Download WORD - UNCTAD Virtual Institute

UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT
Virtual Institute Teaching Material
on Trade, the Environment and Sustainable Development:
Transition to a Low-carbon Economy
New York and Geneva, 2016
Note
The views expressed in this volume are those of the author and do not necessarily reflect
those of the United Nations.
The designations employed and the presentation of the material do not imply the expression
of any opinion on the part of the United Nations concerning the legal status of any country,
territory, city or area, or of its authorities, or concerning the delimitation of its frontiers or
boundaries.
Material in this document may be freely quoted or reprinted, but acknowledgement of the
UNCTAD Virtual Institute is requested, together with a reference to the document number.
A copy of the publication containing the quotation or reprint should be sent to the UNCTAD
Virtual Institute, Division on Globalization and Development Strategies, Palais des Nations,
1211 Geneva 10, Switzerland.
This document has been edited externally.
The UNCTAD Virtual Institute (http://vi.unctad.org) is a capacity-building and networking
programme that aims to strengthen teaching and research of international trade and
development issues at academic institutions in developing countries and countries with
economies in transition, and to foster links between research and policymaking. For further
information, please contact [email protected].
UNCTAD/GDS/2016/2
Copyright © United Nations, 2016
All rights reserved
ii
Acknowledgements
This teaching material was developed by the UNCTAD Virtual Institute, under the overall
guidance of Vlasta Macku, and in cooperation with UNCTAD’s Trade, Environment,
Climate Change and Sustainable Development Branch. The material was researched and
written by Caspar Sauter from the Virtual Institute, and benefitted from comments by
Bonapas Onguglo and Robert Hamwey of the Trade, Environment, Climate Change and
Sustainable Development Branch. The text was English-edited by David Einhorn, and the
design and layout were created by Hadrien Gliozzo.
The financial contribution of the Government of Finland is gratefully acknowledged.
iii
Table of contents
Note ........................................................................................................................................ ii
Acknowledgements ............................................................................................................... iii
Table of contents ................................................................................................................... iv
List of figures ........................................................................................................................ vi
List of tables ......................................................................................................................... vii
List of boxes ......................................................................................................................... vii
List of abbreviations ............................................................................................................. ix
Introduction ...........................................................................................................................1
Module 1 The environment, the economy and trade: The importance of sustainable
development ...........................................................................................................................5
1 Introduction ..........................................................................................................................5
2 The economy and the environment: Unravelling the links ..................................................6
2.1 Links between the economy and the environment ....................................................6
2.2 Environmental impact of economic activities.........................................................12
3 Sustainability of the economic system ...............................................................................19
3.1 Ecocide: Lessons from history ................................................................................19
3.2 Sustainability...........................................................................................................21
3.3 Sustainable development: The international awakening.........................................22
3.4 Trade as a key component of sustainable development ..........................................23
4 Impact of trade on the environment ...................................................................................25
4.1 Trade, trade openness and the environment: A first glance at the data ..................26
4.2 Environmental impact of trade: What we can learn from theory............................29
4.3 Environmental impact of trade: What we can learn from empirical evidence .......33
5 Exercises and questions for discussion ..............................................................................36
Annex 1: Some useful databases...........................................................................................38
Annex 2: Selected additional reading material ....................................................................38
References .............................................................................................................................39
Module 2 The climate science behind climate change .....................................................43
1 Introduction ........................................................................................................................43
2 The theoretical basis of the climate system and climate change .......................................45
2.1 The five components of the climate system ............................................................46
2.2 The earth’s energy balance and the natural greenhouse effect ...............................49
2.3 Internally and externally induced climate change ..................................................53
2.4 Measuring the importance of factors driving climate change: radiative forcing and
effective radiative forcing .............................................................................................56
2.5 Human-induced climate change ..............................................................................59
3 Observed changes in the climate system ...........................................................................67
3.1 Observed changes in temperature ...........................................................................67
3.2 Observed changes in precipitation ..........................................................................69
3.3 Observed changes in ice and snow cover ...............................................................70
3.4 Observed changes in sea levels ...............................................................................70
3.5 Observed changes in extreme events ......................................................................71
3.6 Impacts on natural and human systems ..................................................................72
4 Anticipated changes in the climate system and potential impacts of climate change .......74
iv
5 Exercises and discussion questions ...................................................................................78
Annex 1 .................................................................................................................................80
Annex 2 .................................................................................................................................81
References .............................................................................................................................82
Module 3 The economics of climate change .....................................................................87
1 Introduction ........................................................................................................................87
2 The competitive markets model and Pareto efficiency ......................................................88
2.1 A simple model of market economies ....................................................................88
2.2 Pareto efficiency .....................................................................................................90
2.3 Competitive market equilibria and Pareto efficiency .............................................91
3 Climate change: The greatest market failure in human history .........................................93
3.1 The atmosphere: An open-access resource .............................................................94
3.2 Greenhouse gas emissions: A negative externality problem ..................................96
4 Why is it so hard to solve the climate change problem? ....................................................99
