with and we'll email you a reset link.

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

Environmental migration? A systematic review and meta-analysis of the literature

• Original Paper
• Open access
• Published: 30 March 2024

Cite this article

You have full access to this open access article

• Maria Cipollina   ORCID: orcid.org/0000-0002-1454-4039 1 ,
• Luca De Benedictis 2 &
• Elisa Scibè 3  

This article provides a comprehensive quantitative overview of the literature on the relationship between environmental changes and human migration. It begins with a systematic approach to bibliographic research and offers a bibliometric analysis of the empirical contributions. Specifically, we map the literature and conduct systematic research using main bibliographic databases, reviews, and bibliometric analysis of all resulting papers. By constructing a citation-based network, we identify four separate clusters of papers grouped according to certain characteristics of the analysis and resulting outcomes. Finally, we apply a meta-analysis to a sample of 96 published and unpublished studies between 2003 and 2020, providing 3904 point estimates of the effect of slow-onset events and 2065 point estimates of the effect of fast-onset events. Overall, the meta-analytic average effect on migration is small for both slow- and rapid-onset events; however, it is positive and significant. Accounting for the clustering of the literature, which highlights how specific common features of the collected studies influence the magnitude of the estimated effect, reveals a significant heterogeneity among the four clusters of papers. This heterogeneity gives rise to new evidence on the formation of club-like convergence of literature outcomes.

Avoid common mistakes on your manuscript.

1 Introduction

In a world of changing climate and increasing occurrence of natural hazards, the role of environmental factors in shaping migration patterns has become a most debated topic within institutions and academia. As opposed to a simplistic vision of a general direct role of environmental factors in determining migration flows from environmentally stressed areas and regions hit by calamities, more complex scenarios have emerged, with analyses reporting different and sometimes opposite outcomes. This may not only be due to the intrinsic complexity of their extent and scale, but also to differences in specific characteristics of scientific contributions (International Organization for Migration, 2021 ).

The literature on the relationship between environmental factors and human mobility is characterized by heterogeneous findings: some contributions highlight the role of climate changes as a driver of migratory flows, while others underline how this impact is mediated by geographical, economic, and the features of the environmental shock. This paper aims to map the economic literature on these topics moving away from a classical literature review and offering a methodology that integrates three approaches in a sequence, in this way we believe that our contribution improves the existing literature on several dimensions. First, the analysis starts with systematic research of the literature through main bibliographic databases and collecting previous reviews and meta-analyses, followed by a review and bibliometric analysis of all resulting papers. This step produces a sample of 151 papers empirical and non-empirical contributions, spanning the last 20 years and focusing on different geographical areas, taking into account different socio-economic factors, applying different methodologies and empirical approaches to the analysis of slow-onset climatic events and/or fast-onset natural catastrophic events. Most importantly, the sample provides a variety of different outcomes on the impact of climatic changes and hazards on migration, revealing three main possible scenarios: (1) active role of environmental factors as a driver of migration; (2) environmental factors as a constraint to mobility; (3) non-significant role of environmental factors among other drivers of migration.

Second, to investigate the determinants of this extreme heterogeneity of outcomes, we postulate the assumption that the inter-connectivity of papers may play a role in shaping such different conclusions. Considering the ensemble of papers referenced by each contribution included in the sample, as a second step, we build a bibliographic coupling network, where papers are linked to each other according to the number of shared references. This citation-based method allows for the formation of a network of contributions in the literature space and highlights some potential common grounds among papers. We then run a community detection of the resulting network that produces four main clusters that gather papers together according to not only certain characteristics of the analysis but also resulting outcomes.

Finally, we use the clustered structure in the last step of the analysis: a Meta-Analysis (MA) to summarize and analyze all estimated effects of environmental variables on human mobility. The MA is a “quantitative survey" of empirical economic evidence on a given hypothesis, phenomenon, or effect, and provides a statistical synthesis of results from a series of studies (Stanley, 2001 ). The MA can be applied to any set of data and the synthesis will be meaningful only if the studies have been collected systematically (Borenstein et al., 2009 ). A highly significant result can be potentially considered as a consensual indication of the external validity of the correlation of the phenomena under scrutiny.

Therefore, from the original 151 paper we build - through a replicable process of screening, eligibility, and inclusion of contribution based on PRISMA guideline (see Fig. 1 ) - a unique dataset that synthesises the estimated coefficients of 96 empirical papers released between 2003 and 2020, published in academic journals, working papers series, or unpublished studies, providing 3904 point estimates of the effect of slow-onset events (e.g. climate change) and 2065 point estimates of the effect of fast-onset natural events(e.g. catastrophes) on different kinds of human mobility (international, domestic, and with a clear pro-urban directionality). Overall, the meta-analytic average effect estimates a small impact of slow- and rapid-onset variables on migration, however positive and significant. When the communities of papers are accounted for, however, a significant heterogeneity emerges among the four clusters of papers, giving rise to new evidence on the limits of a consensual effect of climatic shocks on permanent human displacement and the formation of club-like convergence of literature outcomes.

PRISMA Diagram. Note : PRISMA Diagram (Page et al., 2021a ) of identification, screening, eligibility and inclusion stages of academic contributions. The resulting sample is obtained through a search on Scopus, Web of Science, Google Scholar, IDEAS RePEc, and previous meta-analyses (Hoffmann et al., 2020 ; Beine & Jeusette, 2021 )

This is not the first MA on environmental migration. Concerning previous published reviews (Hoffmann et al., 2020 ; Sedova & Kalkuhl, 2020 ; Beine & Jeusette, 2021 ; Hoffmann et al., 2021 ) our article contributes and adds to the existing literature: (a) providing systematic research of the literature through main bibliographic databases, followed by a review and bibliometric analysis of all resulting papers; (b) building a citation-based network of contributions, that allows identifying four separate clusters of papers; (c) applying MA methods on a much larger sample of both micro- and macro-level estimates of environmental factors (slow- and fast-onset events) as a driver of migration (international and internal, including urbanization). Moreover, our overview highlights the role of the interconnectivity of studies in driving some main findings of the environmental migration literature.

Section 2 offers a systematic review of the literature and gives a detailed description of the data collection process; Sect. 3 analyses the structural characteristic of the network of the bibliographically coupled papers; Sect. 4 summarizes and discusses the results of the MA, finally, Sect. 5 concludes and offers some possible future extensions of the analysis.

2 Systematic review

This section reports the different phases of the systematic review. We do it schematically to facilitate the understanding of the proposed procedure.

Setting the boundaries of the literature This first step provides the most comprehensive sample of economic contributions on the relationship between climatic variations (and natural hazards) and human mobility, in all its different forms. We implement a systematic review aimed at mapping the body of literature and defining the boundaries of our focus. Systematic reviews have become highly recommended to conduct bibliographic overviews of specific literature because they provide a tool to report a synthesis of the state of the art of a field through a structured and transparent methodology (Page et al., 2021b ). To allow for comparability with previous MA and reviews, we also add to our sample all articles included in two recently published MA, Hoffmann et al. ( 2020 ) and Beine and Jeusette ( 2021 ) Footnote 1 . We begin with the definition of the research question and the main keywords, to gather and collect data in a sample of contributions. After the definition of inclusion and exclusion conditions, we proceed with a screening by title to exclude off-topic contributions and then to a screening of the text to assure the uniformity of contributions. The resulting sample is then the object of a preliminary bibliometric analysis.

Defining the research question and keywords The purpose of our systematic search is to collect all possible economic contributions to the impact of environmental factors on migration determinants. We define three keywords of the three phenomena under analysis:

climate change, as the most investigated environmental factor in the literature. The events connected to climate change are hereby intended as slow-onset events that gradually modify climatic conditions in the long run. We specifically focus on variations of temperature, precipitation, and soil quality (such as desertification, salinity, or erosion), factors that are not expected to cause an immediate and sudden expected impact, but slowly modify environmental conditions;

natural disasters, defined as fast-onset events that introduce a sudden shock (see Appendix Table 5 );

migration, which captures all possible patterns of human mobility, including within the borders of a country, which might be a potential response to environmental change. Most importantly, internal mobility includes also the process of urbanization of people moving out of rural areas to settle in cities.

Collecting data and initial search results To collect data we use two main literature databases, namely Scopus and Web of Science. Footnote 2 Exploiting the specific indexing and keyword definition of both sources, the search is run allowing for any kind of document type (articles in journals, book chapters, etc.) but limiting the area to economic literature in English. Footnote 3 The obtained sample only includes published documents, however since we perform a MA, it is important to take into account also non-published documents, as a way to control for a well-known publication bias in meta-analytic methodology (see Sect. 4 ). Therefore, we use the bibliographic database IDEAS, based on RePEc and dedicated to Economics, to include unpublished and working papers. Footnote 4 A selection of the contributions is made manually. Finally, to meet the purpose of comparability with other recent meta-analyses on the impact of environmental factors on migration, we also include all the contributions that have been reviewed in two main articles: Hoffmann et al. ( 2020 ) that provide a MA on 30 empirical papers focusing on country-level studies and Beine and Jeusette ( 2021 ) that review 51 papers and offer an investigation of the role of methodological choices of empirical studies (at any level) on the sign and magnitude of estimated results. Merging the results gives a sample of 203 records.

Screening of the results. We manually and meticulously screen the collected items through Scopus and Web of Science by title and we exclude papers on the migration of animals, plants, or other species, or focusing on topics different from human mobility (i.e. discrimination, crime, wars) or on the impact of environmental variables not corresponding to our definition of environmental factors (air pollution, mineral resources). All the papers in Beine and Jeusette ( 2021 ), Hoffmann et al. ( 2020 ) and those manually selected from IDEAS RePEc are automatically included in the sample with no concern of incoherence. The screening by title leads to the exclusion of 20 papers. The remaining 183 documents underwent a text screening process, which involved a careful and thorough reading of each paper to isolate eligible content. This stage leads to the removal of additional 32 documents covering on the one hand the analysis of the impact of environmental variables at destination countries (thus not focusing on their role on migration determinants at origin). We also exclude all the papers in which the dependent variable of the empirical exercise is not a measure of human mobility (i.e. remittances, poverty, wealth, employment, etc.). After duplicates removal, the sample results in 151 documents of different kinds: 35 records are non-empirical and contain an ensemble of literature reviews, qualitative analysis, theoretical modeling, and policy papers; 116 records are categorized as empirical, in which the dependent variable is a measure of human mobility and at least one environmental variable is an independent variable.

The PRISMA flow diagram (Moher et al., 2009 ) in Fig. 1 shows the process of identification, screening, eligibility, and inclusion of contributions in the final sample. It is important to note that there are two levels of inclusion: the first level identifies the sample of contributions included in our network analysis, while the second level is restricted to quantitative analyses suitable for the MA. To conduct a MA it is crucial to select only comparable papers that provide complete information (mainly on estimated coefficients and standard errors) that can then be used to recover the average effect size Footnote 5 . This implies the exclusion of papers that do not comply with the requirements of a MA. However, those excluded papers can be of interest in building the taxonomy of the whole concerned literature, as they may play a role in building links between different contributions (see Sect. 3 ). Similarly, non-quantitative (policy, qualitative or theoretical) papers may participate as well in the development of research fronts or give a direction to a certain thread of contributions and incidentally affect the detection of clusters. These reasons led us to build our citation-based network and perform the network analysis and the community detection on the whole sample, while only the sample for the MA is restricted only to quantitative contributions that meet the coding requirements. Our final database of point estimates for the MA includes 96 papers released between 2003 and 2020, published in an academic journal, working papers series, or unpublished studies, providing 3,904 point estimates of the effect of slow-onset events (provided by 66 studies) and 2,065 point estimates of the effect of fast-onset events (provided by 60 studies). The list of articles is in the Appendix Table 6 .

2.1 Bibliometric analysis

This section summarizes the most relevant features of the ensemble of economic literature collected in our sample. Footnote 6

The economic literature started to pay attention to the potential relevance of environmental events on migration in the early 2000s, although the topic had already gained some relevance in global debate decades before, and scientific production increased sharply in the last 17 years. Figure 2 shows that the scientific production in the specific field is quite recent, spanning from 2003 to 2020, with a peak of 20 contributions in 2016 and an annual growth rate for the overall period at 18.5 percent. Taking a closer look at the cited references, it is possible to trace back an article published before 2003 (Findley, 1994 ), that provides a qualitative analysis of drought-induced mobility in Mali (finding no evidence of any role of 1983-85 droughts on migration). As our research of documents is based on keywords, naturally the three most repeated are those put in the search key (“migration", “climate change" and “natural disasters"). Footnote 7 Within the topic of migration, there’s a greater emphasis on international mobility compared to internal migration. However, internal migration may include also urbanization or rural-urban migration, and when combined, they are as common as international migration (counting 21 repetitions per group). Environmental migration is also explored as a form of forced migration , originating refugees, or specifically environmental refugees. The keywords related to environmental issues are more focused on slow-onset events like ( rainfall, temperature, global warming and climate variability ) rather than rapid-onset events. Although, some of the latter are more recurrent than others, such as drought, floods and ultimately earthquakes .

Number of documents per year. Note : Sample of academic contributions about migration and environmental factors from Scopus, Web of Science, Google Scholar, IDEAS RePEc, and previous meta-analyses (Hoffmann et al., 2020 ; Beine & Jeusette, 2021 ) collected, merged, screened and included by the authors

Overall 288 authors have contributed to this literature, with 372 appearances, 34 documents are single-authored, the mean number of authors per document is 1.88; when considering exclusively multi-authored documents, the number of co-authors per document rises to 2.16, with a maximum of co-authors of 9. Various disciplines have put attention to the topic. Despite journals specializing in economics and econometrics representing the majority of the sources of publication, the literature includes also other disciplines (Fig. 3 ).

The 20 most relevant publication sources by field.  Note : Sample of academic contributions about migration and environmental factors from Scopus, Web of Science, Google Scholar, IDEAS RePEc, and previous meta-analyses (Hoffmann et al., 2020 ; Beine & Jeusette, 2021 ) collected, merged, screened and included by the authors

Specifically, economic environmental migration is the object of publication in journals specialized in environmental sciences, geography, and social sciences such as urban studies, agriculture, demography, and political studies. A special mention has to be done for development studies: many reviews and journals specialized in development have issued contributions on the topic, highlighting the trend of observing the topic through development lenses. As an example, 14 documents in our sample are published in World Development , a multi-disciplinary journal of development studies.

A picture of the most relevant documents included in the sample is provided by simple measures, such as the number of global citations as reported in Scopus (at the moment of the bulk download of all sources), and the number of local citations, which shows how many times a document has been cited by other papers included in the sample. Measures for the most cited documents (global and local citation scores) in the sample are reported in Appendix Table 7 . The difference between global and local citation scores (almost four times higher) reveals that the documents have been cited by papers not included in our sample. It means that environmental migration has attracted the interest of different disciplines or they became part of the two main strands of literature, climate change, and migration, separately. 58 papers have not been cited in any of our samples, while 52 have zero citations globally. A part of it can be explained by the 18 papers that have been published recently in 2020, which could not have been cited yet because of timing (except for some contributions published in early 2020 such as Mueller et al. ( 2020 ) and Rao et al. ( 2020 ). Footnote 8 Position and the number of citations confirm the central role of papers published by Gray Clark and Valerie Mueller (Gray & Mueller, 2012b , a ; Mueller et al., 2014 ), receiving high citations both globally and internally. Some papers seem to be more relevant locally than globally: Marchiori et al. ( 2012 ) and Beine and Parsons ( 2015 ) had a bigger influence on our sample of economic environmental migration literature rather than globally, scoring the highest number of local citations. Conversely, Hornbeck ( 2012 ) seems to be cited more in literature outside the specific literature of environmental migration.

2.2 Overview of major results

The literature on the effects of climate and natural disasters on migration is characterized by a rich variety of studies both in micro- and macro-economic analyses. Country-level analyses tend to find evidence of a direct or indirect impact of environmental factors on migration patterns, either internally or internationally. Barrios et al. ( 2006 ) and Marchiori et al. ( 2012 ) find evidence of an increase in internal migration, especially towards urban areas in the case of Sub-Saharan Africa, according to many specific historical and developmental factors. Both contributions highlight how worsening climatic conditions correspond to a faster urbanization process. Marchiori et al. ( 2012 ) add also that this climate-driven urbanization process results also in higher international migration rates, acting as a channel of transmission of the effect of climate.

The macro literature, in line with most validated theoretical models of migration, also investigates whether the effect is conditioned to income levels of the country of origin of potential migrants (Marchiori et al., 2012 ; Beine & Parsons, 2015 , 2017 ). The role of income in a specific origin country experiencing the effects of environmental events is found to be crucial to determine the sign and the magnitude of the impact. Cattaneo and Peri ( 2016 ) support from one side the active role of those events in fostering migration, but show how this effect is conditioned to middle-income countries. The effect is the opposite when conditioning the analysis to poor countries, highlighting the existence of certain constraints to mobility. Worsened environmental conditions may exacerbate liquidity constraints or lack of access to credit aimed at financing the migratory project, which lead to what has been called poverty trap . Furthermore, these conditioned results seem to be robust even when another important channel is controlled, agricultural productivity. Climatic conditions and disruptive hazards may constitute major drawbacks for agricultural productivity, leading the agriculture-dependent part of the population to move out from rural areas: Cai et al. ( 2016 ) and Coniglio and Pesce ( 2015 ) provide evidence of an indirect link between worsened temperature and precipitation conditions and migration, mediated by the level of agricultural dependency of the country of origin. Sudden and fast-onset hazards, on the other side, are not found to contribute significantly to human mobility, except in the case of a higher-educated population, more mobile than other groups after the disruption of a natural disaster (Drabo & Mbaye, 2015 ).

Number of case studies covered by the micro-level sub-sample per country. Note : Sub-sample of micro-level studies about migration and environmental factors from Scopus, Web of Science, Google Scholar, IDEAS RePEc, and previous meta-analyses (Hoffmann et al., 2020 ; Beine & Jeusette, 2021 ) collected, merged, screened and included by the authors

Micro-level literature provides a vast variety of case studies on different potential impacts of environmental factors on mobility. In our sample, they almost double macro-level contributions (86 contributions against 47) and provide different scenarios. Firstly, while macro-level studies mostly provide analyses at the global level or for some groups of countries or macro-regions, micro-level analyses tend to observe a specific phenomenon hitting a specific area or to study differences in the impact of a common phenomenon in different areas. The most covered region as a whole is Sub-Saharan Africa, with 65 case studies included in the contributions (Fig. 4 ). Footnote 9 When the level of analysis is less aggregated than the national or sub-national level, and individual or household behavior is observed through the use of surveys, the picture gains complexity and less generalized conclusions. This seems clear in Gray and Wise ( 2016 ) who analyze a series of comparable surveys across five Sub-Saharan countries, which have consistent differences. The heterogeneity of responses to climatic variations across those countries is strictly linked to the characteristics of the area and of the specific households. Poorer countries (such as Burkina Faso) mainly experience internal and temporary migration, often on a rural-rural channel as a way to diversify risk (Henry et al., 2003 , 2004 ). Long-distance migration seems to be constrained by liquidity and access to credit to finance those expensive journeys. Migratory trends of Nigerian households are pushed in times of favorable climatic conditions, while the effect of adverse conditions interacts with a negative effect on income and traps populations at origin (Cattaneo & Massetti, 2019 ). Overall, micro-level studies focused on the African continent highlight the importance of considering the interplay of a variety of factors when it comes to the analysis of the role of environmental factors, defining the new path toward hybrid literature.

The single countries that receive singularly the most attention are Mexico, with 10 case studies, and the U.S., with 9 case studies. This should not be a surprise because of two reasons: firstly, the stock of Mexican emigrates has been constantly the highest in the world (in absolute terms) as well as the migratory flow between Mexico and the U.S. But there might also be a publication-related reason based on the fact that the vast majority of journals in our sample are U.S. based. Major findings support the relevance of environmental drivers (mainly precipitation shortage) on push factors from Mexico ( Feng et al. ( 2010 ) estimates that a 10% reduction of agricultural productivity driven by scarce rainfall corresponds to the rise of 2% of emigrants).

Southern and Eastern Asia, representing by far the most disaster-prone area in the world. Footnote 10 also provide a variety of heterogeneous scenarios. The case of Vietnam (Koubi et al., 2016 ; Berlemann & Tran, 2020 ) shows how the Vietnamese population chooses different coping strategies in response to different kinds of environmental stressors. While gradual climatic variations lead to mechanisms of adaptation in loco to new climatic conditions, sudden shocks drive the decision to migrate elsewhere. However, mobility responses to different types of hazards might be different according to their specific consequences and duration (Berlemann & Tran, 2020 ). On the contrary, the case of Bangladesh supports the hypothesis that the existence of previous barriers to access to migration is worsened by the occurrence of disasters, specifically in the face of recurrent and intense flooding (Gray & Mueller, 2012b ).

The specific case of earthquakes across the world (El Salvador in Halliday ( 2006 ), Japan in Kawawaki ( 2018 ) and Indonesia in Gignoux and Menéndez ( 2016 ) for instance) shows a common trend of outcomes: highly disruptive disasters such as earthquakes tend to decrease mobility from the hit area. An interesting mechanism to explain this common trend found in three very different contexts is given by, not only the already mentioned financial constraints but also the possibility of higher local employment opportunities due to post-disaster reconstruction (Gignoux & Menéndez, 2016 ; Halliday, 2006 ). Moreover, households are found to respond to hazard by using the labor force as a buffer to the damages and redistributing labor within the household, with female mobility drastically dropping more than males and being substituted with increased hours of domestic labor (Halliday, 2012 ).

Analyses on South American countries also contribute to giving a hint of the complexity of the phenomenon. Thiede et al. ( 2016 ) show how internal migration is indeed impacted by rising temperature when considering the general effect; however, it hides an extreme heterogeneity of outcomes when specific characteristics of the areas and individuals are taken into account, resulting in a non-uniform effect.

An evident gap in the literature emerges in Fig. 4 : European countries have rarely been the object of study of the impact of environmental factors on mobility. This might be motivated by the fact that the European continent is mostly seen as a destination for migrants than an origin. It should not surprise that the two articles covering European countries, namely Italy (Spitzer et al., 2020 ) and the Netherlands (Jennings & Gray, 2015 ) analyze historical data of mobility at the beginning of the XX century (respectively earthquake in Sicily and Calabria and climate variability associated with riverine flooding in the Netherlands). Nevertheless, figures show that Europe is not unrelated to the occurrence and frequency of hazards as well as to sizable internal mobility that should receive some attention.

3 The inter-connectivity of papers

Our quantitative approach aims at analyzing the connectivity that exists among papers according to a citation-based approach and detecting the existence of communities or clusters. Since our target literature is characterized by a high heterogeneity of results, both in the direction and magnitude of the impact, we try to investigate the existence of potential specific patterns that lead to a certain type of analysis, methodology, or result under network-analysis lenses. We then use all information from this section to implement the meta-analysis.