4.1 Mitigating climate change: A global public good.................................................100
4.2 Basic game theory notions and concepts ..............................................................101
4.3 Climate change mitigation: A game theory perspective .......................................105
5 Exercises and questions for discussion ...........................................................................112
Annex 1 ...............................................................................................................................114
References ...........................................................................................................................115
Module 4 The politics of climate change – towards a low-carbon world .....................117
1 Introduction ......................................................................................................................117
2 Policy options and technological solutions to limit climate change ................................118
2.1 Policy instruments and technologies to stabilize concentrations of greenhouse gases
in the atmosphere ........................................................................................................119
2.2 Policies to promote technological solutions aimed at increasing the amount of
incoming solar radiation reflected back into space .....................................................133
3 Climate change adaptation policy options .......................................................................134
4 The international climate change policy architecture ......................................................141
4.1 From Rio to Paris – 25 years of climate change negotiations ...............................141
4.2 The Paris Agreement.............................................................................................144
5 Exercises and discussion questions ..................................................................................150
Annex 1: Some useful databases.........................................................................................152
Annex 2: Selected additional reading material ...................................................................152
References ...........................................................................................................................153
v
List of figures
Figure 1: The economy and the environment: Two interdependent systems ........................... 7
Figure 2: Classification of natural resources ............................................................................ 8
Figure 3: World trade, exports and imports by broad category ................................................ 9
Figure 4: CO2 – IPAT decomposition ..................................................................................... 15
Figure 5: Kaya decomposition: Annex I countries (left panel) and non-Annex I countries
(right panel) ............................................................................................................................. 16
Figure 6: United Nations 2015 world population projection .................................................. 17
Figure 7: IPAT: The impact of projected population growth on CO2 emissions.................... 17
Figure 8: IPAT: The impact of projected GDP per capita growth on CO2 emissions ............ 18
Figure 9: Trade openness and CO2 emissions, 1960–2014..................................................... 27
Figure 10: Trade openness and CO2 emissions, 2011............................................................. 28
Figure 11: Relationship between agreement, evidence and confidence levels ....................... 43
Figure 12: The climate system ................................................................................................ 46
Figure 13: Layers of the atmosphere ...................................................................................... 47
Figure 14: Photosynthesis – the chemical reaction ................................................................. 49
Figure 15: The global mean energy balance of the earth ........................................................ 51
Figure 16: Extreme events - schematic presentation .............................................................. 54
Figure 17: Correlations of surface temperature, precipitation and mean sea level pressure
with the Southern Oscillation Index ....................................................................................... 55
Figure 18: Externally induced climate changes ...................................................................... 58
Figure 19: Atmospheric carbon dioxide, methane and nitrous oxide concentrations from year
0 to 2005 ................................................................................................................................. 60
Figure 20: Key properties of main aerosols in the troposphere .............................................. 62
Figure 21: Radiative forcing of the climate between 1750 and 2011 ..................................... 64
Figure 22: Positive and negative feedback mechanisms......................................................... 66
Figure 23: Observed surface temperature changes from 1901 to 2012 .................................. 68
Figure 24: Observed change in annual precipitation over the land surface ............................ 69
Figure 25: Selected observed changes in snow cover, ice extent and sea level ...................... 71
Figure 26: Observed impacts attributed to climate change ..................................................... 73
Figure 27: Carbon dioxide emission trajectories according to the four representative
concentration pathways ........................................................................................................... 75
Figure 28: Predicted increases in mean surface temperature .................................................. 76
Figure 29: Key anticipated risks per region ............................................................................ 78
Figure 30: Equilibrium in a single competitive market .......................................................... 90
Figure 31: A Pareto-efficient level of output in a single competitive market ........................ 92
Figure 32: Greenhouse gas emissions: A negative production externality problem .............. 98
Figure 33: Payoff matrix of a two-player game with a dominant strategy equilibrium ....... 103
Figure 34: Payoff matrix of a two-player game with two Nash equilibria ........................... 104
Figure 35: Reducing greenhouse gas emissions in a world composed of two countries ...... 106
vi
Figure 36: Overview of implemented or scheduled carbon pricing policy instruments ....... 123
Figure 37: Carbon dioxide emissions per unit of GDP ......................................................... 124
Figure 38: Share of renewable energies in Germany’s energy market ................................. 130
Figure 39: Number of large-scale carbon capture and storage pilot projects per year ......... 131
Figure 40: Direct industrial air capture system ..................................................................... 132
Figure 41: Potential climate change damage share in relation to population share by region in
2050 (per cent) ...................................................................................................................... 135
Figure 42: The Local Adaptation Plans for Action framework ............................................ 140
Figure 43: Estimated climate financing in 2014 ................................................................... 147
Figure 44: Kenya’s carbon dioxide abatement potential by sector ....................................... 149
List of tables
Table 1: World population, affluence and technology, 2014.................................................. 14
Table 2: Trade-induced scale, composition and technique effects ......................................... 33
Table 3: Selected empirical results on trade and air pollution ................................................ 35
Table 4: Changes in the probability of occurrence of extreme events .................................... 72
Table 5: Reducing greenhouse gas emissions in a world of n countries .............................. 108
Table 6: Factors affecting outcomes of collective actions .................................................... 110
Table 7: Selected policy options to stabilize concentrations of carbon dioxide in the
atmosphere ............................................................................................................................ 120
Table 8: Comparison of policy options limiting climate change .......................................... 134
Table 9: Categories and examples of adaptation options discussed in the fifth IPCC
assessment report .................................................................................................................. 137
List of boxes
Box 1: The Paris Agreement ..................................................................................................... 2
Box 2: Patterns of world trade in natural resources .................................................................. 9
Box 3: Interactions among environmental services: The Ganga River case .......................... 11
Box 4: Kaya decomposition: Providing additional insights ................................................... 15
Box 5: The ecocide of the Mayan society ............................................................................... 20
Box 6: Trade in environmental goods and services ................................................................ 23
Box 7: A decomposition of scale, composition and technique effects ................................... 30
Box 8: Selected empirical evidence on the impact of trade on deforestation and water use .. 34
Box 9: The IPCC’s terminology to report findings to the public ........................................... 43
Box 10: Extreme events .......................................................................................................... 53
Box 11: The El Niño-Southern Oscillation – an example of an internal interaction among
components of the climate system affecting the means and variability of different climate
variables .................................................................................................................................. 55
Box 12: Equilibrium in a single competitive market .............................................................. 89
vii
Box 13: A Pareto-efficient level of production in a single market ......................................... 92
Box 14: Different collective action outcomes for identical collective action problems ....... 109
Box 15: The results of the Swedish carbon tax..................................................................... 124
Box 16: Carbon leakage, competitiveness, and carbon border tax adjustments ................... 126
Box 17: Germany’s Energiewende policy ............................................................................ 130
Box 18: Cyclone shelters and early warning systems in Bangladesh ................................... 138
Box 19: The camellones project in Bolivia........................................................................... 138
Box 20: Nepal’s National Framework on Local Adaptation Plans for Action ..................... 140
Box 21: Kenya’s Nationally Determined Contribution objectives and approaches ............. 148
viii
List of abbreviations
°C
ADP
APEC
Ar
AR4
AR5
BA
C6H12O6
CAT
CCN
CCS
CFC
CO
CO2
CO2-eq
COP3
COP 13
COP15
COP16
COP17
COP18
COP21
CH4
EDGAR
EGS
EIT
EMC
ENSO
ERF
ESCII
ETS
GDP
GHG
GtCO2
GW
H2O
HS
IEA
IN
INDC
IPAT
IPCC
Ka
Degrees in Celsius
Ad Hoc Working Group on the Durban Platform for Enhanced Action
Asia Pacific Economic Cooperation
Argon
Fourth IPCC Assessment Report
Fifth IPCC Assessment Report
Border adjustment
Glucose
Cap and trade
Cloud condensation nuclei
Carbon capture and storage
Chlorofluorocarbon
Carbon monoxide
Carbon dioxide
Carbon dioxide equivalent
United Nations Climate Change Conference of the Parties in Kyoto
United Nations Climate Change Conference of the Parties in Bali
United Nations Climate Change Conference of the Parties in
Copenhagen
United Nations Climate Change Conference of the Parties in Cancún
United Nations Climate Change Conference of the Parties in Durban
United Nations Climate Change Conference of the Parties in Bali
United Nations Climate Change Conference of the Parties in Paris
Methane
Emissions Database for Global Atmospheric Research
Environmental goods and services
Economies in transition
External marginal costs
El Niño-Southern Oscillation
Effective radiative forcing
Energy sector carbon intensity index
Emission trading system
Gross domestic product
Greenhouse gas
Gigatonnes of carbon dioxide
Gigawatt
Water
Harmonized Commodity Description and Coding System
International Energy Agency
Ice nuclei
Intended Nationally Determined Contributions
Environmental Impact, Population, Affluence, and Technology
Intergovernmental Panel on Climate Change
Thousands of years
ix
LAPA
Mt
N2
N2O
NAPA
NDCs
NOx
O2
O3
OA
OECD
PMB
PMC
POA
PPB
PPM
PPP
R&D
RCP
RF
SAR
SMB
SMC
SO
SO2
SOA
SOI
SST
TAR
TPES
TOA
UN
UNCED
UNCTAD
UNDESA
UNFCCC
UV
VDC
WCED
Wm-2
WMGHG
WTO
Local Adaptation Plans for Action
Megatonne
Nitrogen
Nitrous oxide
National Adaptation Programme of Action
Nationally Determined Contributions
Nitrogen oxides
Oxygen
Ozone
Organic aerosols
Organisation for Economic Co-operation and Development
Private marginal benefits
Private marginal costs
Primary organic aerosols
Parts per billion
Parts per million
Purchasing power parity
Research and development
Representative concentration pathway
Radiative forcing
Second IPCC Assessment Report
Social marginal benefits
Social marginal costs
Southern Oscillation
Sulphur dioxide
Secondary organic aerosols
Southern Oscillation Index
Sea surface temperature
Third IPCC Assessment Report
Total primary energy supply
Top of the atmosphere
United Nations
United Nations Conference on Environment and Development
United Nations Conference on Trade and Development
United Nations Department of Economic and Social Affairs Population
Division
United Nations Framework Convention on Climate Change
Ultraviolet
Village Development Committees
World Commission on Environment and Development
Watts per square meter
Well-mixed greenhouse gases
World Trade Organization
x
Introduction
Climate change is the biggest global environmental challenge of the 21st century.
Changes in the climate system that are having a significant impact on the environment
and on human beings can already be seen today. Since the 1950s, our atmosphere and
oceans have been warming, sea levels have been rising, and the amounts of snow and ice
have diminished on a global scale (IPCC, 2014). Today, human influence on the climate
system is a scientifically confirmed fact: according to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change (IPCC),1 human activity is extremely likely
to have been the dominant cause of this global warming. If humanity continues to emit
substantial quantities of greenhouse gases (GHG), additional warming and other longterm changes are likely to occur. IPCC (2014: 64) estimates that these changes will
increase the “likelihood of severe, pervasive and irreversible impacts for people, species
and ecosystems” and would lead to “mostly negative impacts for biodiversity, ecosystem
services and economic development and amplify risks for livelihoods and for food and
human security.” Moreover, the IPCC estimates that these risks are generally greater for
people and communities that are socially, economically or otherwise marginalized.
However, there is still scope for action. Humankind can limit climate risks through a
substantial and sustained reduction of GHG emissions over the next few decades (IPCC,
2014). Sustainable Development Goal 13 of the United Nations 2030 Agenda for
Sustainable Development calls for urgent action to combat climate change and its impact.
This drive to structurally transform our carbon-based economy into a low-carbon
economy is now widely acknowledged by policymakers. At the 21st Session of the
United Nations Framework Convention on Climate Change (UNFCCC) in December
2015, the international community adopted the Paris Agreement (see Box 1), which
uniformly acknowledged the importance of climate change mitigation and adaptation
actions to tackle this global challenge, and recognized the need “to strengthen the global
response to the threat of climate change, in the context of sustainable development and
efforts to eradicate poverty” (Paris Agreement, Article 2, §1). Unlike previous
international agreements on climate change, the Paris Agreement calls upon developed
and developing countries to curb emissions while respecting the principle of common but
differentiated responsibilities. The 196 participating parties (195 countries plus the
European Union) agreed on an ambitious goal to limit global warming to “well below
2°C above pre-industrial levels and pursuing efforts to limit the temperature increase to
1.5°C above pre-industrial levels” (Paris Agreement, Article 2, §1a). Limiting global
warming to at most 2°C above pre-industrial levels would, according to IPCC (2014),
lead to only moderate risks of global aggregate effects on biodiversity and humans. To
achieve this objective, our carbon-based economy needs to be radically transformed into
an economy based on low-carbon power sources. Such structural transformation would in
turn reduce human-induced carbon dioxide (CO2) emissions and thereby help reduce
1
The IPCC, founded in 1988 by the World Meteorological Organization and the United Nations
Environment Programme, is the international entity that is assessing the science related to climate change.