3.1 Bibliographic coupling and citation-based approaches

The citation-based approach we choose is called bibliographic coupling. Footnote 11 Two scientific papers “bear a meaningful relation to each other when they have one or more references in common". Thus, the fundamentals of the link between two papers are depicted by the number of shared papers they both include in their references, which constitute the strength of the connectivity they have. In other words, a reference that is cited by two papers constitutes a “unit of coupling between them" (Kessler, 1963a ). Two articles are then said bibliographically coupled if at least one cited source appears in both articles (Aria & Cuccurullo, 2017 ). Bibliographic coupling is increasingly becoming widely used in citation analysis, thanks to some specific advantages (and despite some disadvantages). Conceptually, through the linkages established, it gives a representation of the basic literature of reference and, incidentally, implies a relation between two papers that reveals a potential common intellectual or methodological approach (Weinberg, 1974 ). The constancy of the links between the papers over time, being based on cited references which, once published and indexed, is also an asset (Thijs et al., 2015 ). Most importantly, the bibliographic coupling is more suitable for recent literature than other citation-based approaches. For reasons of timing and extension of the time window, Footnote 12 using any other citation-based approach would have resulted in a very sparse matrix and created many isolated observations which would not be inter-connected for reasons other than conceptual, but just for the fact that they could not have been cited yet. Not only do the characteristics of our sample motivate the choice of the approach: keeping in mind that this stage of the analysis aims to investigate and map current research fronts in the target literature rather than to look at historical links or the evolution of school of thoughts, bibliographic coupling seems to be the best tool to capture them (Klavans & Boyack, 2017 ).

To obtain the network of bibliographically coupled papers, we initially extract the list of cited references from each article and build a bipartite network, a rectangular binary matrix \(\textbf{A}\) linking each paper in the sample to their reference (Aria & Cuccurullo, 2017 ):

The matrix \(\textbf{A}\) is composed of 151 rows i representing the papers belonging to the sample and 5.433 columns j representing the ensemble of references cited in each paper in the sample. Each element \(a_{ij}\) of the matrix equals 1 when paper i cites paper j in its bibliography; \(a_{ij}\) is equal to 0 otherwise. Starting from matrix \(\textbf{A}\) , we can derive the bibliographic coupling network \(\textbf{B}\) as follows:

where \(\textbf{A}\) is the cited reference bipartite network and \(\mathbf {A^T}\) is its transpose. \(\textbf{B}\) is a symmetrical square matrix 151 \(\times\) 151, where rows and columns are papers included in the sample. Element \(b_{ij}\) of the matrix \(\textbf{B}\) contains the number of cited articles that paper i and paper j have in common. By construction, the main diagonal will contain the number of references included in each paper (as element \(a_{ii}\) defines the number of references that a paper has in common with itself).

The resulting matrix displays an undirected weighted network in which the 151 vertices are the set of papers included in our sample and the edges represent the citation ties between them. An existing tie implies that common reference literature exists between vertex i and j . When two nodes are not linked, the corresponding value of their tie is zero, as they do not share any common reference. Therefore, the network is weighted with the strength of the connections between papers i and j being measured by the weights associated with each tie. To avoid loops, which would be meaningless for our investigation, Footnote 13 we set the main diagonal to zero. Few ties exceed 20 shared cited references, with a maximum value of 48. Footnote 14 It can be argued that the number of references included in an article is not neutral to the resulting tie with any other article. Measuring the correct relatedness of nodes is of primary importance to produce an accurate mapping of literature (Klavans & Boyack, 2006 ). Citation behaviors of authors may interfere with the observation of core reference literature at the basis of coupled nodes. An author may opt for an extensive approach of citations and include a consistent number of references to display some particular links or details of a paper; authors may also decide for a less inclusive approach and include just essential cited references in the list. In other words, the number of included references in one article may dissolve meaningful information about the ties. Furthermore, specific formats or types of articles lead to broader or narrower bibliographies. To address these concerns, a process of normalization is needed so that data can be corrected for differences in the total number of references. Bibliometric literature has dealt with this issue through the calculation of different similarity measures . An accurate overview of the possible measures of similarity is provided in van Eck and Waltman ( 2009 ). Overall, such indices aim to determine the similarity between two units according to their co-occurrence (value of association between them, which in our case, is the number of common references in the bibliography) adjusted in different ways for the number of total occurrences of the single units. However, despite the need to correct data for many purposes in citation-based networks and obtain a size-independent measure of association, there is no consensus on which measure is the most appropriate (van Eck & Waltman, 2009 ): tests of accuracy and coverage proposed by different authors have reached different conclusions (Klavans & Boyack, 2006 ; van Eck & Waltman, 2009 ; Sternitzke & Bergmann, 2009 ). We apply a simple ratio between the observed number of commonly shared references and the product of the number of cited references in each of the two coupled papers. It has been defined as a measure of association strength (van Eck & Waltman, 2009 ) and it can be expressed as:

where \(b_{ij}\) corresponds to the weights of the tie between i and j in the original bibliographic coupling network; \(b_{ii}\) and \(b_{jj}\) are respectively the number of cited references included in paper i ’s bibliography and in paper j ’s bibliography, which corresponds to the original value on the diagonal. The obtained weighted network will serve to detect communities of papers through their common references and investigate if referring to a certain (group of) paper(s) creates meaningful clusters of items aggregating around certain common characteristics.

3.2 Community detection

We intend to identify the existence of communities in our network. The assumption is that papers citing the same references aggregate into a group that shares certain features, which could be methodological approach, level of analysis, specific sub-topics of the literature, and outcomes. The extreme heterogeneity of outcomes in this specific literature may be motivated partially by the heterogeneity of the events themselves (type of environmental factor, type of mobility, preexisting conditions in the specific area) or the theoretical and empirical modeling; it may also be motivated by other factors, that can be traced in some patterns linked to the characteristics of single publications. The procedure of community detection is aimed at investigating which are the “forces" that aggregate or disperse papers with each other, primarily through the direct observation of main characteristics, and then running separate MAs on each cluster. Community detection in the bibliographic network is often made through Louvain community detection algorithm (Blondel et al., 2008 ). In this analysis, a community is thought of as a group of contributions that share common references and form strong common ties with each other, while others have less shared characteristics and structure. The algorithm can detect clusters of contributions with dense interaction with each other and sparse connections with the rest of the network (Fig. 5 ).

Bibliographic coupling network and detected communities Note : Bibliographic coupling network of 151 documents included in the sample obtained from Scopus, Web of Science, Google Scholar, IDEAS RePEc and previous meta-analyses (Hoffmann et al., 2020 ; Beine & Jeusette, 2021 ). Each node represents a paper included in our sample and its size corresponds to its weighted degree. Nodes are tied by links whenever two nodes share at least one common reference. The thickness of links is given by the association strength of the tie between two nodes (to provide a clear visualization, only nodes with weights higher than the mean are displayed). Colors correspond to communities of belonging of each paper: Cluster 1 is represented in violet, Cluster 2 in green, Cluster 3 in blue, and Cluster 4 in yellow. The description of each Cluster is presented in the text

The procedure identifies four main clusters. Our network being relatively small allows analyzing the main characteristics of each cluster. Following the full-text screening made in the first step of our threefold approach, we summarized some meaningful indicators about the analysis (such as type - quantitative, qualitative, theoretical, policy, literature review -, level - macro or micro for quantitative and qualitative studies -, unit - country, household, individual, territorial units), the object of the analysis (concerning the type of migration and environmental factors studied and the area) and theoretical and empirical approach (empirical approach and whether it is theory-based, estimation strategy and potential channel investigated). Finally, we recorded a synthetic indicator of the concluding effect of environmental factors on migration patterns: for each paper, we assigned the value “positive", “negative", “not significant" or a combination of the three (in case a paper contains multiple analysis of different migration or environmental factors that lead to different outcomes). Thanks to these indicators we were able to have a picture of the main common characteristics of the papers included in a cluster (Appendix Table 8 ), which will be tested and eventually confirmed in the MA.

The first cluster (Cluster 1) is the most populated, counting 51 papers spanning the entire period considered (from 2003 to 2020). In terms of the type of analysis, it contains the largest variety: as in all clusters, quantitative studies represent the majority (as they are the 76% of the full sample), but this cluster contains also most of the qualitative analyses (10 out of 13) and policy papers (5 out of 7) of the full sample. Published papers are predominant (47 out of 50). Except for a few papers, the analysis is mainly carried from a micro perspective, with individuals as units of analysis, based on surveys. Interestingly, most of the micro-level studies included in Beine and Jeusette ( 2021 ) can be found in this cluster. Authorship is very concentrated around two main authors, Clark Gray, (co-)authoring 9 papers, and Valerie Mueller, (co-)authoring 4 papers. Many of their co-authors appear in this community, which indeed scores the highest collaboration index of all communities (2.86), much higher than the full sample (2.16). Another important feature is that Cluster 1 includes the micro-level papers with the highest global citations: Gray and Mueller ( 2012b ), Feng et al. ( 2010 ), Gray and Mueller ( 2012a ), Mueller et al. ( 2014 ), Henry et al. ( 2004 ), Henry et al. ( 2003 ) and Gray ( 2009 ). This is also shown by the fact that the number of average citations per document is the highest among all clusters (34.84). Journals are also quite concentrated around a few of them, World Development and Population and Environment mainly. The content of the analyses is mainly focused on climatic change exclusively (precipitation and temperature), while few studies include also natural disasters. All corridors of migration are investigated, with no specific predominance of internal or international migration (which is a characteristic of individual-level studies, mainly based on surveys). Even though the majority of outcomes show a positive coefficient, that can be translated into finding an active role of environmental factors in pushing migrants out of their origin areas, it is not consensual to every paper: variation among results is high compared to other clusters, most paper finding complex relations between the two phenomena and different directions according to different dimensions. Empirical strategies are often based on discrete-time event history models estimated through multinomial logit. This reflects the approach of the main authors included in this community. A strong accent is put on the importance of the agricultural channel and the theme of adaptation to the change in environmental conditions.

The second community (Cluster 2) counts 28 papers, mostly published, except for 4 of them. It is composed of mostly quantitative papers, accompanied by 5 literature reviews. As in the previous cluster, most studies are at a micro level, with all kinds of units of analysis and aggregations. Both patterns of migration are explored, but with special attention to urbanization and internal mobility. Contrarily, it seems to put a stronger accent on natural disasters rather than on slow-onset events. The majority of papers in Cluster 2 have been excluded from Beine and Jeusette ( 2021 ) (only 5 included, compared to the 21 in Cluster 1) and Hoffmann et al. ( 2020 ) (only 1, all others being in Cluster 4). All papers analyzing the impact of different kinds of natural disasters in the U.S. are included in this cluster. Empirical approaches such as the differences-in-difference model and instrumental variable are often used. The papers explore a large variety of potential channels and mechanisms of transmission of the impact of environmental factors on migration (income, agriculture, employment, liquidity constraints), and only in a few cases, a negative direction is found.

The third cluster (Cluster 3) includes the most recent papers: only one paper dates 2011, all other ones are published or issued after 2015. This is part of the reasons why the average citations per document in this cluster is the lowest (10.89) compared to any other cluster. Half of the overall unpublished papers are included in this cluster. In terms of kind of analysis, this cluster appears to be very heterogeneous: even if the micro-level analysis is the majority, 12 papers apply a macro-level analysis on countries. Both cross-country and internal migration are considered, but the majority of them investigate the impact of slow-onset events rather than fast-onset. Many of the analyses are theory-based, especially on classic economic migration theories (Roy-Borjas model, New Economics of Labor Migration), or general or partial equilibrium models. This cluster is also peculiar for the heterogeneity of empirical outcomes, which are often multiple for a single paper: outcomes vary according to the different channels explored, i.e. different levels of agricultural dependency, presence of international aid, and level of income. In many cases, environmental factors are an obstacle to the decision to migrate from an area, or completely neutral. Comparatively, outcomes from this cluster tend to show a complex picture and highlight the many dimensions that may intervene in determining the direction of the impact.

Contrary to the previous one, Cluster 4 is extremely homogeneous. It contains almost exclusively quantitative (32 out of 35) and macro-level studies (30 out of 35). It covers equally slow- and fast-onset events and their impact on mobility. Most importantly, it aggregates 23 of the 30 papers reviewed in Hoffmann et al. ( 2020 ), making this cluster very representative and comparable to Hoffmann et al. ( 2020 )’s MA. Additionally, this community appears to be solid also in terms of theoretical and empirical approaches, as micro-founded gravity or pseudo-gravity models are widely used in it (more than half of them use such models). None of the studies find a negative impact of environmental factors on migration, they mainly estimate positive and significant outcomes, with few not-significant results for specific cases. The most locally cited macro papers are included in this cluster, which also receive high global citations with an average of citations per document 24.91 (even though lower than Cluster 1).

This description of cluster composition serves as a preliminary investigation of which are the main characteristics linking papers together through their citation behavior. It emerges that stronger links are given by diverse indicators varying across clusters. To test which are the sources of heterogeneity between clusters that aggregate papers within a cluster and their impact on the estimated effect size, in the next section, we will use this partitioning to run four separate MAs and compare the conclusions.

4 Meta-analysis

The purpose of our MA is to summarize the results of collected studies and, at the same time, highlight any possible sources of heterogeneity. The analysis is based on four assumptions: (i) our parameter of interest, which we call \(\beta\) , is the effect of climate change on migration; (ii) most researchers believe that \(\beta\) is greater than zero, and this is indeed true; (iii) the sign is not enough for decision-makers; (iv) this has attracted a large literature that has obtained a large number of estimates \(\hat{b}\) of \(\beta\) . Each of the 96 selected papers contains one or more equations that estimate the migration effect due to environmental factors. Footnote 15 In addition to the characteristics specific to migration itself, the estimated impact on migration can also be distinguished according to different features of environmental factors. Since comparability among studies, and more specifically among estimated \(\beta\) s, is a crucial issue for the MA, we group all collected estimates and conduct two separate analyses according to the type of environmental phenomenon: gradual or slow-onset events and sudden or fast-onset events. To compare the estimates and correctly interpret the synthetic results we need to standardize all collected effect sizes \(\beta\) in a common metric. In this MA the estimates from separate, but similar studies, are converted into partial correlation coefficients ( pcc ):

and its standard error, \(se_i\) :

where \(t_i\) and \(df_i\) are the t-value and the degrees of freedom of the i-th estimate \(\beta _i\) . The pcc is commonly used in MA literature (Doucouliagos, 2005 ; Stanley & Doucouliagos, 2012 ; Doucouliagos & Ulubasoglu, 2006 ; Brada et al., 2021 ) and allows to analyze within a single framework of all available studies on the effects of environmental stressors on migration regardless of the specification or measure of migration used. Footnote 16 Summarizing all the different estimates together in a single coefficient raises the question of heterogeneity within the same study and between studies. The summary effect is calculated as follows:

where \(\hat{b}_i\) is the individual estimate of the effect and weight, \(w_i\) , in a fixed effects model (FEM) is inversely proportional to the square of the standard error, so that studies with smaller standard errors have greater weight than studies with larger standard errors. The FEM is based on the assumption that the collected effect sizes are homogeneous (the differences observed among the studies are likely due to chance). Unlike in the FEM, random-effects model (REM) takes into account the heterogeneity among studies and weights incorporate a “between-study heterogeneity", \(\hat{\tau }^2\) . In the presence of heterogeneity, the two models likely find very different results, and it may not be appropriate to combine results. A test of homogeneity of the \(\beta _i\) is provided by referring to the statistic Q to a \(\chi ^2\) -distribution with n - 1 degrees of freedom (Higgins & Thompson, 2002 ): if the test is higher than the degrees of freedom, the null hypothesis is rejected (and thus there is heterogeneity). Another test commonly used is the \(I^2\) inconsistency index by Higgins and Thompson ( 2002 ) describing the percentage of the variability of the estimated effect that is referable to heterogeneity rather than to chance (sample variability). Values of the \(I^2\) range from 0 percent to 100 percent where zero indicates no observed heterogeneity. Since most computer programs report \(I^2\) , and so it is readily available, it is largely used to quantify the amount of dispersion. However, it is a proportion and not an absolute measure of heterogeneity in a meta-analysis (Borenstein et al., 2017 ). To understand how much the effects vary and report the absolute values, we compute the prediction interval as suggested by Borenstein et al. ( 2017 ). The results of the meta-synthesis of the collected estimates (Table 1 ) are statistically significant, except for findings of the slow onset effect of paper included in Cluster 2 (where the most of studies focus on the fast onset effect), in which both FEM and REM give statistically insignificant averages.

The preliminary result of the basic MA is that environmental factors seem to influence migration positively, even if the magnitude is very small and the REM mean is statistically significant only in the case of fast-onset events. The mean effect by cluster becomes negative in the case of estimates of slow-onset events in Clusters 1 and 3 and for the estimates of fast-onset events in Cluster 2.

4.1 Meta-regression tests of publication selection bias

Different findings of the same phenomenon can be explained in terms of heterogeneity of studies’ features, however, the literature also tends to follow the direction consistent with the theoretical predictions causing the so-called publication bias. Footnote 17 Meta-regression tests, such as the funnel asymmetry test (FAT), allow for an objective assessment of publication bias:

Weighted least squares (WLS) corrects the previous equation for heteroskedasticity (Stanley & Doucouliagos, 2017 ) and it can be obtained by dividing \(pcc_i\) by the standard errors:

Results are used to test for the presence of publication selection ( \(H_0:\beta _1 = 0\) ) or a genuine effect beyond publication selection bias ( \(H_0:\beta _0 = 0\) ). According to the Funnel Asymmetry and Precision-Effect Tests (FAT-PET), in the absence of publication selection the magnitude of the reported effect will vary randomly around the “true” value, \(\beta _1\) , independently of its standard error (Stanley & Doucouliagos, 2012 ). Replacing in eq. ( 7 ) the standard error \(se_i\) with the variance \(se_i^2\) , as the precision of the estimate, gives a better estimate of the size of the genuine effect corrected for publication bias (Stanley & Doucouliagos, 2014 ). This model is called “precision-effect estimate with standard error” (PEESE) and the WLS version is:

Table 2 shows results of the FAT-PET using multiple methods for sensitivity analysis and to ensure the robustness of findings. To take into account the issue of the dependence of study results, when multiple estimates are collected in the same study, the errors of meta-regressions are corrected with the “robust with cluster" option, which adjusts the standard errors for intra-study correlation.

Column (1) of Table 2 presents the FAT-PET coefficients, column (2) shows the results of the WLS model to deal with heteroskedasticity, columns (3) and (4) present the results of the panel-random effect model (REM) and multilevel mixed-effect model that treats the dataset as a panel or a multilevel structure.

Looking at the estimates of the effect of climate change on migration, the FAT coefficients ( \(\hat{\beta }_1\) ) are not statistically significant, implying that there is no evidence of publication bias, while the positive and statistically significant PET coefficient ( \(\hat{\beta }_0\) ) indicates a genuinely positive slow-onset effect exists, in particular in the case of Cluster 4. Conversely, in the case of Cluster 3 the REM and multilevel mixed-effect model find that, even if in presence of publication bias, the impact on migration is negative. Table 2 provides evidence of publication bias in the literature focusing on the effect of natural disasters on migration. The estimated FAT coefficient is statistically significant in the overall sample, especially due to papers in clusters 1 and 3, and there is insufficient evidence of a genuinely positive effect (accept \(H_0: \hat{\beta }_0\) ).

4.2 Multiple meta-regression analysis: econometric results and discussion

The multiple meta-regression analysis (MRA) includes an encompassing set of controls for factors that can integrate and explain the diverse findings in the literature. To capture possible sources of bias among all analyses, we code all differences in the features of the various studies and regressions and include a set of dummies to control for them. Specifically, we code left- and right-hand side characteristics of regressions estimated in the collected papers and generate a set of dummies for paper features, dependent variables, independent variables, sample characteristics, and regression characteristics. Footnote 18

The overall sample includes both unpublished and published papers, so we add some moderators variables describing different features of the studies that are published. In particular, we introduce a dummy for Published articles and a control for the quality of the journal in which the study is published by adding the variable Publication Impact-factor . In reporting the main results, some authors emphasize a benchmark regression that produces a preferred estimate, thus we add the dummy preferred specification equal to 1 when the reported effect size is obtained from the main specification. Concerning the measure of migration, the dependent variable in the left-hand side of the regression, original studies mainly distinguish migration by corridor , which are mainly two, internal and international migration. In this context, we distinguish also a special internal corridor, the one characterized by rural-urban mobility, to investigate the potential impact of an environmental variable on the urbanization process. Whenever the corridor is not specified, the variable is categorized as undefined (which will be the reference category in the estimation). Dependent variables differ also in terms of measurement of the phenomenon: specifically, we separate measures that express flows from those expressing stocks. The first category includes both studies that use flows (or an estimation of flows) and rates of migration. The second category captures those cases in which migration is measured as a stock of migrants at the destination. The reference category is direct measures, which mainly capture whether migration has occurred or not (typically dummy variables used on survey-based samples equal to 1 when the individual migrates and 0 otherwise). We also include information about the countries of origin and the destination of migrants. Origins are categorized by macro-regions: Africa, Asia, Europe, Latin America and Caribbean, Middle East and North Africa, and North America. The reference category is “world", identified whenever origin countries are not specified (typically in multi-country settings). Destinations are categorized by level of income. The choice of this categorization is led by the aim to identify differences in the possibility to choose a destination. Categories are divided into high, higher-middle, lower-middle, and low-income.

The specific objective of the study is the impact of environmental variables on migration, thus on the right-hand side of the regression a proxy of the environmental change is included. Slow-onset events are typically defined as gradual modifications of temperature, precipitation, and soil quality. Respectively, three dummies temperature, precipitation and soil degradation are created. Each of these phenomena is measured in different ways, and the use of a specific kind of measurement is relevant to the outcome. Both temperature and precipitation have been measured in levels (simple level or trend of temperature/precipitation); deviation, as the difference between levels and long-run averages; and anomalies, mostly calculated as the ratio of the difference between the level and the long-run mean and its standard deviation. Soil degradation includes events such as desertification, soil salinity, or erosion. Additionally, we also code the time lag considered concerning the time units of the dependent variable: whenever the period considered corresponds to the same period of the dependent variable the lag is zero, while it takes values more than zero for any additional period before the dependent variable time-span. This control also allows us to account for varying time-frames in different studies, including situations where migration spans several years or occurs suddenly in the aftermath of a natural disaster. The second battery of coded variables refers to fast-onset events, which can be also defined as natural hazards or extreme events. The main classification of fast-onset events reflects the one reported in Sect. 2 : geophysical (earthquakes, mass movements, volcanic eruptions), meteorological (extreme temperature, storms - cyclones, typhoons, hurricanes, tropical storms, tornadoes), hydrological (floods and landslides) and climatological (droughts or wildfires). Fast-onset events also differ in the way they are measured. Possible measures are occurrence (when the measure is a dummy capturing if the disaster happened or not), frequency (the count of events that occurred in the area), intensity (i.e. Richter scale for earthquakes, wind speed for tornadoes, etc.), duration (length of the occurrence of the event) and losses (when the disaster is measured in terms of the affected population, number of deaths or injured people, number of destroyed houses or financial value of the damaged goods). As for slow-onset events, we code a continuous variable capturing the time lag of the event concerning the dependent variable. A dummy capturing whether the coefficient refers to multiple disasters is also included.

Characteristics of the sample are one of the main sources of heterogeneity. The level of the analysis varies considerably from paper to paper, as we include both micro-and macro-level studies. we code variables capturing both the specific unit of analysis and the source of the data. Typically micro-level studies use data coming from censuses or surveys where households or individuals are the units of analysis. Country-level studies usually take the source of their data from official statistics . Other kinds of sampling are included in the reference group (for example small territorial aggregates such as districts, provinces, or grid cells). We also code a variable capturing the time span of the analysis, subtracting the last year of observation from the first one. The role of econometric approaches may have an impact on resulting outcomes. Beine and Jeusette ( 2021 ) emphasized in their work the importance of methodological choices, with differentiated results depending on estimation techniques. First of all, we code a panel dummy to capture whether the structure of data and related estimation techniques has an impact. Furthermore, we distinguish Poisson estimations that include the Pseudo Poisson Maximum Likelihood (PPML) estimator and Negative Binomial Models; linear estimators, both Ordinary Least Squares (OLS), linear probability models and maximum likelihood models; conventional Instrumental Variables (IV) estimators, two-stage least squares (2SLS), and other cases of estimators as Generalized Method of Moments (GMM) used to control for endogeneity; and finally, logit which comprises multinomial logit models. Any other estimator (i.e. Tobit, panel VAR) is less frequent and grouped in a category other estimators used as the reference group.

Theoretically, the impact of environmental variables on migration may be mediated, channe-led, or transmitted through other phenomena that can be controlled for or interacted with. Most of models investigating general migration determinants usually control for several possible determinants to recover the effect of the specific objective variable, with all potential other factors being controlled for. The majority of these additional controls are suggested by theoretical models and then introduced in the empirical model. Furthermore, methodological approaches in our sample are found to often include interaction terms to specifically address the combined effect of an environmental variable with other potential factors. Thus, we introduce two groups of variables, controls and interacted terms , categorized both to capture factors or channels such as income, agriculture, conflicts, political stability, cultural or geographical factors. Among the list of controls, we also include a dummy that captures whether both slow- and fast-onset events are included in the regression.

Table 3 shows the results of the multiple MRA on the literature in slow-onset events (precipitation, temperature, and soil quality) in which potential biases are filtered out sequentially by the addition, in a stepwise manner, of statistically significant controls. Column (1) presents results for the whole sample of studies estimating the impact of climatic variations on migration, and columns (2) to (5) show the results of papers grouped by clusters to highlight how specific features characterizing the cluster influence the magnitude of the estimated effect. The results are unfolded below.

Column (1) refers to the overall sample and shows a coefficient of the main variable of interest ( \(\hat{\beta }_0\) ) negative and statistically significant, implying that climatic variations may decrease incentives for migration by exacerbating credit constraints of potential migrants. Looking at results for different clusters (columns 2-5) such a negative effect is generated by studies that are included in clusters 1 and 3. The MRA of papers in clusters 2 and 4, instead, gives positive and statistically significant PET coefficients ( \(\hat{\beta }_0\) ) implying that climate changes induce people to migrate. Concerning the FAT-test, the intercept ( \(\hat{\beta }_1\) ) might deviate from zero confirming the presence of publication bias: the peer-review process seems to particularly affect the magnitude of the estimated effect of studies in all clusters except for Cluster 3.

Most of the papers included in the MRA for slow-onset events are published (52 articles out of 66), indeed the estimated coefficients of controls for published articles are useful to evaluate if the peer-review process exerts some influence on reported results in the collected studies. In Cluster 3 estimates obtained by the Preferred specification tend to be slightly lower while articles published in journals with higher impact factors report lower estimates of the impact of slow-onset events on migration. In Cluster 4, instead, results of Published articles are lower, even if the mean effect of this group of studies remains positive.

From the other sets of controls emerges that specific features of studies included in the MRA differently explain the diversity in the results within clusters. The positive coefficients of controls for corridors such as Internal and Urbanization state that people respond to adverse climatic change with increased internal migration. The only exception is for studies included in Cluster 3, this is the most heterogeneous cluster of most recent papers, where heterogeneous approaches (micro-and macro-level and type of migration) lead to a large heterogeneity in outcomes, varying according to different channels explored. Findings obtained when mobility is measured by Flows seem to be lower in the overall sample. In macroeconomic literature, usually, the measurement of migration is a stock variable, since it is generally easier to find and measure the number of foreign citizens born or resident in a country at any given time. Data on flow variables and migration rates, or the number of people who have moved from an origin to a destination in a specific period, are less available, and analyses often rely on estimates and computations of this data. Therefore, the opposite sign of the coefficient of the variable Flows in Cluster 1 is not surprising since this cluster collects all micro-level studies (where the migration variable refers to the movements of individuals as a unit, based on surveys).

Controls for how the climatic phenomenon is measured, Precipitation measures and Temperature measures , seem to differently affect the heterogeneity of results and, in many cases, the estimated coefficients are statistically significant but very close to zero.

The estimated coefficients of dummies for country groups included in our multiple MRA indicate how results from analyses focusing on specific regions of origin differ. In particular, positive coefficients of controls Asia and Europe support the idea that the results of analyses that focus on the migration from these regions are likely to be positive (with exception of Cluster 1), while if the people move from a country in the region of North America the impact of climate changes on migration is lower and can be negative. The climate impact on migration from LAC (Latin America and the Caribbean) countries are higher in Cluster 3 (where the PET coefficient is negative) and lower in Cluster 4 (where the PET coefficient is positive).

Regarding the heterogeneity produced by the fact that studies use different sources of data for migration, we add dummies for sources used. All estimated coefficients of this set of controls are statistically significant in Cluster 1: the use of different databases might influence the wide variety of findings. Effect sizes in Cluster 2, instead, are not affected by the source of data used.

Since it is natural to expect the adjustment of migratory flows in response to climate change is not instantaneous, especially in the case of gradual phenomena, most of the studies use a panel structure with a macroeconomic focus and attempt to assess the impact of changes in climatic conditions on human migratory flows in the medium-long term. Microeconomic analyses mostly use cross-section data to explain causal relationships between specific features of individuals, collected through surveys and censuses, and various factors determining migration by isolating the net effect of the environment. Analyses at Individual level tend to capture a more negative impact of climate changes on migration, whereas analyses at Country level tend to find a more positive effect. As already said, for micro-level analyses in Cluster 1 controls related to sample characteristics have opposite signs. Looking at dummies for the estimation techniques, our evidence suggests that the diversity in the effect sizes is in part explained by differences in techniques. In particular, positive and significant coefficients are found for controls as OLS and ML estimators for cross-section analyses, same for panel studies that use Panel estimation techniques, and Instrumental Variables ( IV ) or GMM estimators to correct for endogeneity. Micro-economic analyses (Cluster 1) use more disaggregated data, while the high presence of zeros in the dependent variable is treated with a Poisson estimator, which tends to produce lower estimates.

Many authors highlight the importance of variables of political, economic, social, and historical nature, in influencing the impact of climatic anomalies on migration processes, emphasi-zing the role of important channels of transmission of the environmental effect to migrations. We include in the multiple MRA a set of dummies for Controls included in the estimation of the model of migration and dummies for Channels through which the climatic event determines migration phenomena. The idea is that studies based on the same theoretical framework tend to include the same set of control variables or interacted terms and we find that many of these controls may positively and negatively affect the effect size of climate changes on migration.

Table 4 shows the results of the MRA for fast-onset events, or rather natural disasters, more or less related to climate change, which appear as destructive shocks of limited duration and for which the ability to predict is reduced. Footnote 19

The coefficient of \(\hat{\beta }_0\) , is positive and statistically significant in the overall sample and clusters 2 and 4, providing evidence of an increase in migration due to sudden natural hazards. It is worth noting that papers in Cluster 2 (column 3) mainly focus on fast-onset events and the summarized effect size is positive and very high. On the other side, the summarized effect of papers in clusters 1 and 3 is negative and statistically significant.

Results show evidence of publication bias for the overall sample and in Cluster 3, with \(\hat{\beta }_1\) statistically significant signaling that the reported effect is not independent of its standard error. The significant and positive coefficient found for the published dummy confirms that there is a general Publication Impact , so the peer-review process seems to affect the magnitude of the estimated effect, especially in clusters 1 and 2. Articles published in journals with higher Impact-factor get higher estimates of the effects of natural disasters on migration, with exception of published articles in Cluster 2, suggesting that editors prefer to publish results that have a positive but more limited effect. Natural disasters affect domestic and international migration flows. The positive coefficients of the group of controls related to the type of migration, in clusters 2 and 3 confirm that people respond to natural disasters with any kind of mobility. Specifically in Cluster 2 natural disasters increase both Internal and Urbanization migration, while studies in Cluster 3 find a greater effect on Internal and International movements of people. In Cluster 4, instead, estimates of the impact of natural disasters are lower in the case of Internal migration. Hydrological events have a greater impact on migration, the estimated coefficient is statistically significant in all clusters; if the fast-onset event refers to Geophysical , Meteorological and Climatological disasters the effect on migration is lower.

The severity of natural disasters, such as hurricanes, landslides, or floods, affects regional agricultural production and it also has direct effects on employment and income in the agricultural sectors of the affected regions pushing people to migrate. However, on the one hand, natural disasters, such as droughts, floods, and storms, push individuals to move to find new sources of income or livelihood, on the other hand, natural disasters such as earthquakes, tsunamis, or hurricanes cause losses to populations that might lead people into a poverty trap, with potential migrants not having the resources to finance the trip. These effects, already highlighted by the literature, seem to be confirmed. Also in this literature, indeed, various controls and transmission channels analyzed in the original empirical models have a role in determining heterogeneity in results.

5 Conclusions

The present meta-analysis, aimed to systematically review and synthesize the empirical evidence on the relationship between environmental change and human migration, suggests that while there is a small, positive, and significant effect of slow- and rapid-onset environmental variables on migration, the heterogeneity of results in the existing economic literature highlights the need for a nuanced understanding of the causes and effects of environmental migration, as well as the specific characteristics of the places and populations involved.

If a key function of meta-analysis is to challenge and test the results of empirical studies, our study provides important insights that can inform both researchers and policymakers on the relationship between human migration and environmental changes or shocks. Specifically, our findings suggest that a more nuanced and context-specific understanding of environmental migration is needed. Future research could profit from our work by exploring the average effect of specific environmental shocks, such as droughts or floods, and the important role of mediating factors that influence the decision to migrate, such as specific economic and social conditions.

The paper also offers an encompassing methodology for the empirical analysis of very heterogeneous outcomes of a research field. The sample collected through a systematic review of the literature, the bibliometric analysis, the construction of a co-citation network and the community detection on the structure of the network of essays, allow the inspection of a scientific area also in absence of a uniform and cohesive literature. In the case of environmental migration, the too many different characteristics in terms of object of analysis, empirical strategy, and mediating covariates render the meta-analytic average effect estimates just a first approximation of the quantitative evidence of the literature.

As shown in the present meta-analysis, when the level of heterogeneity in the outcome of a literature is relevant, as for the four clusters of papers that compose the economic literature on environmental migration, a group-by-group analysis has to be preferred and compared with the results of an overall meta-analysis.

Moreover, our analysis highlights the need for greater collaboration and standardization of methods in the study of environmental migration. We report a lack of uniform and cohesive literature, with different studies using different methods, covariates, and definitions of key variables. This limits the external validity of existing results and calls for greater efforts by scholars and institutions to validate existing studies and improve the quality of data and methods used in future research.

Overall, our meta-analysis contributes to a better understanding of the complex relationship between slow or rapid environmental change and human migration. The implications of this work extend beyond the academic community to inform public policy and action. As environmental change and human migration continue to characterize the global system, it is crucial for decision-makers to consider the insights provided by scientific research and for the scientific community to continue to produce results that improve the external validity of existing studies and help delineate evidence-based policies.

A detailed table highlighting specific studies featured in other meta-analyses, along with their citations, that have been reviewed in our study is provided in the Supplementary material, Section A.

The extraction is made through bibliometrix , an R tool for science mapping analysis that reads and elaborates the information exported by Scopus and Web of Science (Aria & Cuccurullo, 2017 ).

Scopus: key (“migration" and (“natural disasters" or “climate change")) and ( limit-to ( subjarea ,“econ")) and ( limit-to ( language ,“English")), Date: 24/11/2020. Web of Science: (( AK =(migration and (“natural disasters" or “climate change"))) or ( KP = (migration and (“natural disasters" or “climate change")))) and language : (English) Refined by: web of science categories : (“economics"), Date: 24/11/2020.

We use the Advanced Search tool, searching by Keywords and Title: migration and (“natural disasters" or “climate change").

Our inclusion criteria prioritize studies reporting outcomes in an appropriate and consistent manner. In particular, we have excluded studies that do not rely on a complete set of objective measures. For instance, studies that only present estimated coefficients, solely indicating the significant level, without reporting standard errors or t -ratios have been excluded because they do not allow for the calculation of a meta-synthesis.

All records have been uploaded and summary statistics produced using the R tool bibliometrix (Aria & Cuccurullo, 2017 ). Scopus and Web of Science allow for the download in the bulk of records in .bibtex format, ready to be converted in R objects. Other records are manually entered, depending on the publication status of the single record: for published documents additional research of the specific document is made on Scopus and the relative .bibtex file is downloaded and added to the other results; for unpublished papers, which cannot be found in the two sources, a .bibtex is manually created following the structure of fields and information in the downloaded ready-to-use files. After merging each file and removing duplicates we obtain the data source that contains the bibliographic information of all articles, including publication year/latest draft, author(s), title, journal, keywords, affiliations, and references.

Variants of words or concepts have been aggregated in a unique item i.e. climate change and climatic change or environmental migrants and environmental migration .

The issue of timing will be addressed in the network analysis, choosing a specific type of citation-based network, the bibliographic coupling network, to minimize the risk of missing connections between papers.

Some contributions are not single-case studies.

Asia suffered the highest number of disaster events. In total, between 2000 and 2019, there were 3,068 disaster events in Asia, followed by the 1,756 events in the Americas and 1,192 events in Africa (UNDRR, 2020 ).

Bibliographic coupling, first introduced by (Kessler, 1963b , a ), belongs to the broader class of citation-based approaches to science mapping. Co-citation is based on the relationship established by citing authors of a paper: two papers are linked whenever they jointly appear in the cited references of at least a third paper. Direct citation is the most intuitive approach, linking two papers if one has cited a precedent one. As co-citation, direct citation performs better for long time windows to visualize historical connections (Klavans & Boyack, 2017 ). In terms of accuracy, it has been established that direct citation provides a more accurate representation of the taxonomy of scientific production (Klavans & Boyack, 2017 ), but for the specific requirements the methodology imposes, it has not gained much success (Boyack & Klavans, 2010 ).

Our sampled literature starts in 2003 and ends at the moment the research has been done (November 2020), testifying the recent interest of economic literature on the topic.

It is trivial to observe the value of ties that link a paper with itself, which naturally corresponds to the number of listed references.

This number seems very high, but at a closer look, the two papers that register the highest value are two consecutive papers published by the same author (Naudé, 2008, 2010).

Detailed information on collected coefficients and standard errors are provided in the Supplementary material, Section B.

A summary of the distribution of computed partial correlation coefficients is provided in the Supplementary material, Section C.

The publication bias occurs when (i) researchers, referees, or editors prefer statistically significant results and (ii) it is easier to publish results that are consistent with a given theory. However, the consequences of the peer-review process refer more to a general “publication impact" rather than a “bias" (Cipollina & Salvatici, 2010 ).

The complete description of coded variables is available in Supplementary material, section D.

Potential biases are filtered out sequentially by the addition, in a stepwise manner, of statistically significant controls.

Aria, M., & Cuccurullo, C. (2017). bibliometrix: An r-tool for comprehensive science mapping analysis. Journal of Informetrics, 11 (4), 959–975.

Article   Google Scholar  

Barrios, S., Bertinelli, L., & Strobl, E. (2006). Climatic change and rural-urban migration: The case of sub-Saharan Africa. Journal of Urban Economics, 60 (3), 357–371.

Beine, M. & Jeusette, L. (2019). A meta-analysis of the literature on climate change and migration. IZA Discussion Papers 12639, Institute of Labor Economics (IZA).

Beine, M., & Jeusette, L. (2021). A meta-analysis of the literature on climate change and migration. Journal of Demographic Economics, 87 (3), 293–344.

Beine, M., & Parsons, C. (2015). Climatic factors as determinants of international migration. Scandinavian Journal of Economics, 117 (2), 723–767.

Beine, M., & Parsons, C. R. (2017). Climatic Factors as Determinants of International Migration: Redux. CESifo Economic Studies, 63 (4), 386–402.

Berlemann, M., & Tran, T. (2020). Climate-related hazards and internal migration empirical evidence for rural Vietnam. Economics of Disasters and Climate Change, 4 (2), 385–409.

Blondel, V., Guillaume, J.-L., Lambiotte, R., & Lefebvre, E. (2008). Fast unfolding of communities in large networks. Journal of Statistical Mechanics: Theory and Experiment, 2008 (10), e1008.

Bohra-Mishra, P., Oppenheimer, M., & Hsiang, S. (2014). Nonlinear permanent migration response to climatic variations but minimal response to disasters. Proceedings of the National Academy of Sciences of the United States of America, 111 (27), 9780–9785.

Borenstein, M., Higgins, J. P. T., Hedges, L. V., & Rothstein, H. R. (2017). Basics of meta-analysis: I2 is not an absolute measure of heterogeneity. Research Synthesis Methods, 8 (1), 5–18.

Boyack, K., & Klavans, R. (2010). Co-citation analysis, bibliographic coupling, and direct citation: Which citation approach represents the research front most accurately? Journal of the American Society for Information Science and Technology, 61 (12), 2389–2404.

Brada, J. C., Drabek, Z., & Iwasaki, I. (2021). Does investor protection increase foreign direct investment? A meta-analysis. Journal of Economic Surveys, 35 (1), 34–70.

Cai, R., Feng, S., Oppenheimer, M., & Pytlikova, M. (2016). Climate variability and international migration: The importance of the agricultural linkage. Journal of Environmental Economics and Management, 79 , 135–151.

Cattaneo, C., & Massetti, E. (2019). Does harmful climate increase or decrease migration? Evidence from rural households in Nigeria. Climate Change Economics, 10 (4), 1950013.

Cattaneo, C., & Peri, G. (2016). The migration response to increasing temperatures. Journal of Development Economics, 122 , 127–146.

Cipollina, M., & Salvatici, L. (2010). Reciprocal trade agreements in gravity models: A meta-analysis. Review of International Economics, 18 (1), 63–80.

Coniglio, N., & Pesce, G. (2015). Climate variability and international migration: An empirical analysis. Environment and Development Economics, 20 (4), 434–468.

Doucouliagos, C. (2005). Publication bias in the economic freedom and economic growth literature. Journal of Economic Surveys, 19 (3), 367–387.

Doucouliagos, C., & Ulubasoglu, M. A. (2006). Economic freedom and economic growth: Does specification make a difference? European Journal of Political Economy, 22 (1), 60–81.

Drabo, A., & Mbaye, L. (2015). Natural disasters, migration and education: An empirical analysis in developing countries. Environment and Development Economics, 20 (6), 767–796.

Feng, S., Krueger, A., & Oppenheimer, M. (2010). Linkages among climate change, crop yields and Mexico-US cross-border migration. Proceedings of the National Academy of Sciences of the United States of America, 107 (32), 14257–14262.

Findley, S. (1994). Does drought increase migration? A study of migration from rural mali during the 1983–1985 drought. International Migration Review, 28 (3), 539–553.

Google Scholar  

Gignoux, J., & Menéndez, M. (2016). Benefit in the wake of disaster: Long-run effects of earthquakes on welfare in rural Indonesia. Journal of Development Economics, 118 , 26–44.

Gray, C. (2009). Environment, land, and rural out-migration in the southern Ecuadorian Andes. World Development, 37 (2), 457–468.

Gray, C., & Mueller, V. (2012). Drought and population mobility in rural Ethiopia. World Development, 40 (1), 134–145.

Gray, C., & Mueller, V. (2012). Natural disasters and population mobility in Bangladesh. Proceedings of the National Academy of Sciences of the United States of America, 109 (16), 6000–6005.

Gray, C., & Wise, E. (2016). Country-specific effects of climate variability on human migration. Climatic Change, 135 (3–4), 555–568.

Halliday, T. (2006). Migration, risk, and liquidity constraints in El Salvador. Economic Development and Cultural Change, 54 (4), 893–925.

Halliday, T. (2012). Intra-household labor supply, migration, and subsistence constraints in a risky environment: Evidence from rural El Salvador. European Economic Review, 56 (6), 1001–1019.

Henry, S., Boyle, P., & Lambin, E. (2003). Modelling inter-provincial migration in Burkina Faso, West Africa: The role of socio-demographic and environmental factors. Applied Geography, 23 (2–3), 115–136.

Henry, S., Schoumaker, B., & Beauchemin, C. (2004). The impact of rainfall on the first out-migration: A multi-level event-history analysis in Burkina Faso. Population and Environment, 25 (5), 423–460.

Higgins, J. P., & Thompson, S. G. (2002). Quantifying heterogeneity in a meta-analysis. Statistics in Medicine, 21 (11), 1539–1558.

Hoffmann, R., Dimitrova, A., Muttarak, R., Crespo Cuaresma, J., & Peisker, J. (2020). A meta-analysis of country-level studies on environmental change and migration. Nature Climate Change, 10 (10), 904–912.

Hoffmann, R., Šedová, B., & Vinke, K. (2021). Improving the evidence base: A methodological review of the quantitative climate migration literature. Global Environmental Change, 71 , 102367.

Hornbeck, R. (2012). The enduring impact of the American dust bowl: Short- and long-run adjustments to environmental catastrophe. American Economic Review, 102 (4), 1477–1507.

International Organization for Migration (2021). Environmental migration: Recent trends. migration data portal, international organization for migration.

Jennings, J., & Gray, C. (2015). Climate variability and human migration in the Netherlands, 1865–1937. Population and Environment, 36 (3), 255–278.

Kawawaki, Y. (2018). Economic analysis of population migration factors caused by the great east japan earthquake and tsunami. Review of Urban and Regional Development Studies, 30 (1), 44–65.

Kessler, M. M. (1963). Bibliographic coupling between scientific papers. American Documentation, 14 (1), 10–25.

Kessler, M. M. (1963). An experimental study of bibliographic coupling between technical papers. IEEE Transactions on Information Theory, 9 (1), 49–51.

Klavans, R., & Boyack, K. (2006). Identifying a better measure of relatedness for mapping science. Journal of the American Society for Information Science and Technology, 57 (2), 251–263.

Klavans, R., & Boyack, K. (2017). Which type of citation analysis generates the most accurate taxonomy of scientific and technical knowledge? Journal of the Association for Information Science and Technology, 68 (4), 984–998.