The panel is composed of hundreds of leading scientists who review the work of thousands of scientists.
The IPCC regularly provides policymakers with assessment reports of climate change, observed and
anticipated effects, and mitigation and adaptation options.
1
global warming. It could also unleash new opportunities for sustainable growth and
development, and in turn reduce poverty.
Box 1: The Paris Agreement
The Paris Agreement1 is the outcome of the 21st Session of the UNFCCC held in December 2015
in Paris which brings all its 196 parties under a common, legally binding framework. The
agreement is open for signature and ratification at the UN Headquarters in New York from 22
April 2016 to 27 April 2017 and will take effect in 2020.
The Paris Agreement requires any country that ratifies it to act to reduce its GHG emissions in the
coming century, with the goal of peaking global GHG emissions as soon as possible and
continuing the reductions as the century progresses. Ratifying parties will aim to keep global
temperatures from rising more than 2°C by 2100 with an ideal target of keeping the temperature
rise below 1.5°C (Article 2, §1a). Ratifying parties will have to submit what are called Nationally
Determined Contributions (NDCs), i.e. national commitments to reduce greenhouse gases. By the
end of 2015, 188 NDCs had already been submitted. The agreement provides a mechanism that
aims to increase the ambition of NDCs over time: it requires all countries to deliver, every five
years, a new NDC (Articles 3, 4, 7, 9, 10, 11, and 13). Each new NDC should represent a
“progression” over the prior one, and should reflect the country’s “highest possible ambition”
(Article 4.3). This process is crucial, because current NDCs are not strong enough to limit
warming to below 2°C. Moreover, the agreement states that countries will “engage in adaptation
planning processes” (Article 9) to ensure that they are ready for the effects of climate change. To
finance mitigation and adaptation efforts, the agreement places a legal obligation on developed
countries to provide financial resources (Article 9.1) and invites wealthier developing countries to
contribute as well. The overall implementation of the agreement will be assessed every five years,
starting in 2023 (Articles 14.1 and 14. 2).
Unlike past climate change policy frameworks, which had limited scope and ambitious goals, the
Paris regime is based on broad participation and relatively limited ambition at the initial stage.
This change from limited to broad participation is important given the global nature of the climate
change problem. The broad participation, together with the new architecture of the agreement that
combines bottom-up NDCs with top-down procedures for reporting and synthesis of NDCs by the
UNFCCC Secretariat, represents significant progress towards an effective climate policy solution.
1
The text of the Paris Agreement is available at http://unfccc.int/paris_agreement/items/9485.php.
Source: Author.
This teaching material focuses on the ways and means to address our century’s biggest
global environmental, economic and social, challenge: the structural transformation of
our current carbon-based economy into a low-carbon economy. It provides readers with
the necessary tools to analyse this challenge and assist them in devising appropriate
solutions.
Module 1 offers a general introduction to the topic by highlighting the linkages between
the economy and the environment. The module shows that the environment provides
several services to the economy and demonstrates that the economic and environmental
systems are connected and interdependent. It then explains that human use of the
2
environment’s services can lead to environmental impacts such as climate change, and
discusses factors determining the size of these impacts. Finally it underlines the
importance of sustainable economic systems and concludes by discussing the role of
trade as an enabler of sustainable development and poverty reduction.
Module 2 focuses specifically on the climate and climate change by familiarizing the
reader with the climate science behind climate change. The module discusses the
different components of the climate system and shows that they are all influenced by the
planet’s energy balance. It then explains that climate change can occur due to factors that
are either internal or external to the climate system. External factors such as human
activities can affect the climate by influencing climate change drivers such as
atmospheric GHG concentrations, which in turn change the energy balance of the planet
and thus affect climate. After reviewing different ways in which economic activities
affect climate, the module presents the observed changes in the climate system and
discusses the impact those changes have had on human and natural systems. It then
outlines climate changes that are expected to occur in the future as well as their potential
impact.
Following Module 1 and 2, which have covered the general links between the economy
and the environment and explained how human beings can effect climate change, Module
3 introduces readers to the economics of climate change. The module demonstrates that
climate change is arguably the biggest market failure in human history. It explains that
the atmosphere is an open-access resource and that GHG emissions are negative
externalities. Firms and households produce GHG emissions as an unwanted by-product
of their economic activities, without having to pay the cost of related pollution. Instead,
they dump those gases free of charge into the atmosphere, thereby contributing to climate
change, and thus, indirectly, imposing costs on present and future generations. The
module concludes by highlighting that the reduction of emissions is a global public good,
which generates incentives for nations to free-ride and avoid reducing their own
emissions, and makes implementation of effective solutions to climate change a difficult
task.
Module 3 showed that climate change is the result of a large market failure that can only
be corrected by policy interventions at the global level. Over the past 25 years, the
international community has not found a convincing solution to this market failure, and
global greenhouse gas emissions have continued to increase. Nevertheless, at the 2015
United Nations Climate Change Conference (COP21) in Paris, the international climate
change policy framework took a new and promising direction. Module 4 provides an indepth analysis of climate change policies and the international politics of climate change.
It discusses the different policy instruments and technological solutions that could enable
us to limit climate change and transform economies into low-carbon economies. It
explains that, in parallel with policies aimed at limiting climate change, a society also
needs to undertake actions to adapt human and natural systems so that they are better
prepared for the anticipated impacts of climate change. The module then reviews key
international climate change policy developments and discusses several issues
3
surrounding the Paris Agreement, with a particular focus on the situation of developing
countries.
4
Module 1
The environment, the economy and trade: The importance of
sustainable development
1 Introduction
Climate change is the most important global environmental challenge of our century. In
order to analyse climate change and climate change-related policies from an economist’s
point of view, readers need to be equipped with some general conceptual tools. These
tools, briefly introduced in this module, allow them to (a) understand that human beings
affect the climate because the economy and the environment are related and
interdependent; (b) identify the factors that determine the size of the impact an economy
has on the environment; (c) describe what can happen if this impact becomes too severe,
(d) understand the fundamental importance of the comprehensive concept of sustainable
development in today’s context of climate change; (e) analyse why many see trade as an
important enabler of sustainable development; and (f) assess the overall impact of trade
on the environment.