Koubi, V., Spilker, G., Schaffer, L., & Bernauer, T. (2016). Environmental stressors and migration: Evidence from Vietnam. World Development, 79 , 197–210.

Borenstein, M., Hedges, L. V., Higgins, J., & Rothstein, H. (2009). Introduction to meta-analysis . UK: Wiley.

Book   Google Scholar  

Marchiori, L., Maystadt, J.-F., & Schumacher, I. (2012). The impact of weather anomalies on migration in sub-Saharan Africa. Journal of Environmental Economics and Management, 63 (3), 355–374.

Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred reporting items for systematic reviews and meta-analyses: The Prisma statement. Annals of Internal Medicine, 151 (4), 264–269.

Mueller, V., Gray, C., & Kosec, K. (2014). Heat stress increases long-term human migration in rural Pakistan. Nature Climate Change, 4 (3), 182–185.

Mueller, V., Sheriff, G., Dou, X., & Gray, C. (2020). Temporary migration and climate variation in eastern Africa. World Development, 126 , 104704.

Naudé, W. (2008). Conflict, disasters, and no jobs: reasons for international migration from sub-Saharan Africa. WIDER Working Paper Series RP2008-85, World institute for development economic research (UNU-WIDER).

Naudé, W. (2010). The determinants of migration from sub-Saharan African countries. Journal of African Economies, 19 (3), 330–356.

Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., Chou, R., Glanville, J., Grimshaw, J. M., Hróbjartsson, A., Lalu, M. M., Li, T., Loder, E. W., Mayo-Wilson, E., McDonald, S., … Moher, D. (2021). The Prisma 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 29 , 372.

Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., et al. (2021b). The prisma 2020 statement: an updated guideline for reporting systematic reviews. British Medical Journal, 372.

Rao, N., Singh, C., Solomon, D., Camfield, L., Sidiki, R., Angula, M., Poonacha, P., Sidibe, A., & Lawson, E. T. (2020). Managing risk, changing aspirations and household dynamics: Implications for wellbeing and adaptation in semi-arid Africa and India. World Development, 125 , 104667.

Sedova, B., & Kalkuhl, M. (2020). Who are the climate migrants and where do they go? Evidence from rural India. World Development, 129 , 104848.

Spitzer, Y., Tortorici, G., & Zimran, A. (2020). International migration responses to natural disasters: Evidence from modern Europe’s deadliest earthquake . National Bureau of Economic Research Inc. Technical report.

Stanley, T. D. (2001). Wheat from chaff: Meta-analysis as quantitative literature review. The Journal of Economic Perspectives, 15 (3), 131–150.

Stanley, T. D. & Doucouliagos, H. (2012). Meta-regression analysis in economics and business, volume 5. routledge.

Stanley, T. D., & Doucouliagos, H. (2014). Meta-regression approximations to reduce publication selection bias. Research Synthesis Methods, 5 (1), 60–78.

Stanley, T. D., & Doucouliagos, H. (2017). Neither fixed nor random: Weighted least squares meta-regression. Research Synthesis Methods, 8 (1), 19–42.

Sternitzke, C., & Bergmann, I. (2009). Similarity measures for document mapping: A comparative study on the level of an individual scientist. Scientometrics, 78 (1), 113–130.

Thiede, B., Gray, C., & Mueller, V. (2016). Climate variability and inter-provincial migration in south America, 1970–2011. Global Environmental Change, 41 , 228–240.

Thijs, B., Zhang, L., & Glänzel, W. (2015). Bibliographic coupling and hierarchical clustering for the validation and improvement of subject-classification schemes. Scientometrics, 105 (3), 1453–1467.

UNDRR. (2020). Human Cost of Disasters . United Nations: United Nations Office for Disarmament Affairs.

van Eck, N. J., & Waltman, L. (2009). How to normalize cooccurrence data? An analysis of some well-known similarity measures. Journal of the American Society for Information Science and Technology, 51 (8), 1635–1651.

Weinberg, B. (1974). Bibliographic coupling: A review. Information Storage and Retrieval, 10 (5–6), 189–196.

Download references

Open access funding provided by Università degli Studi del Molise within the CRUI-CARE Agreement.

Author information

Authors and affiliations.

University of Molise, Campobasso, Italy

Maria Cipollina

Universitas Mercatorum, Roma, Italy

Luca De Benedictis

University of Macerata, Macerata, Italy

Elisa Scibè

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Maria Cipollina .

Ethics declarations

Conflict of interest.

The authors declare that there are no competing interests.

Additional information

Publisher's note.

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

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file 1 (pdf 232 KB)

Appendix: environmental migration a systematic review and meta-analysis of the literature.

See Tables 5 , 6 , 7 and 8 .

Rights and permissions

This article is published under an open access license. Please check the 'Copyright Information' section either on this page or in the PDF for details of this license and what re-use is permitted. If your intended use exceeds what is permitted by the license or if you are unable to locate the licence and re-use information, please contact the Rights and Permissions team .

About this article

Cipollina, M., De Benedictis, L. & Scibè, E. Environmental migration? A systematic review and meta-analysis of the literature. Rev World Econ (2024). https://doi.org/10.1007/s10290-024-00529-5

Download citation

Accepted : 13 February 2024

Published : 30 March 2024

DOI : https://doi.org/10.1007/s10290-024-00529-5

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

• Climate change
• Natural disasters
• Systematic literature review
• Meta-analysis
• Co-citations

JEL Classification

• Find a journal
• Publish with us
• Track your research

Environmental Economics Research Paper

This sample environmental issues research paper on environmental economics features: 4800 words (approx. 16 pages) and a bibliography with 29 sources. Browse other research paper examples for more inspiration. If you need a thorough research paper written according to all the academic standards, you can always turn to our experienced writers for help. This is how your paper can get an A! Feel free to contact our writing service for professional assistance. We offer high-quality assignments for reasonable rates.

On the political stage, environmental issues are usually placed at odds with economic issues. This is because environmental goods, such as clean air and clean water, are commonly viewed as priceless and not subject to economic consideration. However, the relationship between economics and the environment could not be more natural.

Academic Writing, Editing, Proofreading, And Problem Solving Services

Get 10% off with 24start discount code.

In its purest form, economics is the study of human choice. Because of this, economics sheds light on the choices that individual consumers and producers make with respect to numerous goods, services, and activities, including choices made with respect to environmental quality. Economics is able not only to identify the reasons that individuals choose to degrade the environment beyond what is most beneficial to society, but also to assist policy makers in developing environmental policy that will provide an efficient level of environmental quality.

Because environmental economics is interdisciplinary in nature, its scope is far-reaching. Environmental economists research topics ranging from energy to biodiversity and from invasive species to climate change. However, despite the breadth of the topics covered by the community of environmental economic researchers, a reliance on sound economic principles remains the constant.

This research paper outlines the basic concepts in environmental economics, including the ways in which environmental economists might estimate the value society holds for the natural environment. Further, the corrective instruments that environmental economists can employ to correct for situations in which markets fail to achieve an efficient outcome are closely examined. This research paper also stresses the important role economic analysis plays in today’s most pressing environmental issues.

Environmental goods are those aspects of the natural environment that hold value for individuals in society. Just as consumers value a jar of peanut butter or a can of soup, consumers of environmental goods value clean air, clean water, or even peace and quiet. The trouble with these types of goods is that though they are valuable to most individuals, there is not usually a market through which someone can acquire more of an environmental good. This lack of a market makes it difficult to determine the value that environmental goods hold for society; although the market price of a jar of peanut butter or a can of soup signal the value they hold for consumers, there is no price attached to environmental goods that can provide such a signal.

To some, it may seem unethical to try to place a dollar value on the natural environment. However, there are plenty of cases in which ethics demands just that. Indeed, in cases of extreme environmental damage, such as the 1989 Exxon Valdez oil spill, an unwillingness to apply a value to environmental loss could be considered equivalent to stating that environmental loss represents no loss to society at all. Because of this, the assessment of appropriate damages, fines, or both, in cases such as this often depends on the careful valuation of varying aspects of the environment.

In the case of environmental policy development, insufficient evidence pertaining to the benefit that environmental goods provide to society could easily skew the results of a cost-benefit analysis against environmental protection. This would, in effect, undermine the value that society holds for environmental goods and could possibly lead policy makers to believe that certain environmental regulations are not worth the costs they impose on society when, in fact, they are.

For these reasons, as well as for other reasons that are covered later in this research paper, economists have long endeavored to develop methods of accurately determining the value of environmental goods to society. This effort has led to the development of several valuation techniques.

Valuing the Environment

Contingent valuation.

Contingent valuation, or stated preferences, is a seemingly simple method of valuation that involves directly asking respondents about their values for a particular environmental good. This method is particularly useful in determining the value of environmental goods that individuals have yet to experience or may never actually experience themselves.

The Exxon Valdez oil spill is an example of a case in which contingent valuation provided a useful tool of valuation (Goodstein, 2008). In this case, contingent valuation was used to determine, among other things, the value that individuals place on simply knowing that a pristine Alaskan wilderness exists, even though many respondents may never actually experience this wilderness for themselves (this value is defined as existence value). More generally, contingent valuation methods are often used in policy development to determine the amount respondents would be willing to pay for a new, higher level of environmental quality.

However, despite its simple concept, the contingent valuation method carries with it a host of complex problems that must be taken into account for the results of a survey to be considered credible. These problems usually stem from one or more of the following: information bias, strategic bias, hypothetical bias, and starting point bias (Tietenberg, 2007). Because any type of bias can hinder the usefulness of a contingent valuation survey, special care must be taken to ensure that any bias in the answers provided by survey respondents is minimized.

With information bias, hypothetical bias, and starting point bias, respondents unintentionally misrepresent the value that they hold for an environmental good. With information bias, respondents lack enough information to form an accurate response. To avoid this type of bias, surveyors will usually provide a great deal of information to respondents pertaining to the topic of the survey.

Hypothetical bias occurs because individuals tend to respond differently to hypothetical scenarios than they do to the same scenarios in the real world. One solution to this problem is to conduct the contingent valuation surveys in a laboratory setting (Kolstad 2000). This solution provides the surveyor with an opportunity to remind respondents to consider the financial ramifications that their responses would produce in a real-world setting. It also allows the surveyor to use experimental techniques that mimic the conditions that respondents would face in a real-world situation.

Finally, starting point bias results when respondents are influenced by the set of responses made available to them by the contingent valuation survey. The solution to this problem requires significant pretesting of a survey to ensure that its design does not influence respondents to provide biased answers (Kolstad, 2000).

Unlike the other types of response bias that can occur in a contingent valuation survey, strategic bias occurs as respondents intentionally try to manipulate the outcome of a survey. It is not always possible to eliminate intentionally biased responses. However, in general, it is best to randomly survey a large number of individuals because this will decrease the likelihood that strategic bias will undermine the overall results of the survey.

Revealed Preferences

The revealed preferences method involves determining the value that consumers hold for an environmental good by observing their purchase of goods in the market that directly (or indirectly) relate to environmental quality. For example, the purchase of air fresheners, noise reducing materials, and water purification systems reveal the minimum amount individuals are willing to pay for improved air and water quality. This particular revealed preferences method is referred to as the household production approach.

Economists can also use revealed preferences to determine the value of clean air and clean water through differences in home prices across both pristine and polluted home locations. This particular revealed preferences method is referred to as the hedonic approach (Kolstad, 2000).

These approaches to valuing the environment have the advantage of relying on actual consumer choices to infer the value society holds for a particular environmental good, rather than relying on hypothetical scenarios. However, there are some environmental goods for which it can be nearly impossible to identify their value through market interactions. For example, using the revealed preferences method to determine the value that society holds for the survival of an endangered species would pose a tremendous challenge. In cases such as these, revealed preferences may not be the preferred method of valuation.

Valuation techniques are useful not only in cost-benefit analysis or in cases of extreme environmental damage but also in the more subtle cases of environmental degradation that occur as a result of market failure.

Market Failure

As was discussed in the previous section, individual consumers will often purchase goods with an environmental component to make up for their inability to directly purchase environmental goods, thus revealing the value they hold for certain aspects of environmental quality. For example, someone may buy a cabin on a lake in order to enjoy not only the home itself but also the pristine environment that comes with such a purchase. As long as this individual is able to exclusively incur the environmental benefits that come from owning a log cabin, the demand for log cabins will reflect the full value of both the home and the environmental goods it provides and the market for log cabins will be efficient.

Unfortunately, in the case of environmental goods, markets often fail to produce an efficient result because it is rare that any one individual can incur the full benefit (or cost) of a particular level of environmental quality. This is because environmental goods commonly suffer from the presence of externalities or a lack of property rights.

There are two types of externalities: negative and positive. Negative externalities exist when individuals in society bear a portion of the cost associated with the production of a good without having any influence over the related production decisions (Baumol & Oates, 1988). For example, parents may be required to pay higher health care costs related to pollution-induced asthma among children because of an increase in industrial activity in their neighborhood.

Because producers do not consider these costs in their production decisions, they produce higher quantities of goods with negative externalities than is efficient, leading to more than the socially desirable level of environmental degradation.

As with negative externalities, positive externalities also result in inefficient market outcomes. However, goods that suffer from positive externalities provide more value to individuals in society than is taken into account by those providing these goods. An example of a positive externality can be seen in the case of college roommates sharing an off-campus apartment. Though a clean kitchen may be valued by all individuals living in the apartment, the person that decides to finally wash the dishes and scrub the kitchen floor is not fully compensated for providing value not only to himself or herself but also to the apartment as a whole. Because of this, the decision to clean the kitchen undervalues the benefits of such an action and the kitchen will go uncleaned more often than is socially desirable.

Such is the case with environmental quality. Because markets tend to undervalue goods that suffer from positive externalities, market outcomes provide a level of environmental quality that is lower than is socially desirable.

Corrective Instruments

Once the market inefficiency relating to a particular environmental good is understood, policy makers can correct for this inefficiency by employing any number of policy instruments. Regardless of the instrument, the goal is to provide incentives to individual consumers and firms such that they will choose a more efficient level of emissions or environmental quality.

Command and Control

Command and control is a type of environmental regulation that allows policy makers to specifically regulate both the amount and the process by which a firm is to reduce emissions. This form of environmental regulation is very common and allows policy makers to regulate goods where a market-based approach is either not possible or not likely to be popular. However, these regulations limit the choices that individual firms can make regarding their pollution levels. Because of this, they do not provide firms with an incentive to develop new pollution-reducing technologies (Kolstad, 2000).

The Coase Theorem

Ronald Coase developed the Coase Theorem in 1960, which, although not necessarily a regulatory framework, paved the way for incentive-driven, or market-based, regulatory systems. According to the Coase Theorem, in the face of market inefficiencies resulting from externalities, private citizens (or firms) are able to negotiate a mutually beneficial, socially desirable solution as long as there are no costs associated with the negotiation process (Coase, 1960). This result is expected to hold regardless of whether the polluter has the right to pollute or the average affected bystander has a right to a clean environment.

Consider the negative externality example given previously, in which parents face soaring health care costs resulting from increased industrial activity. According to the Coase Theorem, the firm producing the pollution and the parents could negotiate a solution to this externalities issue, even without government intervention. In this example, if the legal framework in society gave the firm the right to produce pollution, the parents with sick children could possibly consider the amount they are spending on medical bills and offer a lesser sum to the firm in exchange for a reduced level of pollution. This would save the parents money (as compared with their health care costs), and the firm may find itself more than compensated for the increased costs that a reduction in emissions can bring.

If it is the parents, instead, that have a right to clean, safe air for their children (this is more typically the case), then the firm could offer the parents a sum of money in exchange for allowing a higher level of pollution in the area. As long as the sum offered is less than the cost of reducing emissions, the firm will be better off. As for the parents, if the sum of money more than compensates the health care costs they face with higher pollution levels, they may also find themselves preferring the negotiated outcome.

Unfortunately, because the fundamental assumption of the Coase Theorem (costless negotiation) often falls short, this theorem is not commonly applicable as a real-world solution. Despite this fact, the Coase Theorem is an important reminder that even in the case of complex environmental problems, there may be room for mutually beneficial compromises. This theorem also sheds light on cases in which firms are willing to voluntarily comply with environmental regulations as well as on possible solutions to complex international environmental agreements.

In 1920, Arthur C. Pigou (1920) developed a taxation method for dealing with the goods suffering from externalities. The idea behind his tax, now known as the Pigouvian tax, is to force producers to pay a tax equal to the external damage caused by their production decisions in order to allow the market to take into consideration the full costs associated with the taxed goods. This process is often referred to as internalizing an externality.

This concept can also be applied to goods that suffer from positive externalities. However, in this case, a negative tax (or subsidy) is provided to allow an individual to gain an additional benefit from providing the subsidized good. A common example of this type of subsidy can be seen each time an individual receives a tax break for purchasing an Energy Star appliance.

Of course, because the amount of the tax (or subsidy) must equal the value of the external environmental damage (or benefit) in order to correct for market inefficiencies, the valuation techniques detailed previously are crucial in the development of a sound tax policy.

Permit Markets

The concept of using a permit market to control pollution levels was first developed by John Dales (1968). Through this method of regulation, pollution permits are issued to firms in an industry where a reduction in emissions is desired. These permits give each firm the right to produce emissions according to the number of permits it holds. However, the total number of permits issued is limited to the amount of pollution that is allowed industry wide. This means that some firms will not be able to pollute as much as they would like, and they will be forced to either reduce emissions or purchase permits from another firm in the industry (Barde, 2000).

Those firms able to reduce their emissions for the lowest possible cost benefit from this type of regulation. This is because these firms can sell their permits for an amount greater than or equal to the cost of their own emissions reduction, resulting in profits in the permit market. However, even firms for which it is very costly to reduce pollution experience a cost savings through this type of regulation because they are able to purchase pollution permits at a price that is less than or equal to the cost they would face if they were required to reduce emissions. Ultimately, permit markets make it less costly for an industry to comply with environmental regulations and, with the prospect of profits in the permit market, this type of regulation provides an incentive for firms to find less costly pollution reducing technologies.

Applications and Empirical Evidence

Valuing the environment: practical applications.

Both the methods of valuation and the corrective instruments described previously have been applied quite extensively to real-world environmental problems. In fact, according to Barry Field and Martha Field (2006), contingent valuation methods have been used to determine the amount respondents would be willing to pay for a myriad of environmental goods. For example, respondents have been surveyed to determine the value they would place on increased air visibility in places such as the White Mountains (located in New Hampshire) and the Grand Canyon (located in Arizona). Further, contingent valuation methods have been used to determine the value of old-growth forest preservation in the face of industrial expansion (Hagen, Vincent, & Welle, 1992).

Revealed preferences methods have increased in popularity in recent history and are commonly used by researchers to determine the value society holds for clean air and clean water. Though there are many recent cases in which researchers have used revealed preferences methods, Eban Goodstein (2008) provided a particularly useful example of the way in which this method has been used in a real-world setting.

The example given involves the decline in housing prices that occurred in the town of New Bedford, Massachusetts, in the early 1980s following severe contamination of the nearby harbor. Using the hedonic approach, economists were able to determine that those homes closest to the contamination experienced a $9000 reduction in value while the overall loss to homeowners in New Bedford was estimated to be approximately $36 million (Goodstein, 2008).

Although this type of analysis provides only a minimum value of the loss experienced due to the pollution of the harbor, it can be a valuable component in determining an appropriate fine for the firms responsible for the pollution. More generally, these results also shed light on the value that individuals place on clean water.

Corrective Instruments: Practical Application

Though many of the concepts in environmental economics predate the 1970s, the implementation of the Clean Air Act of 1970 represents the first major application of these concepts to government policy. Through these amendments, strict ambient air quality standards were set, and in some cases, specific technologies were required for compliance (Tietenberg, 2007). This regulatory framework is consistent with the command-and-control framework described previously.

However, since the Clean Air Act Amendments of 1990, pollution taxes and permit markets have taken center stage in terms of environmental regulation. In fact, though permit markets were used in the United States as early as the 1970s, the Clean Air Act Amendments of 1990 ushered in an era of increased popularity for this type of regulation by requiring the development of a nationwide permit market for sulfur dioxide emissions.

According to Jean-Philippe Barde (2000), the Environmental Protection Agency implemented a program in response to this requirement that was expected to result in a significant cost savings (20%-50%) as compared with other types of regulation. Further, Thomas Tietenberg (2007) asserted that the development of permit markets increased compliance with federally mandated pollution reduction requirements.

Additional programs have been used to reduce ozone-related emissions, including California’s Regional Clean Air Incentives Market (RECLAIM), established in the Los Angeles basin, and the Ozone Transport Commission NOX Budget Program, which spans approximately 10 states in the eastern United States. (Both of these programs were originally implemented in 1994. However, the NOX Budget Program has since undergone several program modifications, including slight changes to the program name.)

The Ozone Transportation Commission program aimed to reduce nitrogen oxide emissions in participating states in both 1999 and 2003 (U.S. Environmental Protection Agency [EPA], n.d.). The results of this program, as reported by the Environmental Protection Agency, have included a reduction in sulfur dioxide emissions (as compared with 1990 levels) of over 5 million tons, a reduction in nitrogen oxide emissions (as compared with 1990 levels) of over 3 million tons, and nearly 100% program compliance (EPA, 2004).

In terms of taxation programs aimed at reducing pollution levels, Finland, Sweden, Denmark, Switzerland, France, Italy, and the United Kingdom have all made changes to their tax systems in order to reduce environmental degradation. Some of these changes include the introduction of new taxes, such as Finland’s implementation of the 1990 carbon tax; other changes involve using tax revenue to increase environmental quality, such as Denmark’s use of tax revenue to fund investment in energy-saving technologies (Barde, 2000).

In the United States, local grocery markets are at the center of a large tax system aimed at reducing environmental degradation: the deposit-refund system. This system effectively rewards individuals willing to return bottles and cans to an authorized recycling center. Such an incentive represents a negative tax (or subsidy) to individuals in exchange for recycling behavior that benefits society as a whole.

Policy Implications

The policy implications of work done by environmental economists are far-reaching. As countries deal with issues such as water quality, air quality, open space, and global climate change, the methodologies developed in environmental economics are key to providing efficient, cost-effective solutions.

Although command and control remains a common form of regulation, the previous sections detail the ways that countries have begun to use market-based approaches such as taxation and permit markets within the regulatory framework.

Examples of these types of programs continue to develop. For example, in an attempt to comply with the provisions of the Kyoto Protocol, which was implemented to control greenhouse gas emissions, the European Union has established a carbon dioxide permit market aimed at reducing greenhouse gases (Keohane & Olmstead, 2007).

Even the Coase Theorem comes into play as global environmental problems demand mutually beneficial agreements to be voluntarily negotiated across countries. In fact, the Montreal Protocol, which was implemented to control emissions of ozone-depleting chemicals, makes use of a multilateral fund that compensates developing countries for the costs incurred in phasing out ozone-depleting chemicals (Field & Field, 2006). This is very similar to the example in which parents in a community may find it beneficial to compensate a polluting firm in order to induce a reduction in emissions.

Future Directions

Because of the interdisciplinary nature of environmental economics, the discipline constantly presses forward in many directions. Many of the most pressing environmental issues involve both local and global pollutants. These range from local water quality issues to the reduction of greenhouse gas emissions.

In terms of local, regional, and national environmental issues, the application of currently available corrective instruments is quite feasible. However, an evaluation of the value of regulated environmental goods as well as the proposed regulatory instruments is still the topic of ongoing research.

In terms of global issues, such as global climate change, there is still much work to be done regarding the economic impact of changes to the earth’s climate. In addition, solutions relying on government enforcement are less possible when it comes to global climate change. This means that there is likely to be more emphasis placed on voluntary compliance.

For example, in the wake of the Kyoto Protocol, there have been regional agreements that have been formed that have a reduction in greenhouse emissions as a primary goal. One such agreement, known as the Western Climate Initiative, was developed in February 2007. This initiative is a voluntary agreement between seven U.S. states and four Canadian provinces. Its goal is to reduce greenhouse gas emissions by 15% (as compared with 2005 emissions levels) by the year 2020 (Western Climate Initiative, n.d.).

Finally, countries have long suffered from the production decisions of their neighbors. However, since the availability of clean water in the border regions of developing countries remains an issue, solutions to these problems (and similar transborder problems) remain the focus of ongoing research.

Environmental economics provides a set of tools that are crucial in understanding today’s most pressing environmental problems. Through the use of valuation techniques such as contingent valuation and revealed preferences, economists are able to estimate the value society holds for a variety of environmental goods. These values allow policy analysts to consider the impact that a proposed public policy might have on the natural environment. Economists are also able to use these techniques to provide an accurate description of the loss that occurs in cases of both extreme environmental damage and more subtle environmental degradation that occurs daily.

Environmental economics explains the role that externalities play in excessive environmental degradation because the failure of markets to capture the full value of environmental goods consistently results in the overproduction of those goods that can damage the environment and an underprovision of those goods that improve environmental quality. Further, through corrective instruments developed by economists such as Pigou (1920), Coase (1960), and Dales (1968), environmental economics has provided society with innovative solutions to excessive environmental degradation resulting from market failure.

Finally, the application of the techniques developed by environmental economists has become increasingly popular as concern over environmental issues has become a common staple in public policy. As environmental problems continue to become increasingly complex, environmental economists continue to press forward, applying the solutions provided by the fundamentals of economics to these problems.

Bibliography:

• Arrow, K. J., Cropper, M. L., Eads, G. C., Hahn, R. W., Lave, L. B., Noll, R. G., et al. (1996). Is there a role for benefit-cost analysis in environmental health and safety regulation? Science, 272, 221-222.
• Barde, J.-P. (2000). Environmental policy and policy instruments. In H. Folmer & H. L. Gabel (Eds.), Principles of environmental and resource economics: A guide for students and decision-makers (2nd ed., pp. 157-201). Cheltenham, UK: Edward Elgar.
• Baumol, W., & Oates, W. (1988). The theory of environmental policy (2nd ed.). New York: Cambridge University Press.
• Coase, R. H. (1960). The problem of social cost. Journal of Law and Economics, 3, 1-44
• Crocker, T. D. (1965). The structuring of atmospheric pollution control systems. In H. Wolozin (Ed.), The economics of pollution (pp. 61-86). New York: W. W. Norton.
• Dales, J. H. (1968). Pollution, property, and prices: An essay in policy-making and economics. Toronto, Ontario, Canada: University of Toronto Press.
• DeLong, J. V, Solow, R. M., Butters, G., Calfee, J., Ippolito, P., & Nisbet, R. A. (1981, March/April). Defending cost-benefit analysis: Replies to Steven Kelman. Regulation, 39-43.
• Ellerman, A. D., Joskow, P. L., Schmalensee, R., Montero, J.-P., & Bailey, E. M. (2000). Markets for clean air: The U.S. Acid Rain Program. New York: Cambridge University Press.
• Field, B. C., & Field, M. K. (2006). Environmental economics: An introduction (4th ed.). New York: McGraw-Hill.
• Freeman, A. M., III. (2003). The measurement of environmental and resource values (2nd ed.). Washington, DC: Resources for the Future.
• Goodstein, E. S. (2008). Economics and the environment (5th ed.). New York: John Wiley.
• Hagen, D. A., Vincent, J. W., & Welle, P. G. (1992). Benefits of preserving old-growth forests and the spotted owl. Contemporary Policy Issues, 10, 13-26.
• Keohane, N. O., & Olmstead, S. M. (2007). Markets and the environment. Washington, DC: Island Press.
• Kelman, S. (1981, January/February). Cost-benefit analysis: An ethical critique. Regulation, 33-40.
• Kolstad, C. D. (2000). Environmental economics. New York: Oxford University Press.
• Kolstad, C. D., & Freeman, J. (Eds.). (2006). Moving to markets in environmental regulation: Lessons from twenty years of experience. New York: Oxford University Press.
• Maler, K. G. (1974). Environmental economics: A theoretical inquiry. Baltimore: Johns Hopkins University Press.
• Mitchell, R. C., & Carson, R. T. (1989). Using surveys to value public goods: The contingent valuation method. Washington, DC: Resources for the Future.
• Pigou, A. C. (1920). The economics of welfare. London: Macmillan.
• Sen, A. K. (1970). Collective choice and welfare. San Francisco: Holden-Day.
• Shechter, M. (2000). Valuing the environment. In H. Folmer & H. L. Gabel (Eds.), Principles of environmental and resource economics: A guide for students and decision-makers (2nd ed., pp. 72-103). Cheltenham, UK: Edward Elgar.
• Stavins, R. N. (2003). Experience with market based environmental policy instruments. In K. G. Maler & J. Vincent (Eds.), Handbook of environmental economics (Vol. 1, pp. 335-435). Amsterdam: Elsevier Science.
• Stavins, R. N. (Ed.). (2005). Environmental and natural resource economics (6th ed.). Boston: Addison-Wesley.
• Tietenberg, T. H. (1990). Economic instruments for environmental regulation. Oxford Review of Economic Policy, 6(1), 17-33.
• Tietenberg, T. H. (1992). Environmental and natural resource economics. New York: HarperCollins.
• Tietenberg, T. H. (2007). Environmental economics and policy (5th ed.). Boston: Pearson.
• S. Environmental Protection Agency. (2004). Acid Rain Program 2004 progress report. Retrieved January 18, 2009, from http://www.epa.gov/airmarkets/progress/arp04.html
• S. Environmental Protection Agency. (n.d.). Overview of the Ozone Transportation Commission (OTC) NOX Budget Program. Retrieved January 18, 2009, from http://www.epa.gov/ airmarkets/progsregs/nox/otc-overview.html
• Western Climate Initiative. (n.d.). U.S. states, Canadian provinces announce regional cap-and-trade program to reduce greenhouse gases. Retrieved January 18, 2009, from http:// www.pewclimate.org/docUploads/Sept%2023%20PR_0.pdf

ORDER HIGH QUALITY CUSTOM PAPER

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
• Springer Nature - PMC COVID-19 Collection

Twenty Key Challenges in Environmental and Resource Economics

Lucas bretschger.

1 CER-ETH Centre of Economic Research at ETH Zurich, ZUE F7, 8092 Zurich, Switzerland

Karen Pittel

2 ifo Center for Energy, Climate and Resources, ifo Institute and LMU Munich, Munich, Germany

Economic and ecological systems are closely interlinked at a global and a regional level, offering a broad variety of important research topics in environmental and resource economics. The successful identification of key challenges for current and future research supports development of novel theories, empirical applications, and appropriate policy designs. It allows establishing a future-oriented research agenda whose ultimate goal is an efficient, equitable, and sustainable use of natural resources. Based on a normative foundation, the paper aims to identify fundamental topics, current trends, and major research gaps to motivate further development of academic work in the field.

Introduction

Research frontier.

The research agenda in environmental and resource economics has always been very broad and dynamic, reflecting the ways our economies interact with the natural environment. While in classical economics of the eighteenth century the factor land played a dominant role, the effects of pollution externalities, resource scarcities, ecosystem services, and sustainability became important in subsequent time periods. These issues have triggered different waves of research with very prominent results, specifically on optimal policies in the presence of externalities (Pigou 1920 ), optimal extraction of non-renewable resources (Hotelling 1931 ), optimal capital accumulation in the presence of resource scarcities (Dasgupta and Heal 1974 ), and sustainable development (Hartwick 1977 ; Pearce et al. 1994 ). Of course, the list of topics has already been very diverse in the past but has increasingly become so with recent global environmental problems challenging the functioning of a world economy which is growing at a high rate and heavily relies on an international division of labour and trade.

In the past, new research challenges emerged and manifested in different ways: Some topical fields became increasingly relevant due to new technological developments, new ecological or societal challenges or new political agendas. Others arose in fields that were already well researched but rose in importance. Not all challenges were of a topical nature. In some fields, we found our methodological tool-kit not equipped to deal with new problems or in need of extension to find new (and better) answers to old questions. At the same time, it has become increasingly clear that we have to reach out to other disciplines to meet new and often immense challenges. In environmental economics it is key to seek a good balance between disciplinary excellence, interdisciplinary collaboration, and political impact.

Environmental and resource economics is a dynamic field, in which new key topics emerge frequently. So, while the topical and methodological challenges that the paper identifies will be important for some time to come, they will and should also be subject to further development over the next years and decades. The paper aims to identify and address the variety of new complex problems generated by humans when they exploit natural resources and the environment. We specifically identify Twenty Challenges that we feel will be important for environmental and resource economists to address. We are aware that such a list will never be unanimously agreed upon and we do not even lay claim on the list being complete; the next section provides a background to the compilation of the list. Nevertheless, we feel it to be important to (at best) point researchers in directions important to work in in the future or (at least) to launch a new—controversial but productive—discussion on the development of our field. In any case, the paper should support the profession to operate at the research frontier generating novel theories, empirical designs, and workable policies. But, before we turn to the Twenty Challenges , we aim to motivate the framing of research in our field—past, present and future.

Identification of Research Challenges

To provide a normative foundation for our research agenda we characterize our underlying assumptions and generalized views on the nature of research in the field. This set of basic assumptions motivates the criteria of importance, activeness, and distinction of the selected topics as well as our choices with respect to design, methodology and research methods. Identifying the relevant issues, i.e. the mere choice of what to study in environmental economics imposes specific values on the subjects. In our view, the guiding principle in the normative framework is that environmental economics differs from general economics by its ontology, i.e. the system of belief that reflects the interpretation of what constitutes an important fact. It is a deep and serious concern about the state of the natural environment that drives the economic analysis of ecological processes. Nature is not simply part of the economic system but a different system with its own very complex regularities and dynamics; ecosystem values are not reducible to market exchange values. The task to integrate the ecological and economic systems to a holistic framework in an appropriate manner and to derive valid guidelines for the economy under the restrictions imposed by the environment lies at the heart of our research. Central parts of the ontology are the valuation of ecosystems, the increasing scarcities in natural resources and sinks, the effects of environmental externalities, the long-term orientation of planning, an important role of uncertainty, and the existence of irreversible processes. The anthropocentric view and the use of utilitarianism do not imply that individuals are purely self-centered and narrowly selfish. It highlights the indistinguishable role of human decision making for the future of the planet and aims at decision making that cares for efficiency, equity, and posterity. Based on a broad utilitarian setup, growth is not valued in terms of material consumption but in terms of wellbeing, which includes elements like social preferences, work-life balance, appreciation of nature etc. Posterity reflects our care for future generations, whose welfare should not be harmed by the activities of current generations. Fundamental changes of the economy e.g. the phase-out of fossil fuels, includes policy-induced decrease of activities, a role for technology, substitutability in production and consumption, a decoupling from natural resource use, and internalizing cost to correct market failures. Substantive transitions are very difficult to implement, as important lock-in mechanisms such as habit persistence, built infrastructure, and supporting policies such as subsidies stabilize current practices. To achieve a change of mindset in politics to achieve a transition to a green economy is a difficult task. A fundamental systems change, as discussed by many these days, is undoubtedly much more complex to accomplish; its impacts are uncertain and may delay the necessary steps which are important to rapidly improve the state of our ecosystems.

We acknowledge that one can always challenge an ontological position because it reflects ethical principles. In our research agenda there is no external reality, independent of what we may think or understand it to be. We reduce economic and ecological complexity through our personal system of belief to design our preferred map, which by definition is not the territory. In his survey of ecological research issues for the economists, Ehrlich ( 2008 ) refers to his ”own mental meta-analysis” to motivate his choices and to alert us to the importance of research on big issues like the meaning of life, mortality, and death. At the same time, he acknowledges that the emergence of pervasive new environmental problems, such as climate change and biodiversity loss, requires to flexibly adjust research programs to societal demand. Adjustments of the agenda may also be supply driven, when new methods allow for more effective engagement with important issues like risk and uncertainty or assessment of empirical regularities with superior estimation methods.

Forming a Research Agenda

Environmental economics is closely linked to general economics in its epistemology, i.e. the validity, scope and methods of acquiring knowledge by using models, distinguishing between positive and normative models, and testing hypotheses with empirical methods and experiments. An important cornerstone for economic research has always been the analysis of economic efficiency. Since the early days of environmental economics research, this has also held for our field whether it concerned the efficiency in the use of natural resources or the design of policies. Although research in our field has become much more interdisciplinary and policy-oriented, this still constitutes common ground. It is still a prime duty of the economist to point at the potentially vast allocative inefficiencies of the use of natural resources in pure market economies. Efficiency is a necessary condition for optimal states of the economic-ecological system and the foundation for policies maximizing social welfare.

The pursuit of optimality has to be complemented by a requirement to take care of equity and posterity enabling sustainability of development. In this long-run perspective, economics has to highlight the substitution effect as a powerful mechanism establishing consistency between humanity and its natural environment. Substitution comes in many guises, e.g. as substitution between clean and dirty production, renewable and exhaustible resources, extractive and conservationist attitude, pollution intensive and extensive consumption, etc. This dynamic analysis is crucial in many respects. It has recently been included at all levels of research in the fields. The same holds for the issue of risk and uncertainty, a pervasive topic when dealing with the environment.

In many cases, there has been a significant discrepancy between the theoretical derivation of social optima in academia and the attempts to foster their implementation under realistic policy conditions. As a consequence, policies dealing with environmental issues have been of very different quality and effectiveness. The reduction of acid rains, the protection of the ozone layer, and cutbacks of particulate matter emissions in many world regions were among the prominent successes. Global warming, extraction of rare earth elements, and loss of biodiversity are not yet addressed in a comprehensive manner. Political resistance against the protection of nature often refers to the economic costs of policies, including the concerns of growth reduction, employment loss, and adverse effect on income distribution. The lack of success in many policy areas has led to reformulation and extension of the research agenda. In the future, research should focus more on strengthening the links between theory and policy.

Our selection of the Twenty Challenges is also based on the potential of research in these areas to contribute and leverage social welfare and sustainable development. We specifically look for areas that are either inherently new to the research agenda in environmental and resource economics or in which research stagnates. We present the challenges in a specific order and like to highlight the links between them before we enter into the details. The aim of net zero carbon emission by the mid of the century dominates current policy debates and unites basically all important elements of our discipline; it thus constitutes a good starting point. Decarbonization necessarily involves a deep understanding of systems dynamics and of risk and resilience, which are presented next. An important and not sufficiently addressed research issue is the emergence of disruptive development during a substantive transition, the next challenge for our research. Extending the scope, we then address human and government behaviour. In the context of environmental policy, the popular and sometimes underrated request of an equitable use of the environment has emerged as a dominant topic, a next issue for further research. As natural capital involves many more elements than the climate, biodiversity and general ecosystem services are included in the sequence. Broadening the scope to the big problems of human behaviour with natural resources we then turn to political conflicts, population development and conflicting land use. Shifting the focus on induced movements of the labour force we go on by dealing with environmental migration and urbanization. These affect welfare of the individuals in a major way, like health and the epidemiological environment as a next research challenge. In terms of the reorganization of the transition to a green economy we highlight the central role of finance and the implementation of new measures in the dominant energy sector. The final three research challenges are motivated by advances in the methodology. Big data and machine learning offer new perspectives in sustainability research, refined methods and increasing experience improve our simulation models and structural assessment modelling, which forms the last three challenges of our list.

Links to Current Research

In order to put our agenda into a broader perspective and to concretize the selected challenges, we believe it is important to show the relationship between our research agenda and the priorities in current literature and policy debates. We have considered three main links. First, we conducted a quantitative and qualitative literature review and analyzed current research as presented at international conferences (World Conference of Environmental and Resource Economics in 2018, the SURED conference in 2018, Meetings of the American, European, and Asian Associations of Environmental and Resource Economics in 2019). The aim of this analysis was to see where our profession moves and which of the currently hotly debated topics offers a high potential for future research. Second, we took the discussions in interdisciplinary research fora into consideration to identify further fields that are of high importance for future resource use, sustainable development and environmental outcomes but have so far not been adequately addressed from an economics perspective. Information on this research was gained through interdisciplinary research initiatives (for example The Belmont Forum, Future Earth and National Research Funding Activities). Involvement in interdisciplinary and globally oriented research councils provided further access to the discussions in other disciplines. Third, we draw conclusions from current policies and news as well as our involvement in the policy arena. The authors are involved in a number of institutionalized policy-oriented activities on the regional, national and international level (Regional Climate Councils, National Climate Policy Platforms as well as the UN climate negotiations).

The paper relates to similar contributions in recent literature. Based on citation data Auffhammer ( 2009 ) identifies important topics and scholars and provides a brief historical overview of the discipline from exhaustible and renewable resources to sustainability, pollution control, development, international trade, climate change, international agreements, and non-market valuation. Polyakov et al. ( 2018 ) analyze authorship patterns using text analysis for classification of articles in Environmental and Resource Economics. Based on 1630 articles published in the Journal from 1991 to 2015 they document the importance of applied and policy-oriented content in the field. They identify non-market valuation, recreation and amenity, and conservation, as popular topics and growing when measured by both number of articles and citations. Costanza et al. ( 2016 ) investigate the most influential publications of Ecological Economics in terms of citation counts both within the journal itself and elsewhere. Important topics turn out to be social aspects of environmental economics and policy, valuation of environmental policy, governance, technical change, happiness and poverty, and ecosystem services. A contemporary analysis of how research issues have developed in the Journal of Environmental Economics and Management in the time of its existence is provided by Kubea et al. ( 2018 ). These authors show that the sample of topics has broadened from the core issues of non-market valuation, cost-benefit analysis, natural resource economics, and environmental policy instruments to a more diversified array of research areas, with climate change and energy issues finding their way into the journal. In addition, increasing methodological plurality becomes apparent. They conclude that energy, development, and health are on the rise and that natural resources, instrument choice, and non-market valuation will endure; multidisciplinary work will be increasingly important. An excellent survey on research in the central field of sustainable development is provided in Polasky et al. ( 2019 ), which explicitly shows where the collaboration between economists and the other disciplines is currently insufficient and how it should be intensified in the future.

Regarding the literature that we connect our Twenty Challenges to, we naturally face the problem that some challenges have so far not been addressed adequately in the (economics) literature. In these cases we also reference papers from other disciplines. We, however, also take basic literature and recent research in environmental and resource economics into account. As we often deal with emerging topics, we cite some of this work even when not yet published. In other cases, where future research can build on or learn from past research, we also go back in time and reference older papers. Ultimately, neither our list of challenges nor the literature we base our analysis on will be satisfying to everybody. Our selection cannot be comprehensive and does not claim to be. But the specific task to identify future-oriented topics ultimately lasts on a subjective individual assessment of the authors. Nevertheless, hopefully it imparts impulses for future research in the different subfields of environmental and resource economics.

Twenty Challenges

The ordering of the following challenges should not be understood to perfectly reflect their individual importance (beyond what we explained in the previous sections). Also, many of the fields discussed are inherently related, creating some unavoidable overlap. We feel that efforts to bring the challenges into some complete ’natural order’ are not only doomed to fail but also would not do them justice as they relate to very different areas and can/should not be weighed against each other. Also, attempting to show their interrelations would result in a 20-by-20 matrix that would not provide more clarity.