Section 2 investigates the general relationship between the economy and the environment
and the ways in which they interact. The environment provides four key services to the
economy: it serves as a natural resource base, provides life support services, provides
amenity services, and acts as the economy’s waste sink. In turn, the economy affects the
environment through these four key services. This section shows that climate change is
the result of the negative impact caused by the economy’s use of the environment’s
resource and waste services. It then explains the factors that influence the overall size of
an economy’s impact on the environment using a simple model and applying it to the
emissions of CO2, a particularly important greenhouse gas that will be discussed
extensively in the remainder of this teaching material.
Having examined how the economy affects the environment and which factors determine
the overall size of this impact, Section 3 starts by asking what could happen if the size of
this impact were to become too large. It shows that some societies collapsed in the past
because of their excessively negative impact on the environment: this illustrates that
economic systems need to be sustainable. Consequently, the section highlights the
importance of sustainable development in today’s world.
While trade is currently viewed by many as an important enabler of sustainable
development, various concerns have been voiced in recent decades with regard to the
effects of trade on the environment. Section 4 therefore attempts to provide answers to
the question of whether international trade harms or benefits the environment. After
taking a first glance at data, it introduces the reader to a comprehensive theoretical
framework that allows for assessing different channels through which trade affects the
environment. These theoretical tools are then applied to review recent empirical evidence
on the impact of trade on CO2 emissions.
At the end of this module, readers should be able to:
5







Describe why economic activities can affect the climate by discussing the general
linkages between the economy and the environment and understanding why and
how the two systems are considered to be interdependent;
Use a simple model to analyse the effects of changes in population, affluence and
technology on the size of an environmental impact such as GHG emissions;
Understand the potential consequences of unsustainable human behaviour that
ignores environmental constraints by referring to examples of collapses of past
societies;
Describe the fundamental importance of sustainable development for humankind
and assess the role of trade as a key enabler of sustainable development;
Identify the main effects that trade has on the environment;
Analyse important empirical contributions on the impact of trade on emissions;
Assess the role of trade in the transfer of green technologies by discussing the role
of environmental goods and services in the world economy.
To support the learning process, readers will find several exercises and discussion
questions in Section 5 covering the issues introduced in Module 1. Useful data sources
and additional reading material can be found in Annexes 1 and 2.
2 The economy and the environment: Unravelling the links
The economy and the environment are interconnected and interdependent systems, and
both have an impact on economic and social development. Economic activities can
therefore have an impact on the environment and vice versa. This section explains how
the environmental system influences the economy, how economic activities affect the
environment, and what factors determine the size of the economy’s environmental
impact.2
2.1 Links between the economy and the environment
As shown in Figure 1, the economic system and the environmental system are closely
related and influence one another.3
2
The structure of this section was partially inspired by Chapter 2 of Perman et al. (2011) and Chapters 4
and 7 of Common and Stagl (2004).
3
When using the terms “economic system” or “economy” in this section, we are always referring to the
world economic system or the world economy.
6
Figure 1: The economy and the environment: Two interdependent systems
Source: Author's elaboration based on Common and Stagl (2004: 112).
Note: G&S: goods and services.
It is fundamental to understand that the economic system is a part, or sub-system, of the
environmental system (in simple terms, the earth and its atmosphere). This can be seen
graphically in Figure 1, as the boundary of the economic system lies entirely within the
environmental system’s boundary. The environmental system, with the economic system
inside, is in turn part of a larger system – the rest of the universe, with which it exchanges
energy.4
Within the economic system, one finds households that buy and consume goods and
services (G&S) produced by firms, and that provide labour (L) to firms. Firms use labour
(L) provided by households, capital (K) from the capital stock, and natural resources
provided by the environmental system (R) as inputs to produce goods and services. They
sell a part of the goods and services they produce to households and other firms, and
invest (I) the other portion in the capital stock. Firms and households both produce waste
that is partially recycled and reused by firms and partially discharged into the
environment.
The environmental system, which consists of planet earth and its atmosphere, provides
services to the economy. Common and Stagl (2004) define four classes of environmental
services, each of them represented by a green box in Figure 1. In particular, the
environment provides natural resources, amenity services, and life support services to the
economy, and acts as the economy’s waste sink. At the same time, the economy affects
the environment by using these services. Each of the services is analysed in a greater
detail in the sections that follow.
4
The environmental system receives energy inputs from the rest of the universe (in the form of solar
radiation emitted by our sun) and emits energy outputs to the rest of the universe (as thermal radiation).
Refer to Module 2 for a detailed discussion.
7
2.1.1 Providing natural resources to the economy
Firms extract natural resources from the environmental system and use them as inputs in
their production processes. Natural resources can be classified according to a common
classification system used in recent textbooks on natural resource economics or
ecological economics (Common and Stagl, 2004; Perman et al., 2011). This classification
system is based on the two questions presented in Figure 2.
The first question is whether the current use of a resource influences the future
availability of that resource. If the current use of the resource reduces its availability in
the future, the resource is called a stock resource. For instance, the future availability of
animal species, healthy soil, fossil fuels, or minerals is affected by the current use of
them. If firms extract coal, they will reduce the remaining quantity of coal available for
future usage. Firms can thus affect the environment by extracting stock resources and
thereby changing their future availability. If the current use of the resource does not affect
the quantity of the resource available for future use, that resource is called a flow
resource. Typical examples of flow resources are wind or solar radiation. No matter how
much solar radiation firms use today, or no matter how many wind turbines the economy
deploys today, there is strictly no impact on the quantity of solar radiation or wind that
will be available tomorrow. The distinction between flow and stock resources is
fundamental – stock resources can be completely depleted by the economic system, while
flow resources cannot. As we will see in Section 3, completely depleting stock resources
like healthy soils or forests can have devastating effects on the economic system.