• Deep decarbonization and climate neutrality To limit global warming to a maximum of 1.5 degrees Celsius, a state of net zero greenhouse gas emissions—i.e. climate neutrality—should be reached by the mid of the century (IPCC 2018 ). The directly following and unprecedented challenge is to decarbonize the global economy in very a narrow time window (Hainsch et al. 2018 ). This holds especially as the threshold for 1.5 degrees is expected to be passed around 2040 (IPCC 2018 ). Countries must increase their NDC ambitions of the Paris Agreement more than fivefold to achieve the 1.5 degree goal (UN - United Nations 2019 ). The time window for necessary decisions is closing fast. Infrastructure that is installed today often has a life span that reaches until and beyond 2050. Decisions on investments today therefore affect the ability to reach climate targets not only in 2030 but also 2050 and beyond. And while the necessity of reaching net zero emissions by mid century is reflected by, e.g., the European Commission’ Green Deal, much uncertainty remains regarding its implementation. This holds to an even larger extent with respect to other countries and regions. The fundamental challenge is to better understand economically viable deep decarbonization paths and then to implement incentives for input substitution, technology development, and structural change. More specifically, the vision of these policies has to be long-term and reach beyond phasing out coal and increasing energy efficiency. However, despite recent research efforts in climate economics, many issues around decarbonization, negative emissions and economic development are still controversial or insufficiently understood by economists. Specifically, industry applications for which alternative technologies are not available yet as well as agricultural emissions will have to be addressed. Also, the later greenhouse gas emissions start to fall, the faster their decline will have to ultimately be in order not to overshoot temperature targets (Agliardi and Xepapadeas 2018 ), leading to an increased need for negative emissions. However, potential trade-offs and synergies in the use of land for negative emission technologies, food production and biodiversity are still underresearched. Identifying technologies today that are the most promising in the very long run is subject to high uncertainty. Yet, while investing too early might be costly, delaying investment might cost even more or might lead to a weakening of future climate targets (Gerlagh and Michielsen 2015 ). Also, transition processes may involve strong scale effects implying nonlinear development of abatement cost. Once certain thresholds are reached, lower abatement cost or even disruptive development completely altering the production process could emerge in a later phase of decarbonization. Given the dramatic increase needed in mitigation efforts to reach the 1.5 or even 2 degree target, more attention also has to be devoted to the question of adaptation. Until today, the focus of research as well as policy has been primarily on mitigation rather than adaptation, partially because of expected substitution effects between mitigation and adaptation and partially because adaptation was taken to be automatic (Fankhauser 2017 ). However, as Fankhauser lays out “knowledge gaps, behavioral barriers, and market failures that hold back effective adaptation and require policy intervention”. All of these topics present a wide scope for substantial further research.
• Dynamics of the economic-ecological system Depletion of exhaustible resources, harvesting of renewable resources, recycling of raw materials, and accumulation of pollution stocks require basic societal decisions which are of an inherently dynamic nature. Whether the world society will be able to enjoy constant or increasing living standards under such dynamic natural constraints depends on another dynamic process, which is the accumulation of man-made capital. To derive the precise laws of motion in all the stock variables is challenging because general solutions of dynamic systems with several states are usually hard to obtain. An adequate procedure to obtain closed-form solutions may be to link several stocks in a reasonable way, e.g. when simultaneously dealing with resource, pollution, and capital stocks (Peretto 2017 ; Bretschger 2017b ). The specific challenge is then to find the best possible economic justification to motivate the links. One may also focus on a few stocks which are considered the main drivers of economic development and sustainable growth on a global scale (Marin and Vona 2019 ; Borissov et al. 2019 ). When resorting to numerical simulation methods it is a main challenge to provide basic economic results which are sufficiently robust and supported by ample economic intuition. Social-ecological systems are increasingly understood as complex adaptive systems. Essential features of these systems - such as nonlinear feedbacks, strategic interactions, individual and spatial heterogeneity, and varying time scales—pose another set of substantial challenges for modeling in a dynamic framework. A main challenge is the characterization and selection of dynamic paths with multiple equilibria and the overall tractablility of the models, given the diversity of interlinkages and nonlinear relationships. The complexity of economic-ecological systems lead to a main challenge for designing effective policies is taking account of network effects, strategic interaction, sectoral change, path dependencies, varying time lags, and nonlinear feedbacks have to be considered as well as different regional and temporal scales, interdependencies between ecosystems, institutional restrictions and distributional implications (see, e.g., Engel et al. 2008 ; Levin et al. 2013 ; Vatn 2010 ). Optimal policies should also acknowledge the balance between the preservation of the ecology and the development of the economy especially for countries growing out of poverty. Setting a price for ecosystem services and natural capital via policy is important for preventing innovation incentives from being skewed against maintaining natural capital and ecosystem services.
• Risk, uncertainty, and resilience The vast majority of contributions in environmental economics use models with a purely deterministic structure. However, large negative environmental events require a completely different framework, which poses specific challenges for modelling. Heatwaves, floods, droughts, and hurricanes are shocks that are very uncertain, arriving at irregular times and with varying intensity. Also, risk and uncertainty about socio-economic impacts and technological development affect the optimal design of policies (see, e.g., Jensen and Traeger 2014 ). Moreover, uncertainty changes the political economy of climate policy and, finally, regulatory and policy uncertainty might create obstacles to reach climate targets through, for example, distortions of investment decisions (Pommeret and Schubert 2018 ; Bretschger and Soretz 2018 ). Stern ( 2016 ) argued forcefully that climate economics research needs to better integrate risk and uncertainty. Bigger disasters or so-called ”tipping points” such as the melting of the Greenland ice sheet, the collapse of Atlantic thermohaline circulation, and the dieback of Amazon rainforest involve an even higher level of uncertainty (Lenton and Ciscar 2013 ) with implications for optimal policy design and capital accumulation (Van der Ploeg and de Zeeuw 2018 ). Understanding the implications of tipping points is further complicated as the different tipping points are not independent of each other (Cai et al. 2016 ). The Economy and the Earth system both form non-deterministic systems; combining the two in an overarching framework and adding institutions for decision making multiplies the degree of complexity for adequate modelling and methods (Athanassoglou and Xepapadeas 2012 ). It is thus a main challenge for further research to provide analytic foundations and policy rules for rational societal decision-making under the conditions of risk and uncertainty up to deep uncertainty (Brock and Xepapadeas 1903 ; Baumgärtner and Engler 2018 ). Future work on policy design under deep uncertainty can build on a wide range of literature ranging from the assessment of the precautionary principle in this context to the fundamental contributions by Hansen and Sargent ( 2001 ) and Klibanoff et al. ( 2005 ) as well as on more recent analyses in the context of environmental and resource economics, e.g. Manoussi et al. ( 2018 ). An important challenge of the environmental discipline is to provide a framework for the global economy providing the conditions for resilience against major shocks and negative environmental events (Bretschger and Vinogradova 2018 ). With deep uncertainty one has to generate rules for deep resilience. Including uncertainty is especially important when environmental events do not occur constantly but cause the crossing of tipping points involving large and sudden shifts. Economic modeling needs to increasingly incorporate tipping points and the value of resilience in theory and to generate and use data supporting the empirical validity. The combination of uncertainty and potential irreversible outcomes (e.g., species extinction) is another big challenge for research.
• Disruptive development and path dependencies Substantial and sometimes disruptive changes in behavioral patterns, economic structure and technologies will be required if net zero GHG emissions and the UN sustainable development goals are to be reached. On the bright side, development may exhibit favorable disruptions. Consumers’ preferences and political pressure coupled with new technology achievements may alter certain sectors in a short period of time. Similar to the communication industry which has completely changed, transportation and heat generation could and mst probably will undergo fundamental changes in the near future. The research challenge here is to provide adequate models predicting and adequately analyzing such important transitions and to highlight resisting forces at the same time. In fact, the change of trajectories in development is often hampered by technological, economic and behavioral lock-ins, resulting in path dependencies and inertia. In such situations, history influences current development through, for example, past investment in R&D, the size of established markets, increasing returns or habits acquired (Aghion et al. 2016 ; Barnes et al. 2004 ; Arthur 1989 ). Behavioral path dependencies affect acceptance and adoption of new technologies, hinder social innovation and might render policies aimed at marginal changes ineffective. They can thus postpone the transition to a low-carbon economy, harm efforts in biodiversity conservation and prolong unsustainable resource use patterns and lifestyles, even if they are welfare enhancing in the long-run (e.g. Acemoglu et al. 2012 ; Kalkuhl et al. 2012 ). Inertia and lock-ins may also be policy driven with, for example, political or economics elites trying to block change (Acemoglu and Robinson 2006 ) or clean energy support schemes fostering new technology lock-ins. Whether disruption or a lock-in emerges depends, for example, on expectations determining the steady state of an economy (Bretschger and Schaefer 2017 ). This requires nonlinearities e.g. in capital return, generating overlap regions in which the growth path is indeterminate and could be either driven by history or by expectations. The challenge is to add more substantial research into system dynamics and the political economy of change, to gain a better understanding of the different mechanisms responsible for inertia and disruptive change. So far, the role of path dependencies has often been neglected in empirical as well as theoretical analyses (Calel and Dechezlepretre 2016 ). Also, understanding the triggers or tipping points for disruptive change can help to identify policies that have a big environmental impact with moderate costs in terms of environmental policy.
• Behavioral environmental economics Traditionally, economics focuses predominantly on the supply side when analyzing potentials and challenges for environmental policies. Preferences of individuals are mostly assumed to be given with economic analysis confining itself to studying the effects of changing incentives and altering constraints. The change and development of preferences over time plays only a comparative minor role for economic research. Also, the follow-up question whether policies should be allowed to tamper with preferences is rarely discussed with nudging being one big exception to this rule (e.g. Strassheim and Beck 2019 ). While the traditional, supply-side oriented analysis has provided powerful results in positive analysis, it proves to be limited in a field which inherently includes normative conclusions like environmental economics. The path toward sustainable development requires behavioral changes and political actions changing our relationship to the environment. Ultimately, environmental policies have to be decided by the same people overusing the environment in the absence of a policy. In situations where outcomes are inefficient because individuals and political actors follow their own self-interest and ignore external costs and benefits of their actions, it is clearly not sufficient for economists to advocate the implementation of environmental policies. It is crucial to understand under what conditions preferences change and agents support green policies (Casari and Luini 2009 ). So, the challenge to economic research is to better understand the evolution of green attitudes, the emergence of preferences for a clean environment, and expectations in the case of multiple equilibria (Cerda Planas 2018 ). The formation and development of preferences is also not independent from cultural, regional and community aspects. Research that ignores heterogeneity among actors or the role of social and group dynamics and only relies on the traditional, isolated analysis of individual preferences is likely to lead to an incomplete understanding of preference dynamics. As the example of discounting shows, the social context has an impact on myopic attitudes and the motivation to undertake sacrifices for a cleaner future (Galor and Özak 2016 ). Also, attention to behavioral details, that economists might find rather uninteresting from a research perspective, might influence effectiveness of policies tremendously (Duflo 2017 ). Especially with the natural environment, the choice and guise of policy instruments should take these mechanisms into account.
• Institutional analysis of environmental policy Virtually every contribution to the environmental and resource economics literature culminates in one or several policy conclusions. However, these results are often received with skepticism from industry and public. Therefore, a continuing key challenge for our profession is a thorough understanding of environmental policy institutions, processes and decision-making; this task has become even more important given the enormous scale and global nature of future policies. Research in this area has, however, the advantage of already looking back on a long tradition (see e.g. the body of work by Daniel Bromley, e.g. Bromley 1989 ). Well-designed institutions support and create incentives to drive development toward a welfare-improving state. Absent, weak, inefficient, or even corrupt governments and institutions are detrimental to successful environmental policy (Pellegrini and Gerlagh 2008 ; Dasgupta and De Cian 2016 ) or might lead to detrimental effects of resource wealth (see Badeeb et al. 2017 for an overview of the related literature). To effectively increase social welfare by, for example, conservation of ecological services, one has to design policies in a way that allow implementation under realistic policy conditions (Rodrik 2008 ). Pure reference to the construct of a social planner is not sufficient. For increasing efficiency in problem solving, the ex-post evaluation of policies has to be expanded and improved. Policy evaluation should not only analyze if regulatory objectives have been reached but also which side-effects arise (OECD 2017 ). Moreover, the comparison with alternative measures and a continuous international exchange of best practices have to be supported by science. A proactive environmental policy analysis should furthermore include studying vested interests, lobbying, political power, policy communication, and voting behavior. Especially insights from behavioral economics may add to our understanding of a proper design of environmental institutions. On the international level, the adequate institutional design for global environmental policy still poses great challenges. Beyond traditional research fields like international environmental agreements in specific areas like climate change, the multi-dimensionality of the sustainable development goals (SDGs) and potential trade-offs between different goals need to be explored further. This holds especially given the vast differences in income, vulnerability, and resilience between countries, as well as the need for unanimity and voluntary contributions on the UN level. Relating national to international policies has the potential to be especially rewarding in this context given the SDGs relevance for and acceptance in national as well as international politics. Insights from the analysis of institutions in traditional economic sectors (e.g. on the efficiency of capital markets) should be transferred and applied to the global level (e.g. with respect to investment in the world’s natural capital stock).
• Equitable use of the environment We place equity and fairness in dealing with the natural environment on the priority list of our challenges because first and foremost equity is a central requirement for sustainability of development. By definition, sustainable development seeks an equitable treatment across different generations as well as agents living today. We also believe that for successful environmental policies, equity and fairness are crucial complements to the dominant efficiency requirement (Sterner 2011 ). It is a specific challenge of our field to study equity in an economic context and to demonstrate its importance for sustainability to mainstream economics and the public. The first aspect of the problem is the aforementioned unequal vulnerability of countries to environmental changes such as global warming. If vulnerability is higher in less developed countries, the equity perspective is especially striking. As a matter of fact, most of the climate vulnerable countries have a low average income. Global environmental policy is then motivated not only by efficiency but also by the aim of preventing increasing inequalities (Bretschger 2017a ). Global efforts are also indicated to avoid adverse feedback effects of induced inequalities like environmental migration. The second aspect is that acceptance of public policies sharply increases with the perceived fairness of the measure (Pittel and Rübbelke 2011 ; IPCC 2018 ). In the past, economists have often underestimated political resistance against efficient environmental protection, which was mostly related to negative impacts on income distribution. Take carbon pricing and emission regulation as a current example. Although evidence from cross-country studies suggests that regressivity of carbon pricing is much less frequent than often assumed in the public (Parry 2015 ), the perceived distributional impact is often very different (Beck et al. 2016 ). Therefore the impact of environmental policies on income groups, regions, and countries should be better integrated in our analysis and policy recommendations. Where efficient policies are regressive, economists have to evaluate and propose alternative or complementary policy designs. Benefits and costs need to be disaggregated by group (country) with a special attention on the poorest members of society (countries). Internationally, equity concerns need to be addressed especially in situations where the entire world benefits from the protection of natural capital and ecosystem services in poor countries (e.g., of carbon sinks and biodiversity hubs like tropical rain forests). The experience with the REDD+ process shows the complexity of designing such international approaches to incentivize and enable developing countries to protect these global public goods. More economic analysis is needed on all of the above aspects, giving rise to a rich research agenda in theory and applied work.
• Loss of biodiversity and natural capital The rate of species extinction today is estimated to be up to 1000 times higher than without human interference (Rockstrom 2009 ). Human activities impact biodiversity through land use change, pollution, habit fragmentation and the introduction of non-native species but also increasingly through climate change and its interaction with already existing drivers of biodiversity change (IPCC 2002 ). In view of this, biodiversity conservation has long been a focus of politics. In 1992, the United Nations Convention on Biological Diversity main objectives were stated as ”the conservation of biological diversity, the sustainable use of its components and the fair and equitable sharing of the benefits arising out of the utilization of genetic resources” (UN - United Nations 1992 ). Yet, although economists have developed conceptual and theoretical frameworks addressing the valuation of biodiversity (Weitzman 1998 ; Brock and Xepapadeas 2003 ) and despite data on valuation having become increasingly available (see, e.g. TEEB 2020 ), Weitzman ( 2014 ) points out, that an objective or even widely agreed measure of biodiversity and its value is still missing. The same holds for an underlying theory framework and a comprehensive measure of natural capital that not only includes biodiversity but also its links to regulating services (e.g., pollution abatement, land protection), material provisioning services (e.g., food, energy, materials), and nonmaterial services (e.g., aesthetics, experience, learning, physical and mental health, recreation). How biodiversity and natural capital should be measured, which societal, political and economic values underlie different measures and valuation and how ecological and economical trade-offs should be dealt with are big challenges left for future research. In order to address these issues, not only do we need to develop appropriate assessment methods, but we also need to disclose the theoretical basics of this assessment and which trade-offs go hand in hand with different assessments (Brei et al. 2020 ; Antoci et al. 2019 ; Drupp 2018 ). Completely new issues for the valuation of biodiversity and natural capital arise with the development of new technologies. Take DSI (digital sequence information), for example. DSI are digital images of genetic resources (DNA) that can be stored in databases. This gives rise not only to new challenges regarding their valuation but also about the fair and equitable sharing of the benefits arising out of the utilization of these resources.
• Valuing and paying for ecosystem services Related to the question of biodiversity valuation is the market and non-market valuation of ecosystem services in general and the adequate design of payment for ecosystem services (PES). Overall, research on ecosystem services valuation has made significant progress in the last decades. Nevertheless, challenges remain even in traditional valuation fields (for example, valuation of non-use or interconnected ecosystems). Other, so far underresearched areas that constitute promising fields for future research are health-related valuation aspects (Bratman et al. 2019 ) and nonmaterial ecosystem services, such as amenities of landscapes or cultural ecosystem services (Small et al. 2017 ; James 2015 ). Also, data availability remains a problem in many valuation areas. Although digitized observation and information systems offer large potentials for previously unknown data access, they also raise a whole slew of new ethical, privacy as well as economic questions, especially in areas like health. While a lot of progress has been made in the valuation of ecosystem services, their impact on decision making still lags behind. One factor contributing to this disconnect are prevalent mismatches between regional and temporal scales of economic, institutional and ecological systems that make valuation and policy design complex (Schirpke et al. 2019 ). The challenge is to develop combined natural science-economic models that allow better insights into how changes in economic systems lead to changes in the flows of ecosystem services and vice versa (Verburg et al. 2016 ). This requires a deep understanding of ecological and economic systems as well as other aspects like technologies, regional heterogeneity and system boundaries, i.e. catastrophic events. It also raises classic economic problems, such as choosing an appropriate discount rate and degree of risk aversion. Regarding tools to include ecosystem services in economic decision making, PES are a, by now, well-established (Salzman et al. 2018 ) and also quite well-researched approach for promoting environmental outcomes. Still, the literature has identified a number of aspects to be addressed in the design of PES to make them more effective as well as efficient and to simultaneously improve social outcomes (Wunder et al. 2018 ; Chan et al. 2017 ). A promising area of research rarely addressed are PES to preserve transboundary or global ecosystem services through international payment schemes (for example, in tropical forest preservation). While some work has been done on the conceptual level (e.g. Harstad 2012 ), the REDD+ process (Maniatis et al. 2019 ) and the failure of the Yasuni initiative (Sovacool and Scarpaci 2016 ) show the complexity of such approaches for which a thorough economics analysis is still missing.
• Conflicts over natural resources Climate change and decarbonization transform regional and global geopolitical landscapes and might give rise to future domestic as well as international conflicts (Mach et al. 2019 ; Carleton and Hsiang 2016 ). First, decarbonization changes the role of resources and of resource- and energy-related infrastructures. Climate policies affect the rent allocation between different fossil fuels like, for example, coal and natural gas, but might also change the overall rent level (Kalkuhl and Brecha 2013 ). Asset stranding can endanger stability in resource (rent) dependent countries. Conflicts may also arise over materials critical to new, low-carbon energy technologies like rare earth elements but also over access to sustainable energy (Goldthau et al. 2019 ; O’Sullivan et al. 2017 ). Further research is needed to design policies that are better equipped to reduce the vulnerability of economies to changes in resource availability and resource rents. This opens up challenges for future research, especially as restrictions from very diverse institutional capacities have to be considered to render policies efficient and effective. Second, climate change will affect the ability to meet basic human needs through food, land and water. Sulemanaa et al. ( 2019 ) find a positive effect of the occurrence of temperature extremes on conflict incidence. They stress the need for more advanced spatial econometric models to identify effects that are transmitted across space. More research is also needed on the role of institutions and interaction with other phenomena like population dynamics, migration, and environmental degradation. Currently, the role of climate for conflict is still small compared to other causes, many linkages between conflicts and climate change as well as other factors promoting conflict are still uncertain (Mach et al. 2019 ). The challenge to economic research is to get early insights into the nexus of historical and cultural factors, vested interests, population dynamics and climate change in order to help to prevent resource-related conflicts.
• Population development and use of the environment Already since antiquity, demographic analysis has been a central topic of human thinking. With the Malthusian predictions of catastrophes caused by population growth, the topic is firmly related to the natural environment and the limits of planet Earth. While limited food production was the dominant topic in the 18th century, the impact of world population on global commons, availability of renewable and exhaustible resources, and ecosystem services have been dominant topics in the last decades. Still, while it is often argued in the public and in natural sciences that world population size should be a concern because of ecological constraints, economics has largely left the topic on the side; the few exceptions (Peretto and Valente 2015 ) and (Bretschger 2013 , 2020 ) point in a different direction, namely the compatibility of population growth and sustainable development under very general conditions. Current trends of demographic transition show significant signs of population degrowth for leading economies while trends for developing countries vary substantially (UN - United Nations 2019 ). Population is forecasted to expand especially in Africa, accounting for more than half of the world’s population growth over the coming decades, raising questions about the effect of this population increase on fragile ecosystems, resource use and ultimately the potential for sustainable growth (African Development Bank 2015 ). Population growth will also promote further urbanization and migration triggered by environmental and resource depletion but also giving rise to new environmental problems (Awumbila 2017 ). Challenges from population development and environment are thus closely linked to the other research topics highlighted in this article. However, population growth is not exogenously given but determined by economic, social as well as environmental factors. Education and income or economic development have long been established as crucial for fertility (see e.g. the reviews of the literature provided by Kan and Lee 2018 ; Fox et al. 2019 ). To integrate these findings into a holistic approach is a mediating challenge for future research. Climate change might affect these channels in different ways, potentially exacerbating global inequality (Casey et al. 2019 ). However, population development, fertility, and mortality are not only affected by climate change but also by other environmental stresses like air pollution (Conforti et al. 2018 ). A successful combination of endogenous fertility and mortality with natural resource scarcity, agricultural production, and pollution accumulation as well as capital and knowledge build-up in a comprehensive framework is a respectable challenge for an economic modeller; we suggest that in the future it should be considered by economists more intensively.
• Land use and soil degradation The terrestrial biosphere with its products, functions and ecosystem services is the foundation of human existence, not only for food security but far beyond. Currently, about a quarter of ice-free land area is degraded by human impacts (IPCC 2019 ). The optimal use of scarce land resources becomes an even more urgent topic in the face of the biodiversity crisis and the onset of climate change. This holds especially as the physical and economic access to sufficient, safe and nutritious food is the basic precondition for human existence. Climate change challenges this access on different levels. On the one hand, climate change increases the pressure on productive land areas (due to extreme weather events such as droughts, floods, forest fires or the shifting of climatic zones). On the other hand, land plays a major role in many climate protection scenarios by reducing emissions from land use and land use change, protecting carbon stocks in soils and ecosystems, and conserving and expanding natural carbon sinks. Also, the capture and storage of CO 2 through carbon dioxide removal technologies plays an increasing role for reaching the Paris climate goals (IPCC 2018 ). The induced increase in the demand for the different services from land inevitably implies trade-offs. However, neither the trade-offs nor the potentials for synergic uses are, as of now, comprehensively understood from an economic point of view and thus pose a challenge for future research. While there is a growing literature on negative emission technologies, their costs, potentials and side effects (Fuss et al. 2019 and references within) as well as on the interaction between climate goals and other SGDs on the global level (von Stechow et al. 2016 ), many research questions still remain to be addressed (Minx et al. 2018 ). This concerns especially a better understanding of opportunity costs, governance requirements, regional and distributional effects as well as of acceptance and ethical considerations. With respect to land degradation and land use for food production, changing climate and weather conditions as well as regional population pressure may raise the rate of land degradation (Fezzi and Bateman 2015 ), hurting food security and calling for preservation policies (Brausmann and Bretschger 2018 ). The overuse of ecosystems like forests and water, which protect and complement land, can accelerate the risk of adverse shocks and thus lower soil fertility, which reveals the close link between the different research subjects. However, much of the agricultural research in this field is still quite distant from mainstream environmental economics which can harm research productivity substantially. It remains a challenge to integrate agricultural and environmental research better, for example by bringing together food production, population, and the environment into a macrodynamic framework (Lanz et al. 2017 ).
• Environmental migration Migration in times of climate change is an extraordinarily complex, multicausal and controversial challenge (Adger et al. 2014 ). Heatwaves, droughts, hurricanes, and rising sea levels are likely to motivate or even force a growing number of people to leave their homes moving to presumably safer places. Climate-related migration can take a variety of different forms (Warner 2011) from voluntary to involuntary, from short- to long-distance and from temporary to permanent. Migration decisions are usually based on different motives and personal circumstances (climatically, politically, economically, socially), leading to heterogeneous reactions to climate events and making it often problematic to identify and delineate climate-induced migration. Due to these and other methodological difficulties and the small number of studies so far, no globally reliable forecasts for climate induced migration exist (WBGU - German Advisory Council on Global Change 2018a , b ). At present, the forecasted magnitude of the phenomenon ranges from 25 million up to 1 billion people by 2050 (Ionesco et al. 2017 ). Much of this migration can be expected to take place within countries, for example, from rural to urban areas or from drylands to coastal zones (Henderson et al. 2014 ) with environmental migration being one possible adaptation and survivor strategy in the face of climate change (Millock 2015 ). Given the uncertainty in future migration projections, the challenge is to improve migration models (Cattaneo et al. 2019 ) which includes a better understanding and integration of the microfoundation of agents’ migration decisions. Migration, and especially mass-migration, can have a profound impact on the environment of the new as well as the old settlement location and on their economic structure. Labor and commodities markets will be affected the most, with challenges arising also for education and health systems, government budgets and public spending. By affecting public institutions and the skill-mix of the labor force, migration alters economic development both in the sending and in the receiving countries or regions. More research is needed on these impacts. The influx of environmental migrants to new settlement locations may also trigger hostile attitudes and lead to clashes and even armed conflicts. The migrants may be perceived as rivals for scarce resources (land, clean water) or jobs. The situation may be aggravated by lack of political stability and poor-quality political institutions. Dealing with these aspects gives rise to new challenges in environment and resource economics. Traditional analysis of economic costs and benefits of migration have to be complemented by behavioral economic and political economy analyses.
• Urbanization as a key for environmental development In the last 70 years, the urban population has increased fivefold with more than half of the world’s population living in cities today and forecasts projecting the share of urban population to rise to almost 70% in 2050 (UN - United Nations 2018 ). Cities are responsible for about 70% of the world energy use and global CO 2 -emissions (Seto et al. 2014 ) and ecological footprints are positively correlated to the degree of urbanization (WBGU - German Advisory Council on Global Change 2016 ). In 2014, about 880 million people were living in slums (UN - United Nations 2016 ) elucidating the problems to make urban development environmentally as well as economically and socially sustainable. The speed of urbanization is projected to be the fastest in low and middle income countries, especially in Africa and Asia (UN - United Nations 2018 ), leading to new challenges for the provision of infrastructure, housing, energy supply, transport and even health care. Climate change can be expected to not only foster urbanization trends (Henderson et al. 2017 ) but also increase the magnitude of urbanization-related challenges. Urban areas are often located close to the coast or rivers basins, making them susceptible to rising sea levels and impacts of extreme weather events. Risks can be expected to be higher for poor households due to settlement in less safe areas and poorer housing (Barata et al. 2011 ), potentially perpetuating existing inequalities. On the other hand, cities might offer more efficient adaptation potentials. To date the consequences of climate change for cities and urbanization are still to be determined in detail but depend heavily on factors like location, size and level of development as well as governance capacities. Making cities, their population and their infrastructure resilient to climate change will be decisive for future development. The main challenge here is to better connect the research fields of environmental and urban economics to understand the drivers and dynamic effects of climate change on urbanization and resulting economic development, on adaptation costs and benefits and on the role of institutions. Insights from regional, political and behavioral economics can help shape effective governance to enhance resilience of cities to climate change.
• Health and epidemiological environment Environmental degradation can have profound implications for human health. These implications lead to direct as well as indirect challenges for economic decision making, economic development and thus economic research. While many of these challenges might not be new per se, they can be severely exacerbated by, for example, climate change. Economic implications of long-term increases in vector-borne diseases and heat stress as well as pandemics like the COVID-19 and ozone formation still remain to be analyzed in depth, as do the costs and benefits of adaptation measures dedicated to mitigating these effects (Mendelsohn 2012 ). Climate change also affects human health indirectly through impacts on economic development, land use, and biodiversity - and vice versa. Failed emission reductions and bad environmental management especially impact developing countries negatively through direct effects on health but also through health effects of delayed poverty reduction (Fankhauser and Stern 2020 ). Exposure to diseases or epidemics can increase the risk of civil conflicts and violence (Cervellati et al. 2016 , 2018 ). While research has addressed effects of life-expectancy, diseases and premature mortality on long-run economic development (e.g. Ebenstein et al. 2015 ; Acemoglu and Johnson 2007 ), a thorough analysis of the climate-health-development nexus is still missing. Overall, most research carried out on the interaction between environment, climate and human health has focused on physical health and mortality. The effects of air pollution from the burning of fossil fuels or agriculture on premature deaths, cardiac conditions and respiratory diseases, for example, received not only renewed interest in the wake of recent scandals (see e.g. Alexander and Schwandt 2019 ) but have been an active field of research for a number of years (Schlenker and Walker 2016 ; Tschofen et al. 2019 ). Mental health implications like stress, anxiety or depression on the other hand have received much less attention although, for example, Chen et al. ( 2018 ) in a study on air pollution in China estimate these effects to be on a similar scale to costs arising from impacts on physical health. Also, Danzer and Danzer ( 2016 ) find substantial effects of a large energy-related disaster (the Chernobyl catastrophe) on subjective well-being and mental health. Economic research should take up the challenge and put more effort into the economic evaluation of mental health related effects of climate change and environmental degradation in general. Potential to analyze these and other health-related questions have risen substantially in the last years, method-wise as well as topical, with new large data sets becoming available. Big data from insurance companies, satellite imagery on pollution dispersion and effects of draughts, for example, can provide new insights into the dynamics between environmental changes and health. But digital technologies themselves also generate new research questions addressing, for example, risks, costs and benefits of these new technologies.
• Carbon exposure and green finance The impact of climate change and of climate policy on the financial system is a topic of increasing public concern. The transition to a low-carbon economy poses a lot of challenges not only from physical risks and damages but also from transition risks. These accrue in such different areas as climate-related policy making, altered market behavior, changes in international trade patterns, technology development, and consumer behavior. To support a safe and gradual transition to a low-carbon economy, the financial sector needs to evaluate and eventually address the new risks associated with climate change and decarbonization in an efficient manner. There is widespread concern that financial markets currently lack sufficient information about the carbon exposure of assets, resulting in risks from climate change and climate policy for investments (Karydas and Xepapadeas 2018 ). If not anticipated by the markets, climate shocks also cause asset stranding, i.e. unanticipated and premature capital write-offs, downward revaluations, and conversion of assets to liabilities (Rozenberg et al. 2020 ; Bretschger and Soretz 2018 ). The same holds true for climate policies which are not or cannot be correctly anticipated by investors (Dietz et al. 2016 ; Stolbova et al. 2018 ; Sen and von Schickfus 2020 ). The growing awareness of these risks is reflected in the attention that policy makers have devoted to the development of transparency improving information systems and indicators in recent years. However, challenges related the design of these systems and indicators, e.g. with respect to an accurate and encompassing risk assessment, still remain. The importance of addressing these challenges is excerbated by prevalent network effects and counterparty risks that transmit climate-induced financial shocks from individual firms to the broad public holding their capital in stocks of fossil-fuel-related firms, investment funds, and pension funds, which all could suffer from stranded assets (Battiston et al. 2017 ). Divestment campaigns, shareholder engagement, and mandatory disclosure of climate-relevant financial information by companies and investors warrant further theoretical and empirical analysis. Also, a better understanding of the economics behind financing instruments like green bonds is only recently emerging (Agliardi and Agliardi 2019 ). Despite some early studies there is a knowledge gap with respect to the extent of climate and policy risks for central banks and regarding the potential significance of different channels connecting the risks in the real economy with monetary policy. Given the environmental and international policy perspective of the climate problem, the specific contribution of the financial sector and the central banks in the architecture of global climate policy has to be subject to further investigation.
• Energy system transformation The transition from a fossil-based to a green economy is needed to combat climate change but requires a thorough transformation of energy systems (Pommeret and Schubert 2019 ) in developed as well as in developing countries. In industrialized countries, challenges arise from the structural transformation of highly complex energy systems and their linkage with other economic sectors. While one hundred years ago, it was the rapid dissemination of fossil-based industrial processes, transportation, and heating that resulted in wide-spread sectoral change, similar adjustments can be expected with the increasing importance of electricity for decarbonization. However, changing the use of energy technologies in practice involves decisions on different levels and constitutes a highly nonlinear process. Future power generation in many countries will increasingly rely on renewable energies like wind and solar energy. To offset intermittent power generation, more and better storage capacities of batteries or pumped hydropower will be needed (Ambec and Crampes 2019 ). Synthetic fuels, heat pumps, fuel cells and e-mobility will increasingly use electricity to replace fossil fuels not only in the power sector but also in traffic and heat generation. While the adoption of renewable technologies like wind and solar was often much faster than predicted in the past, the critical mass of market penetration has still to be reached in other areas to benefit from potential scale effects and cost decreases. Shape and speed of the energy transition are, however, highly dependent on a political process which is hard to predict for market participants. Policy and ecological risks, together with the long-run character of the energy and related infrastructure investments, pose a big challenge for research and practice. In this context, it is especially the economic potential of green hydrogen and/or synthetic fuels that is controversially discussed at present. As production costs are expected to fall (Glenk and Reichelstein 2019 ), interest in hydrogen is increasing sharply (IEA 2019 ) and new research questions arise. For developing countries, clean and decentralized renewable energy technologies offer big potentials for electrification and economic development. However, despite the potential for decarbonization and the reduction of other externalities and health hazards and despite the fact that more than 90% of the annual increase in power generation comes from emerging economies, research on the development and adoption of clean energy technologies still focuses mainly on the developed world. More research on the barriers and challenges for adoption in developing countries is needed, including sustainable financing, institutional framing and the design of regionally tailored policies.
• Sustainability perspective on digitalization Digitalization and artificial intelligence are often seen as opportunities for enhancing the efficiency of energy and resource use. They offer new opportunities for circular economy, agriculture, monitoring of ecosystems and biodiversity, sustainable finance and decarbonization (see WBGU 2019 and literature within). However, they may also accelerate energy and resource use, increase inequality between regions and income groups and endanger sustainable development. Digitalization offers new access to markets, impacts market forms and shapes consumer behavior all of which can have extensive implications for the ecological, social and economic dimensions of sustainable development. Digitalization is a cross-cutting theme that reaches across spatial scales (from regional development to globalization) as well as temporal scales (from short-run impacts on energy systems to long-run adaptation to climate change). So far, the potentials and challenges for sustainable development that are associated with digital technologies have mostly been addressed outside of environmental and resource economics. The focus has been on topics such as data security and privacy or, for example, on the implications of the ”fourth industrial revolution” on employment and labor markets. Costs and benefits of digitization, the design and effectiveness of policies in industrialized as well as developing countries have garnered much less attention in the context of environmental, resource, energy and climate economics. Also, impacts of digitization on the behavior of economic agents resulting in, for example, rebound effects or changes in consumption patterns and environmental awareness, have not been addressed comprehensively (Gossar 2015 ). In all of these areas, our limited knowledge base creates opportunities and challenges for future research in the field. But, digitalization not only creates new research questions, it also provides new means to answer them. It has led to new developments in data science, big data analysis, machine learning and artificial intelligence that allow new insights into, for example, material flows, emission patterns and technology diffusion as well as the optimal design, implementation and effectiveness of regulation (Fowlie et al. 2019 ; Weersink et al. 2018 ; Graziano and Gillingham 2015 ).
• Quantitative analysis of environmental use Recently, there has been a significant shift in the empirical methods used in economics from traditional regression analysis to random assignment and quasi-experiments. Arguably this can improve the capturing of causal relationships and reduce the biases of traditional study designs. In environmental economics, experimental and quasi-experimental approaches have been applied mainly for capturing individuals’ or firms’ decisions on the use of land, water, resources, and energy (e.g. Allcott 2011 ; Duflo et al. 2013 ; Deschenes et al. 2017 ). Wider applications of these rigorous methods in environmental economics and well-suited empirical designs are desirable but certainly challenging e.g. when assessing aggregate environmental costs from climate change or biodiversity loss. An important but underrated field in applied environmental economics is the ex-post empirical assessment of environmental policies. The challenge is not only to identify environmental externalities, causalities, and impact intensities but also to provide an accurate valuation of the cost of policies, because they vary widely especially in environmental economics. The traditional empirical methods remain to be important and are not simply replaced. The same holds true for empirical designs in a time, cross-country, or panel structure. The increasing availability of large or very large datasets with observations varying widely across time and space offers a different set of options to provide evidence on the impact of environmental damages or policies to abate them (e.g. Currie and Walker 2011 ; Martin et al. 2014 ; Zhang et al. 2018 ). Fast-growing computational power and machine learning provide even more avenues for fruitful applications in environmental economics (see e.g. Abrell et al. 2019 ) but the challenge to use computer power wisely and to derive results which are sufficiently robust remains demanding .
• Structural assessment modelling and modelling transparency In an effort to better understand the ramifications of political decisions and technological developments on climate change, energy supply and resource extraction (to name but a few examples), increasingly sophisticated numerical models have been developed in recent decades. It is evident that quantitative economics analysis is important for policy advice. Yet despite their complexity, these models usually still adopt some very simplifying and sometimes ad-hoc assumptions. In particular assumptions used in integrated valuation models have come under heavy criticism in recent years (Stern 2013 ; Pindyck 2013 ). Simplifications concern market structures and market failures, the integration of risk and uncertainty as well as societal, institutional and cultural detail. Also, manifestations of climate change and damages come at very different regional and temporal scales, making a truly integrated assessment of the climate-ecosystem-economy nexus next to impossible. We see it as a major challenge for future research to provide more accurate foundations for integrated assessment models. While simplifications are needed to reduce computational complexity, they raise the question to which extent the results obtained render reliable insights into future developments. Asking for models that are detailed in every dimension and can answer every question resembles of course the search for the holy grail. However, the need for a better understanding of the model dynamics has already led to the development of a new generation of models which have a stronger foundation in theory (Golosov et al. 2014 , Bretschger and Karydas 2019 ). A better understanding of the limits of models and of the questions specific models can and cannot address is still needed as well as transparency in model development. More applied studies, assessments of global environmental trends under different economic assumptions often use ”scenarios” to describe future trajectories. The scenarios are mostly based on expert opinion and do not rely on estimates about the likelihood that such a trajectory will occur. It is also critical that the economics behind the scenarios is often neglected. Prominently, per capita income can be projected using endogenous growth theory, while population development can be evaluated using state-of-the-art theories on fertility and morbidity.

Conclusions

This article set out to highlight a number of challenges that are highly relevant for future research in the field of environmental and resource economics. The focus was mainly, although not exclusively, on topical issues. We only briefly touched upon on some methodological advancements that might have the power to further parts of our field. Big data, machine learning and artificial intelligence hold high promise in this regard but their limits and potentials for environment, climate and resource economics have yet to be fully understood.

It should have become clear, that a number of the challenges presented can only be addressed adequately by interdisciplinary research teams with relevant disciplines ranging from climate science, (computer) engineering, sociology, virology to soil sciences. In many cases, economists’ analysis and the derivation of sound policy recommendations require the knowledge available in these fields. However, such research cooperations are by no means one-way streets: Other disciplines need the input of economists in order to assess future development scenarios and implementability of solutions. The knowledge and data required for economics analysis does not always exist yet, but interdisciplinary cooperation can help to identify and close these gaps. Overall, the less economists have already worked on specific challenges, the harder it is to assess best research strategies and the potential for success. Take the digitization-sustainable-development-nexus as an example: best research strategies and success are extremely difficult to predict as not only is the related economics research still in its infancy but also the field itself is extremely dynamic.

As already pointed out in the beginning: We are aware that our selection is bound to create discontent and disagreement. Having said this, it should also be stated that we expect some of our challenges to be more or less universally agreed upon. This holds especially for the broader topics: for example, how to accomplish deep decarbonization; how to deal with risk and uncertainty; or how to assess the role of disruptive development. One reason for this lies in the encompassing nature of these topics. They are relevant for many of the other fields that we have pointed out: For behavioral analyses, the capacity to deal with disruptive change in the face of risk and uncertainty are essential. Loss of biodiversity and natural capital, land degradation, conflicts over resources and migration are exacerbated by climate change. The potential of digitization for sustainable development constitutes disruptive change in itself. Yet, all of these fields are not merely subfields of the more overarching themes, they raise important research questions in their own right.

Nevertheless, it is to be expected that it will be the more specific fields over which disagreement will arise: Are ‘land use and soil degradation’ more important than ‘fisheries’? Is the ‘institutional analysis of environmental policies’ of higher relevance than the ‘development of alternative welfare concepts’ (to pick out some random examples). Of course, there are more fields that could have been included and also, of course, there is no objective criterion for the inclusion or exclusion of fields. The selection of the challenges is based on the analysis and criteria presented in the first section but it is ultimately ours; we are happy if this paper contributes to a lively and constructive discussion about the future of our field.

Open access funding provided by Swiss Federal Institute of Technology Zurich.

Publisher's Note

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

Contributor Information

Lucas Bretschger, Email: hc.zhte@reghcsterbl .

Karen Pittel, Email: ed.ofi@lettip .