Figure 2: Classification of natural resources
Source: Author's elaboration based on the classification in Common and Stagl (2004) and Perman et al.
(2011).
When it comes to stock resources, there is a second question to ask: does the stock
resource regenerate itself? If it does, the stock resource is called a renewable resource. If
it does not, it is called a non-renewable resource. Fossil fuels are examples of nonrenewable resources. No oil will be available for further use once the economic system
has depleted all current oil reserves.5 Animal species like fish are examples of renewable
resources. Fish do reproduce at a certain rate, which is often called a natural growth rate.
5
Note that this is not entirely correct. While the natural growth rate of oil stocks is indeed zero if we
consider human time scales (e.g. years, decades, centuries), there are positive growth rates for longer time
horizons (millions of years).
8
Consequently, as long as the amount of fish the economy extracts (the fish harvest rate) is
equal to the fish natural growth rate, the fish stock remains stable over time.
Box 2: Patterns of world trade in natural resources
From a trade statistics perspective, natural resources are difficult to define precisely. The World
Trade Organization (WTO, 2010: 46) defines natural resources as “stocks of materials that exist
in the natural environment that are both scarce and economically useful in production or
consumption, either in their raw state or after a minimal amount of processing.” Based on this
definition, WTO (2010) classifies fish, forestry products, fuels, ores and others minerals, and nonferrous metals as natural resources.
Figure 3: World trade, exports and imports by broad category
Source: UNCTAD (2015: 17–18).
Note: BRICS: Brazil, Russia, India, China and South Africa; LDCs: least developed countries.
Natural resources are unevenly distributed over the planet’s surface. Several resources are highly
concentrated in specific geographic areas. Trade in natural resources is thus unsurprisingly
substantial. The share of natural resources in world merchandise trade rose from 11.5 per cent in
1998 to 23.8 per cent in 2008 (WTO, 2010). While manufacturing remained by far the largest
broad category of goods traded in 2014, with a share of almost 70 per cent, natural resources were
the second largest. In addition, unlike developed countries, developing countries export more
natural resources than they import (see Figure 3). Moreover, natural resources still represent a
large share of developing countries’ exports, particularly in energy-exporting countries in the
Middle East and primary commodity-exporting countries in Africa (UNCTAD, 2015).
Source: Author's elaboration based on UNCTAD (2015) and WTO (2010).
2.1.2 Providing life support services
The second class of environmental services identified by Common and Stagl (2004) are
life support services, i.e. basic conditions that enable human life. These conditions
include, for instance, liveable temperatures, gravity levels, oxygen levels, etc. Without
these services, no human life could exist on earth.
2.1.3 Providing amenity services
The environmental system provides amenity services to households. Such services are
diverse: the joy of taking a walk in a forest, or the pleasure of swimming in a lake are just
9
two examples. These amenity services often do not need any transformation and can be
directly consumed by households. Moreover, consuming environmental amenity services
does often not have an impact on the future availability of these services. Enjoying flora
and fauna by looking at it does not reduce its future quantity or quality. There are
however exceptions. Mass tourism or illegal harvesting of endangered species, for
instance, can have a negative impact on wildlife and thus affect the environmental system
if organized unsustainably.
2.1.4 Serving as the economy’s waste sink
The environmental system acts as the economy’s waste sink. Households and firms create
waste, an unwanted by-product of economic activity. Waste is understood as a rather
broad category containing such items as chemical, paper, plastic and metal waste from
consumers, organic food waste, CO2 emissions, smoke from burning biomass, etc. It is
important to understand that waste discharges into the environment are direct and
necessary consequences of the extraction and use of resources from the environment. In
physics, the law of conservation of mass explains this phenomenon. Economists often
refer to it using the term “material balance principle,” which states that matter can neither
be created nor destroyed. Economic activity can thus only transform inputs into outputs
but can never create or destroy matter. Consequently, all extracted material will have to
return to the environment in some form and at some point in time. One can therefore look
at the waste sink function as a type of complement of the resource extraction function.
Waste can be partially recycled and subsequently reused as an input in the production
process (see Figure 1). Unrecyclable or unrecycled waste discharged in the environment
may accumulate as a stock. Environmental processes may subsequently reduce the waste
stock. Whether or not the particular stock is increasing depends on the type of waste, the
rate of discharge, and the environment’s capacity to reduce the waste stock. Take, for
instance, plastic waste discharged into oceans. The stock of this particular type of waste
is growing fast because the world economy’s rate of discharge is high and the
environment’s capacity to biochemically transform plastic – and thereby reduce the
plastic stock – is low.
Waste is often not harmful to the environment, but in some cases it induces chemical or
physical changes that can harm living organisms. This is another example of how the
economy affects the environment by using one of the four environmental services. There
are a variety of relationships between the quantity of waste and resulting environmental
damage. Sometimes, waste discharged into the environment has no damaging impact
until a certain threshold is reached. Sometimes damage increases gradually with the
quantity of waste discharged. Take, for instance, CO2 emissions. CO2 itself is not a priori
harmful to living organisms. After all, humans and other animal species produce CO2 by
breathing, and plants consume it in their respiration and photosynthesis processes.
However, although CO2 does not have direct effects on biological systems, it does have a
direct physical impact on the earth’s atmosphere. When emitted in large quantities, CO2
triggers a greenhouse effect (a phenomenon that will be discussed in detail in Module 2)
that contributes to changing the climate of the planet and poses a threat to a wide range of
living organisms, including humans.