• Abrell J, Kosch M, Rausch S (2019) How effective was the UK carbon tax? A machine learning approach to policy evaluation, CER-ETH working paper 19/317, ETH Zurich
• Acemoglu D, Johnson S. Disease and development: the effect of life expectancy on economic growth. J Polit Econ. 2007; 115 (6):925–985. [ Google Scholar ]
• Acemoglu D, Robinson JA. Economic backwardness in political perspective. Am Polit Sci Rev. 2006; 100 (1):115–131. [ Google Scholar ]
• Acemoglu D, Philippe Aghion P, Bursztyn L, Hemous D. The environment and directed technical change. Am Econ Rev. 2012; 102 (1):131–166. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Adger NM, Pulhin JM, Barnett J, Dabelko GD, Hovelsrud GK, Levy M, White LL (Eds) (2014) Climate change 2014: impacts, adaptation, and vulnerability. Part a: global and sectoral aspects. Contribution of working group II to the fifth assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge
• African Development Bank (2015) African Ecological Futures Report 2015, https://www.afdb.org
• Aghion P, Dechezleprêtre A, Hémous D, Martin R, Van Reenen J. Carbon taxes, path dependency, and directed technical change: evidence from the auto industry. J Polit Econ. 2016; 124 (1):1–51. [ Google Scholar ]
• Agliardi E, Xepapadeas A (2018) Optimal scheduling of greenhouse gas emissions under carbon budgeting and policy design, Athens University of Economics and Business, DEOS Working Paper No. 1808
• Agliardi E, Agliardi R (2019) Financing environmentally-sustainable projects with green bonds, Environment and Development Economics 24, Special Issue 6 (The Economics of Climate Change and Sustainability (Part A)), pp 608–662
• Alexander D, Schwandt H (2019) The impact of car pollution on infant and child health: evidence from emissions cheating, WP 2019-04, The Federal Reserve Bank of Chicago
• Allcott H. Social norms and energy conservation. J Public Econ. 2011; 95 (9–10):1082–1095. [ Google Scholar ]
• Ambec S, Crampes C. Decarbonizing electricity generation with intermittent sources of energy. J Assoc Environ Resour Econ. 2019; 6 (6):1105–1134. [ Google Scholar ]
• Antoci A, Borghesi S, Russu P. Don’t feed the bears! environmental defense expenditures and species-typical behaviour in an optimal growth model. Macroecon Dyn. 2019 doi: 10.1017/S1365100519000397. [ CrossRef ] [ Google Scholar ]
• Arthur WB. Competing technologies, increasing returns, and lock-in by historical events. Econ J. 1989; 99 (394):116–131. [ Google Scholar ]
• Athanassoglou S, Xepapadeas A. Pollution control with uncertain stock dynamics: When, and how, to be precautious. J Environ Econ Manage. 2012; 63 :304–320. [ Google Scholar ]
• Auffhammer M. The state of environmental and resource economics: a google scholar perspective. Rev Environ Econ Policy. 2009; 3 (2):251–269. [ Google Scholar ]
• Awumbila M. Drivers of migration and urbanization in Africa: key trends and issues. background paper prepared for UN expert group meeting on sustainable cities. New York: Human Mobility and International Migration; 2017. [ Google Scholar ]
• Badeeb RA, Lean HH, Clark J. The evolution of the natural resource curse thesis: a critical literature survey. Resour Policy. 2017; 51 :123–134. [ Google Scholar ]
• Barata M, Ligeti E, De Simone G, Dickinson T, Jack D, Penney J, Rahman M, Zimmerman R (2011) In: Climate change and human health in cities. Climate change and cities: first assessment report of the urban climate Change Research Network. Rosenzweig C, Solecki WD, Hammer SA Mehrotra S (eds) Cambridge University Press, Cambridge
• Barnes W, Gartland M, Stack M. Old habits die hard: path dependency and behavioral lock-in. J Econ Issues. 2004; 38 (2):371–377. [ Google Scholar ]
• Battiston S, Mandel A, Monasterolo I, Schutze F, Visentin G. A climate stress-test of the financial system. Nat Clim Change. 2017; 7 (4):283–288. [ Google Scholar ]
• Baumgärtner S, Engler J-O (2018) 2018. An axiomatic foundation of entropic preferences under Knightian uncertainty, Paper presented at the SURED conference in Ascona
• Beck M, Rivers N, Yonezawa H. A rural myth? Sources and implications of the perceived unfairness of carbon taxes in rural communities. Ecol Econ. 2016; 124 :124–134. [ Google Scholar ]
• Borissov K, Brausmann A, Bretschger L. Carbon pricing, technology transition, and skill-based development. Eur Econ Rev. 2019; 118 :252–269. [ Google Scholar ]
• Bratman GN, Anderson CB, Berman MG, Cochran B, de Vries S, Flanders J, Daily GC. Nature and mental health: an ecosystem service perspective. Sci Adv. 2019 doi: 10.1126/sciadv.aax0903. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Brausmann A, Bretschger L. Economic development on a finite planet with stochastic soil degradation. Eur Econ Rev. 2018; 108 :1–19. [ Google Scholar ]
• Brei M, Pérez-Barahona A, Strobl E. Protecting species through legislation: the case of sea turtles. Am J Agric Econ. 2020; 102 (1):300–328. [ Google Scholar ]
• Bretschger L. Population growth and natural resource scarcity: long-run development under seemingly unfavourable conditions. Scand J Econ. 2013; 115 (3):722–755. [ Google Scholar ]
• Bretschger L. Equity and the convergence of nationally determined climate policies. Environ Econ Policy Stud. 2017; 19 (1):1–14. [ Google Scholar ]
• Bretschger L. Climate policy and economic growth. Resour Energy Econ. 2017; 49 :1–15. [ Google Scholar ]
• Bretschger L. Malthus in the light of climate change. Eur Econ Rev. 2020; 127 :103477. doi: 10.1016/j.euroecorev.2020.103477. [ CrossRef ] [ Google Scholar ]
• Bretschger L, Schaefer A. Dirty history versus clean expectations: Can energy policies provide momentum for growth? Eur Econ Rev. 2017; 99 :170–190. [ Google Scholar ]
• Bretschger L, Vinogradova A (2018) Escaping Damocles’ Sword: endogenous climate shocks in a growing economy, economics working paper series 18/291, ETH Zurich
• Bretschger L, Soretz S (2018) Stranded assets: how policy uncertainty affects capital, growth, and the environment, economics working paper series 18/288, ETH Zurich
• Bretschger L, Karydas C. Economics of climate change: Introducing the basic climate economic (BCE) model. Environ Develop Econ. 2019; 24 (6):560–582. [ Google Scholar ]
• Brock W, Xepapadeas A. Valuing biodiversity from an economic perspective: a unified economic. Ecol Genet Approach Am Econ Rev. 2003; 93 :597–1614. [ Google Scholar ]
• Brock W, Xepapadeas A (1903) (2019) Regional climate policy under deep uncertainty: robust control. Athens University of Economics and Business, Discussion Paper No, Hot Spots and Learning
• Bromley D. Economic interests and institutions: the conceptual foundations of public policy. Oxford: Blackwell; 1989. [ Google Scholar ]
• Cai Y, Lenton TM, Lontzek TS. Risk of multiple interacting tipping points should encourage rapid CO2 emission reduction. Nat Clim Change. 2016; 6 :520–525. [ Google Scholar ]
• Calel R, Dechezlepretre A. Environmental policy and directed technological change: evidence from the European carbon market. Rev Econ Stat. 2016; 98 (1):173–191. [ Google Scholar ]
• Carleton TA, Hsiang SM. Social and economic impacts of climate. Science. 2016 doi: 10.1126/science.aad9837. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Casari M, Luini L. Cooperation under alternative punishment institutions: an experiment. J Econ Behav Organ. 2009; 71 (2):273–282. [ Google Scholar ]
• Casey G, Shayegh S, Moreno-Cruz J, Bunzl M, Galor O, Caldeira K. The impact of climate change on fertility. Environ Res Lett. 2019 doi: 10.1088/1748-9326/ab0843. [ CrossRef ] [ Google Scholar ]
• Cattaneo C, Beine M, Fröhlich CJ, Kniveton D, Martinez-Zarzoso I, Mastrorillo M, Millock K, Piguet E, Schraven B. Human migration in the era of climate change. Rev Environ Econ Policy. 2019; 13 (2):189–206. [ Google Scholar ]
• Cerda Planas L. Moving toward greener societies: moral motivation and green behaviour. Environ Resource Econ. 2018; 70 :835–860. [ Google Scholar ]
• Cervellati M, Esposito E, Sunde U, Valmori S (2018) Malaria and violence. Econ Policy, pp 403–446
• Cervellati M, Sunde U, Valmori S. Pathogens, weather shocks and civil conflicts. Econ J. 2016; 127 :2581–616. [ Google Scholar ]
• Chan K, Anderson E, Chapman M, Jespersen K, Olmsted P. Payments for ecosystem services: rife with problems and potential - for transformation towards sustainability. Ecol Econ. 2017; 140 :110–122. [ Google Scholar ]
• Chen Z, Oliva P, Zhang P (2018) Pollution and mental health: evidence from China, NBER Working Paper Series 24686
• Conforti A, Mascia M, Cioffi G, De Angelis C, Coppola G, De Rosa P, Pivonello R, Alviggi C, De Placido G. Air pollution and female fertility: a systematic review of literature. Reproduct Biol Endocrinol. 2018; 16 :117. doi: 10.1186/s12958-018-0433-z. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Costanza R, Howarth R, Kubiszewski I, Liu S, Ma C, Plumecocq G, Stern D. Influential publications in ecological economics revisited. Ecol Econ. 2016; 123 :68–76. [ Google Scholar ]
• Currie J, Walker R. Traffic congestion and infant health: evidence from E-ZPass. Am Econ J Appl Econ. 2011; 3 (1):65–90. [ Google Scholar ]
• Danzer AM, Danzer N. The long-run consequences of chernobyl: evidence on subjective well-being, mental health and welfare. J Public Econ. 2016; 135 (2016):47–60. [ Google Scholar ]
• Dasgupta P, Heal G. The optimal depletion of exhaustible resources. Rev Econ Stud. 1974; 41 (5):3–28. [ Google Scholar ]
• Dasgupta S, De Cian E (2016) Institutions and the environment: existing evidence and future directions, FEEM Working Paper No. 41.2016
• Deschenes O, Greenstone M, Shapiro JS. Defensive investments and the demand for air quality: evidence from the NOx budget program. Am Econ Rev. 2017; 107 (10):2958–2989. [ Google Scholar ]
• Dietz S, Bower A, Dixon C, Gradwell P. Climate value at risk of global financial assets. Nat Clim Change Lett. 2016; 6 :676–679. [ Google Scholar ]
• Drupp MA. Limits to substitution between ecosystem services and manufactured goods and implications for social discounting. Environ Resour Econ. 2018; 69 :135–158. [ Google Scholar ]
• Duflo E. Richard T. Ely lecture: the economist as plumber. Am Econ Rev Papers Proc. 2017; 107 (5):126. [ Google Scholar ]
• Duflo E, Greenstone M, Pande R, Ryan N. Truth-telling by third-party auditors and the response of polluting firms: Experimental evidence from India. Q J Econ. 2013; 128 (4):1499–1545. [ Google Scholar ]
• Ebenstein A, Fan M, Greenstone M, He G, Yin P, Zhou M. Growth, pollution, and life expectancy: China from 1991 to 2012. Am Econ Rev Paper Proc. 2015; 105 (5):226–231. [ Google Scholar ]
• Ehrlich P. Key issues for attention from ecological economists. Environ Dev Econ. 2008; 13 (1):1–20. [ Google Scholar ]
• Engel S, Pagiola S, Wunder S. Designing payments for environmental services in theory and practice: an overview of the issues. Ecol Econ. 2008; 65 :663–674. [ Google Scholar ]
• Fankhauser S. Adaptation to climate change. Ann Rev Resour Econ. 2017; 9 (1):209–230. [ Google Scholar ]
• Fankhauser S, Stern N, et al. Climate change, development, poverty and economics. In: Basu K, et al., editors. The state of economics, the state of the world. Cambridge: MIT Press; 2020. [ Google Scholar ]
• Fezzi C, Bateman I. The impact of climate change on agriculture: nonlinear effects and aggregation bias in Ricardian models of farmland values. J Assoc Environ Resour Econ. 2015; 2 (1):57–92. [ Google Scholar ]
• Fowlie M, Rubin E, Walker R. Bringing satellite-based air quality estimates down to earth. AEA Paper Proc. 2019; 109 :283–288. [ Google Scholar ]
• Fox J, Klüsener S, Myrskyla M. Is a positive relationship between fertility and economic development emerging at the sub-national regional level? Theoretical considerations and evidence from Europe. Eur J Populat. 2019; 35 :487–518. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Fuss S, Lamb WF, Callaghan MW, Hilaire J, Creutzig F, Amann T, Minx JC. Negative emissions-Part 2: costs, potentials and side effects. Environ Res Lett. 2019; 13 :063002. doi: 10.1088/1748-9326/aabf9f. [ CrossRef ] [ Google Scholar ]
• Galor O, Özak Ö. The agricultural origins of time preference. Am Econ Rev. 2016; 106 (10):3064–3103. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Gerlagh R, Michielsen TO. Moving targets-cost-effective climate policy under scientific uncertainty. Clim Change. 2015; 132 :519–529. [ Google Scholar ]
• Glenk G, Reichelstein S. Economics of converting renewable power to hydrogen. Nature Energy. 2019; 4 :216–222. [ Google Scholar ]
• Goldthau A, Westphal K, Bazilian M, Bradshaw M. How the energy transition will reshape geopolitics. Nature. 2019; 569 :29–31. doi: 10.1038/d41586-019-01312-5. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Golosov M, Hassler J, Krusell P, Tsyvinski A. Optimal taxes on fossil fuel in general equilibrium. Econometrica. 2014; 82 (1):41–88. [ Google Scholar ]
• Gossar C (2015) Rebound Effects and ICT: a review of the literature. In: Hilty LM, Aebischer B (eds) ICT Innovations for sustainability, pp 435–448
• Graziano M, Gillingham K. Spatial patterns of solar photovoltaic system adoption: the influence of neighbors and the built environment. J Econ Geogr. 2015; 15 (4):815–839. [ Google Scholar ]
• Hainsch K, Burandt T, Kemfert C, Löffler K, Oei PY, von Hirschhausen C (2018) Emission pathways towards a low-carbon energy system for Europe: a model-based analysis of decarbonization scenarios, DIW Discussion Paper 1745
• Hansen LP, Sargent TJ. Robust control and model uncertainty. Am Econ Rev. 2001; 91 (2):60–66. [ Google Scholar ]
• Harstad B. Buy Coal! A case for supply-side, environmental policy. J Polit Econ. 2012; 120 (1):77–115. [ Google Scholar ]
• Hartwick JM. Intergenerational equity and the investment of rents from exhaustible resources. Am Econ Rev. 1977; 67 :972–74. [ Google Scholar ]
• Henderson JV, Storeygard A, Deichmann U (2014) 50 years of urbanization in Africa, Examining the Role of Climate Change. World Bank Development Research Group Policy Research Working Paper no. 6925. Washington DC: World Bank Group
• Henderson JV, Storeygard A, Deichmann U. Has climate change driven urbanization in Africa? J Dev Econ. 2017; 124 :60–82. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Hotelling H. The economics of exhaustible resources. J Polit Econ. 1931; 39 (2):137–175. [ Google Scholar ]
• IEA (2019) The Future of Hydrogen, International Energy Agency https://www.iea.org/hydrogen2019
• Ionesco D, Mokhnacheva D, Gemenne F. The atlas of environmental migration. London: International Organization for Migration; 2017. [ Google Scholar ]
• IPCC (2002) Climate Change and Biodiversity, IPCC Technocal Paper V, https://www.ipcc.ch/site/assets/uploads/2018/03/climate-changes-biodiversity-en.pdf
• IPCC (2018) Summary for Policymakers. In: Global warming of 1 . 5 ∘ C. An IPCC Special Report on the impacts of global warming of 1 . 5 ∘ C above pre-industrial levels and related global greenhouse gas emission pathways, World Meteorological Organization, Geneva, Switzerland
• IPCC (2019) Special report on climate change, desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in Terrestrial Ecosystems, Summary for Policymakers
• James SP. Cultural ecosystem services: a critical assessment. Ethics Policy Environ. 2015; 18 (3):338–350. [ Google Scholar ]
• Jensen S, Traeger CP. Optimal climate change mitigation under long-term growth uncertainty: Stochastic integrated assessment and analytic findings. Eur Econ Rev. 2014; 69 :104–125. [ Google Scholar ]
• Kalkuhl M, Brecha RJ. The carbon rent economics of climate policy. Energy Econ. 2013; 39 :89–99. [ Google Scholar ]
• Kalkuhl M, Edenhofer O, Lessmann K. Learning or lock-in: Optimal technology policies to support mitigation. Resour Energy Econ. 2012; 31 (1):1–23. [ Google Scholar ]
• Kan K, Lee M-J. The effects of education on fertility: evidence from Taiwan. Econ Inq. 2018; 56 (1):343–357. [ Google Scholar ]
• Karydas C, Xepapadeas A (2018) Pricing climate change risks: CAPM with rare disasters and stochastic probabilities. CER-ETH Working Paper Series Working Paper 19/311
• Klibanoff P, Marinacci M, Mukerji S. A smooth model of decision making under uncertainty. Econometrica. 2005; 73 (6):1849–1892. [ Google Scholar ]
• Kubea R, Löschel A, Mertense H, Requate T (2018) Research trends in environmental and resource economics: Insights from four decades of JEEM, paper presented at the WCERE 2018 in Gothenburg
• Lanz B, Dietz S, Swanson T. Global population growth, technology, and Malthusian constraints: a quantitative growth theoretic perspective. Int Econ Rev. 2017; 58 (3):973–1006. [ Google Scholar ]
• Lenton TM, Ciscar J-C. Integrating tipping points into climate impact assessments. Clim Change. 2013; 117 :585–597. [ Google Scholar ]
• Levin S, Xepapadeas T, Crépin A-S, Norberg J, de Zeeuw A, Folke C, Hughes T, Arrow K, Barrett S, Daily G, Ehrlich P, Kautsky N, Maeler K-G, Polasky S, Troell M, Vincent JR, Walker B. Social-ecological systems as complex adaptive systems: modeling and policy implications. Environ Dev Econ. 2013; 18 (2):111–132. [ Google Scholar ]
• Mach KJ, Kraan CM, Adger WN, Buhaug H, Burke M, Fearon JD, Field CB, Hendrix CS, Maystadt J-F, O’Loughlin J, Roessler P, Scheffran J, Schultz KA, von Uexkull N. Climate as a risk factor for armed conflict. Nature. 2019; 571 :193–197. [ PubMed ] [ Google Scholar ]
• Maniatis D, Scriven J, Jonckheere I, Laughlin J, Todd K. Toward REDD+ Implementation. Annu Rev Environ Resour. 2019; 44 :373–98. [ Google Scholar ]
• Manoussi V, Xepapadeas A, Emmerling J. Climate engineering under deep uncertainty. J Econ Dyn Control. 2018; 94 :207–224. [ Google Scholar ]
• Marin G, Vona F. Climate policies and skill-biased employment dynamics: evidence from EU countries. J Environ Econ Manage. 2019; 98 :1–18. [ Google Scholar ]
• Martin R, De Preux LB, Wagner UJ. The impact of a carbon tax on manufacturing: evidence from microdata. J Public Econ. 2014; 117 :1–14. [ Google Scholar ]
• Mendelsohn R. The economics of adaptation to climate change in developing countries. Clim Change Econ. 2012; 3 (2):1250006-1–1250006-21. [ Google Scholar ]
• Millock K. Migration and environment. Ann Rev Resour Econ. 2015; 7 (1):35–60. [ Google Scholar ]
• Minx JC, Lamb WF, Callaghan MW, Fuss S, Hilaire J, Creutzig F, del Mar Zamora Dominguez M. Negative emissions: Part 1: research landscape and synthesis. Environ Res Lett. 2018; 13 :063001. doi: 10.1088/1748-9326/aabf9b. [ CrossRef ] [ Google Scholar ]
• O’Sullivan M, Overland I, Sandalow D (2017) The geopolitics of renewable energy, working paper, Belfer Center for Science and International Affairs, Harvard Kennedy School
• OECD (2017) Closing the Regulatory Cycle: effective ex post evaluation for improved policy outcomes. In: 9th OECD conference on measuring regulatory performance, key findings and conference proceedings, http://www.oecd.org/gov/regulatory-policy/Proceedings-9th-Conference-MRP.pdf
• Parry I (2015) Carbon Tax Burdens on Low-Income Households: A Reason for Delaying Climate Policy?, In: Clements B, de Mooij R, Gupta S, Keen M (2015) Inequality and Fiscal Policy, Ch. 13, International Monetary Fund, Washington
• Pearce DW, Atkinson G, Dubourg WR. The economics of sustainable development. Annu Rev Energy Environ. 1994; 19 :457–474. [ Google Scholar ]
• Pellegrini L, Gerlagh R. Corruption, democracy, and environmental policy, an empirical contribution to the debate. J Environ Dev. 2008; 15 (3):332–354. [ Google Scholar ]
• Peretto P (2017) Through scarcity to prosperity: a theory of the transition to sustainable growth, Economic Research Initiatives at Duke (ERID) Working Paper No. 260
• Peretto P, Valente S. Growth on a finite planet: resources, technology and population in the long run. J Econ Growth. 2015; 20 (3):305–331. [ Google Scholar ]
• Pigou AC. The economics of welfare. London: Macmillan; 1920. [ Google Scholar ]
• Pindyck RS. Climate change policy: what do the models tell us? J Econ Literat. 2013; 51 :860–872. [ Google Scholar ]
• Pittel K, Rübbelke DTG. International climate finance and its influence on fairness and policy. World Econ. 2011; 36 (4):419–436. [ Google Scholar ]
• Polasky SCL, Kling SA, Levin SR, Carpenter GC, Daily PR, Ehrlich GM Heal, Lubchenco J. Role of economics in analyzing the environment and sustainable development. PNAS. 2019; 116 (12):5233–5238. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Polyakov M, Chalak M, Iftekhar S, Pandit R, Tapsuwan S, Zhang F, Ma C. Authorship, collaboration, topics, and research gaps in environmental and resource economics 1991–2015. Environ Resource Econ. 2018; 71 (1):217–239. [ Google Scholar ]
• Pommeret A, Schubert K (2019) Energy transition with variable and intermittent renewable electricity generation, CESifo Working Paper Series 7442, CESifo Group Munich
• Pommeret A, Schubert K. Intertemporal emission permits trading under uncertainty and irreversibility. Environ Resour Econ. 2018; 71 :73–97. [ Google Scholar ]
• Rockstrom J. A safe operating space for humanity. Nature. 2009; 461 :472–475. [ PubMed ] [ Google Scholar ]
• Rodrik D. Second-best institutions. Am Econ Rev Paper Proc. 2008; 98 (2):100–104. [ Google Scholar ]
• Rozenberg J, Vogt-Schilb A, Hallegatte S. Instrument choice and stranded assets in the transition to clean capital. J Environ Econ Manage. 2020; 100 :102277. [ Google Scholar ]
• Salzman J, Bennett G, Carroll N, Goldstein A, Jenkins M. The global status and trends of payments for ecosystem services. Nat Sustain. 2018; 1 :136–144. [ Google Scholar ]
• Schirpke U, Tappeiner U, Tasser E. A transnational perspective of global and regional ecosystem service flows from and to mountain regions. Nature. 2019; 9 :6678. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Schlenker W, Walker RW. Airports. Air pollution, and contemporaneous health. Rev Econ Stud. 2016; 83 :768–809. [ Google Scholar ]
• Sen S, von Schickfus MT. Climate policy, stranded assets, and investors’ expectations. J Environ Econ Manage. 2020; 100 :102277. [ Google Scholar ]
• Seto KC, Dhakai S, Bigio A, Delgado Arias S, Dewar D, Huang L, Ramaswami A (2014) Human settlements, infrastructure and spatial planning. In: Intergovernmental panel on climate change (eds.), Climate Change 2014: Mitigation of Climate Change. Cambridge University Press, Cambridge
• Small N, Munday M, Durance I. The challenge of valuing ecosystem services that have no material benefits. Glob Environ Change. 2017; 44 :57–67. [ Google Scholar ]
• Sovacool BK, Scarpaci J. Energy justice and the contested petroleum politics of stranded assets: Policy insights from the Yasun-ITT Initiative in Ecuador. Energy Policy. 2016; 95 :158–171. [ Google Scholar ]
• Stern N. The structure of economic modeling of the potential impacts of climate change: Grafting gross underestimation of risk onto already narrow science models. J Econ Literat. 2013; 51 (3):838–59. [ Google Scholar ]
• Stern N. Current climate models are grossly misleading. Nature. 2016; 530 :407–409. [ PubMed ] [ Google Scholar ]
• Sterner T. Fuel taxes and the Poor: the distributional consequences of gasoline taxation and their implications for climate policy, Routledge Journals. New York: Taylor & Francis; 2011. [ Google Scholar ]
• Stolbova V, Monasterolo I, Battiston S. A financial macro-network approach to climate policy evaluation. Ecol Econ C. 2018; 149 :239–253. [ Google Scholar ]
• Strassheim H, Beck S. Handbook of behavioural change and public policy. Edward Elgar: Handbooks of Research on Public Policy series; 2019. [ Google Scholar ]
• Sulemanaa I, Nketiah-Amponsaha E, Codjoea EA, Andoh JAN. Urbanization and income inequality in Sub-Saharan Africa. Sustain Cities Soc. 2019; 48 :101544. doi: 10.1016/j.scs.2019.101544. [ CrossRef ] [ Google Scholar ]
• TEEB (2020) The economics of ecosystems and biodiversity, https://www.teebweb.org
• Tschofen P, Azevedo IL, Muller NZ. Fine particulate matter damages and value added in the US economy. PNAS. 2019; 116 (40):19857–19862. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• UN - United Nations (1992) Convention on biological diversity https://www.cbd.int/doc/legal/cbd-en.pdf
• UN - United Nations (2015) Millenium Development Goals and Beyond 2015, Target 7, https://www.un.org/millenniumgoals/environ.shtml
• UN - United Nations (2016) Urbanization and development: emerging futures, world cities report, http://wcr.unhabitat.org/wp-content/uploads/2017/02/WCR-2016-Full-Report.pdf
• UN - United Nations (2018) 68% of the world population projected to live in urban areas by 2050, says UN, https://www.un.org/development/desa/en/news/population/2018-revision-of-world-urbanization-prospects.html
• UN - United Nations (2019) World Population Prospects 2019: Highlights. UN Department of Economic and Social Affairs, Population Division, New York
• UNEP - United Nations Environmet Programme (2019), Emissions Gap Report 2019, Nairobi
• Van der Ploeg F, de Zeeuw A. Climate tipping and economic growth: precautionary capital and the price of carbon. J Eur Econ Assoc. 2018; 16 (5):1577–1617. [ Google Scholar ]
• Vatn A. An institutional analysis of payments for environmental services. Ecol Econ. 2010; 69 (6):1245–1252. [ Google Scholar ]
• Verburg PH, Dearing JA, Dyke JG, van der Leeuw S, Seitzinger S, Steffen W, Syvitski J. Methods and approaches to modelling the Anthropocene. Glob Environ Change. 2016; 39 :328–340. [ Google Scholar ]
• von Stechow C, Minx JC, Riahi K, Jewell J, McCollum DL, Callaghan MW, Bertram C, Luderer G, Baiocchi G. 2 ∘ C and SDGs: united they stand, divided they fall? Environ Res Lett. 2016; 11 :034022. doi: 10.1088/1748-9326/11/3/034022. [ CrossRef ] [ Google Scholar ]
• WBGU - German Advisory Council on Global Change (2016) Humanity on the Move: Unlocking the transformative power of cities. WBGU, WBGU Flagship Report, Berlin
• WBGU - German Advisory Council on Global Change (2018a) Just & In-Time Climate Policy. Four Initiatives for a Fair Transformation. Policy Paper 9. Berlin: WBGU
• WBGU - German Advisory Council on Global Change (2018b) Towards our common digital future. WBGU, WBGU Flagship Report, Berlin
• Weersink A, Fraser E, Pannell D, Duncan E, Rotz S. Opportunities and challenges for big data in agricultural and environmental analysis. Ann Rev Resour Econ. 2018; 10 :19–37. [ Google Scholar ]
• Weitzman M. The Noah’s Ark approach. Econometrica. 1998; 66 (6):1279–1298. [ Google Scholar ]
• Weitzman M. Book review on nature in the balance: the economics of biodiversity, edited by Dieter Helm and Cameron Hepburn. J Econ Literat. 2014; 52 (4):1193–1194. [ Google Scholar ]
• Wunder S, Brouwer R, Engel S, Ezzine-de-Blas D, Muradian R, Pascual U, Pinto R. From principles to practice in paying for nature’s services. Nat Sustain. 2018; 1 :145–150. [ Google Scholar ]
• Zhang P, Deschenes O, Meng K, Zhang J. Temperature effects on productivity and factor reallocation: Evidence from a half million Chinese manufacturing plants. J Environ Econ Manage. 2018; 88 :1–17. [ Google Scholar ]