10
2.1.5 Interaction among environmental services: Fossil fuel extraction, CO2
emissions, and climate change
The sections above discussed the links between the economic and environmental systems
separately for the four classes of environmental services. In reality, however, in addition
to interacting with the economic system, these services also interact among themselves,
which makes the overall picture even more complicated. Let us take a first glance at
climate change, which is a good example of such multiple interactions.6 Currently, our
economic system relies heavily on energy from fossil fuels. Firms extract fossil fuels like
coal, petroleum or natural gas from the environment. These resources are then consumed
and, as a by-product, emissions of CO2, a major greenhouse gas, are released into the
atmosphere where they accumulate. This is an illustration of a typical interaction between
the environment’s resource service (firms extract fossil fuels) and its waste sink service
(firms emit CO2 from fossil fuel combustion into the atmosphere). The increased
concentration of CO2 in the atmosphere then leads to the so-called greenhouse effect, and
thus to climate change.7
Climate change in turn affects the four environmental services in numerous ways. It can
induce an increase in extreme weather events like severe droughts, which can, for
instance, reduce the availability of healthy soils. It can also cause changes in temperature
and precipitation, altering life support services and thus causing losses of biodiversity and
instability of ecosystems. Amenity services can be affected by the retreat of glaciers,
shortened ski seasons, and unusual heat waves during the summer. And the waste sink
function of rivers can be altered by changes in the assimilative capacity of rivers due to
the rise in temperatures. This illustrative list of potential interactions could be extended
almost indefinitely given the wide range of effects predicted by the IPCC (2014). Finally,
it is important to note that all these changes in turn affect economic activities, for
example by slowing economic growth, lowering agricultural productivity, reducing
tourism activity, decreasing food security, and making it more difficult to reduce poverty.
Box 3: Interactions among environmental services: The Ganga River case
An illustrative example of interacting environmental services is provided by the pollution of the
Ganga River in India. For centuries, the Ganga River has served as a major resource base for
India’s population by providing fish, water, trade routes, etc. The Ganga basin also provides
important amenity services, such as the Hindus’ use of the river for mass bathing and ritual
ceremonies that include cremation of dead bodies. Moreover, the Ganga basin also increasingly
serves as a major waste sink for domestic and industrial waste. Over 1.3 billion litres of sewage
per day and an estimated 260 million litres of largely untreated industrial wastewater pass directly
into the river (Behera et al., 2013). At the same time, the Ganga ecosystem plays an important
part in providing various life support services for numerous animal species, including endangered
ones like the Ganga River dolphin. The Ganga ecosystem has thus been under considerable
pressure due to the inflow of household and industrial waste, intensive water and fish extraction,
and the religious usage of the river that regularly floods it with remains of the dead. These
complex interactions of the Ganga’s waste sink, resource extraction, and amenity service
6
7
A second example of interactions among environmental services is provided in Box 3.
Refer to Module 2 for a detailed discussion.
11
functions has led over recent decades to severe levels of pollution that not only threaten different
resource stocks (e.g. water quality) but also the possibility of benefitting from amenity services
(e.g. health risks associated with taking a bath in the river). Moreover, current pollution levels
have an impact on the life support function of the river and threaten the existence of entire species
like the Ganga River dolphin (Behera et al., 2013). Awareness of the problem has been increasing
over recent decades. Since 1985, several policy initiatives have been put forward, and in 2011 the
World Bank decided to support the Indian National Ganga River Basin Authority with a US$1
billion project called the National Ganga River Basin Project.
Source: Author's elaboration based on Behera et al. (2013).
Short summary
Section 2.1 shed some light at the complex interrelationship between the economic and
environmental systems, which in turn has an impact on economic and social
development. Readers learned that the environmental system provides four key services
to the economy, and that the economy has an impact on the environment by using these
services. A first glance at climate change showed that this phenomenon is the result of
such impacts. The example of climate change also showed how these services can affect
one another and lead to severe cases of environmental damage, in turn affecting
economic activities.
2.2 Environmental impact of economic activities
Section 2.1 showed that the economy affects the environment by using the four
environmental services, and illustrated this mechanism using the example of climate
change. However, we have not yet addressed the important related question of timing.
Why is climate change the biggest environmental problem of the 21st century rather than
having been the biggest problem during the 19th or 20th centuries? Which factors explain
why today’s GHG emissions are at levels that start to threaten the climate system of the
entire planet? Or, put more generally, which factors explain the magnitude of the
environmental impact of economic activities?
Assessing the role of factors that influence the magnitude of the environmental impact of
economic activities is a challenging endeavour. It is therefore useful to rely on simplified
models that allow us to identify the main factors that influence the magnitude of such an
impact. These models can also be used to simulate future impact and thereby provide
useful information about potential future scenarios. The following two sections discuss
two of these models and apply them to the analysis of the size of one particular impact of
economic activity: CO2 emissions.
2.2.1 Unravelling the role of main drivers of environmental impacts by using the
IPAT identity8
A simple but very useful way to think about the size of any environmental impact is
provided by the Environmental Impact, Population, Affluence and Technology (IPAT)
8
Note that an identity is an equation that is always true by definition (see footnote 11).
12
identity. IPAT emerged during a debate on drivers of environmental impact between Paul
R. Ehrlich, John Holdren, and Barry Commoner, three leading ecologists of the 1970s.9
The IPAT identity states that three factors jointly determine the size of any environmental
impact (I): population (P), affluence (A), and technology (T). Intuitively this seems rather
straightforward. All else being equal, the larger the population, the higher the average per
capita consumption, and the more resources a production technology uses and/or the
more waste a technology generates, then the more one can expect an economy’s impact
on the environment to be bigger.
Let us take a look at the climate change example discussed above. One generally expects
that more fossil fuels are extracted and more CO2 emissions from fossil fuel combustion
are emitted if more people live on the planet. This is relatively easy to understand, as
more people means, for instance, increased demand for cars. As most of the cars run on
energy from fossil fuels, this will increase CO2 emissions. At the same time, one expects
that CO2 emissions will increase if these people are more affluent, i.e. they consume more
per person. Higher per capita consumption implies, for instance, increased industrial
production, which requires more energy and in turn increases CO2 emissions from fossil
fuel combustion. Finally, CO2 emissions are expected to be higher the dirtier the
production technology. Put differently, if production technology evolves and becomes
cleaner, one would expect, all else being equal, fewer CO2 emissions. If factories start to
use machines that require less energy to do the same job, this should reduce CO2 emission
levels.
IPAT allows us to visualize the relationship between population, affluence, technology,
and the magnitude of the environmental impact they can create. The general form of the
IPAT identity captures the three main factors driving the size of environmental impact
and can be written as:
𝐼 ≡ 𝑃 ∗ 𝐴 ∗ 𝑇,
where, as stated earlier, I stands for environmental impact, P for population, A for
affluence, and T for technology.10
The general IPAT identity can be applied to a variety of different environmental effects.
Depending on how I is defined, IPAT allows for analysing insertions into the
environment (e.g. GHG emissions) and extractions from the environment (e.g. fish or
coal extraction).
To illustrate how one can use the IPAT identity, we will continue to focus on one specific
environmental impact – CO2 emissions – and apply the IPAT identity to global
anthropogenic (i.e. human-induced) CO2 emissions. The sources of data used in this
analysis are detailed in Table 1.
9
See Chertow (2000) for an overview on the history of the IPAT identity.
Depending on the application, there are a variety of sets of units that can be used with the IPAT identity.
According to Common and Stagl (2004), I can be measured in tons or litres, depending on the particular
impact one attempts to analyse. P is always measured in numbers of people. A is measured as the total
economic output in currency units (often gross domestic product – GDP - in US dollars) divided by the
number of people. T is measured as units of impact (for instance tons or litres) per US dollar of GDP.
10
13
Table 1: World population, affluence and technology, 2014
Source: Author.
1
CO2 emissions include CO2 emissions generated by the use of fossil fuels and industrial processes but
exclude CO2 emissions generated by short-cycle and large-scale biomass burning.
Note: PPP: purchasing power parity. International dollar is a hypothetical unit of currency widely used in
economics. By construction, it has the same purchasing power parity that the US dollar had in the United
States at a given point in time (in this case, 2011).
Using the data for P, A, and T we can calculate the impact I as
$
ktons
𝐼=⏟
7,260,710,677 persons ∗ 14,287
∗ 0.000000344
person ⏟
$
⏟
P
T
A
= 35,684,418.06 ktons ,
and find that the world economy emitted 35,684,418.06 kilotons of CO2 in 2014. At this
point, it is important to note that IPAT is an accounting identity and thus holds by
construction.11
Applying IPAT to a particular year allows us to understand how the identity works, but it
is by far more interesting to apply the method over a longer time horizon. By doing so,
one can analyse which of the three factors have contributed to increase or reduce
emissions in the recent past. Figure 4 shows the change in anthropogenic CO2 emissions
from the use of fossil fuels and industrial processes over the period of 1970-2013. CO2
emissions from fossil fuel use and industrial processes increased by roughly 125 per cent
over the last four decades. Figure 4 also shows the decomposition of these emissions into
factors that caused them, namely P, A and T. World population and world affluence have
been increasing fast (by 97 per cent and 89 per cent, respectively) while technology has
become cleaner (by 39 per cent).
Overall, Figure 4 reveals that the emission-reducing effect of cleaner technology has not
been able to offset the emission-generating effects of increased population and affluence.
In other words, the sharp increase of CO2 emissions over recent decades has mainly been
driven by population growth and increased affluence. Thus, to answer our initial question,
climate change is the biggest environmental problem of the 21st and not the 19th or 20th
century because of the joint evolution of population, affluence, and technology: world
population and affluence have been increasing rapidly and are currently at their highest
levels in human history, while technology is still relatively dirty and cannot yet cancel out
11
To see this, note that we calculated T=
𝐶𝑂2
𝐺𝐷𝑃
. The IPAT identity 𝐶𝑂2
⏟ = 𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛
∗
⏟
𝐼
thus collapses to 𝐶𝑂2 = 𝐶𝑂2, which by definition always holds.
14
𝑃
𝐺𝐷𝑃
𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛
⏟
𝐴
𝐶𝑂2
∗⏟
𝐺𝐷𝑃
𝑇
the emission-increasing effects of population and affluence. Together, these three factors
explain why CO2 emissions in the early 21st century have risen to such high levels,
causing accumulations of CO2 in the atmosphere that threaten the climate of our planet.
IPAT can thus be very useful in identifying which factors contributed to increasing or
reducing the overall size of the economy’s environmental impact. Box 4 shows a related
decomposition using the Kaya identity, which provides additional interesting insights into
the drivers of past CO2 emissions. Section 2.2.2 will show that IPAT can be used not only
to analyse the past, but also to glance at the future.
Figure 4: CO2 – IPAT decomposition
Source: Author's elaboration based on data on population and GDP at market prices (constant 2005 US
dollars) from the World Bank’s Data Catalog, and on CO2 emissions data from The Netherlands
Environmental Assessment Agency (Oliver et al., 2015).
Box 4: Kaya decomposition: Providing additional insights
The IPAT identity discussed in this section has been extended by the Japanese economist Yoichi
Kaya (1990) to provide additional insights in the context of CO2 emissions. This identity is
frequently used to decompose CO2 emissions (Nakicenovic and Swart, 2000). The so-called Kaya
identity can be expressed as follows:
𝐸
𝐶𝑂2
𝐸
≡𝑇
𝐶𝑂2 = 𝑃 ∗ 𝐴 ∗ ⏟
∗
𝐺𝐷𝑃
,
where CO2 = anthropogenic CO2 emissions; P = population; A = GDP per capita; E = primary
energy consumption; and GDP = gross domestic product.
The Kaya identity decomposes IPAT’s technology variable T into two parts: energy intensity of
GDP (E/GDP) and carbon intensity of the energy mix (CO2/E). IEA/OECD (2015) use the Kaya
identity and decompose CO2 emissions for two groups of countries: Annex I and non-Annex I
countries of the UNFCC.1
Figure 5 displays the decompositions by country group. According to IEA/OECD (2015), the
recent decline in emissions in Annex I countries was driven by a significant reduction in the
energy intensity of GDP (TPES/GDP), and by a slight fall in the CO2 intensity of the energy mix
(CO2/TPES). These two emission-reducing effects have offset the emission-generating effects of
15
the growth in GDP per capita and in population. In non-Annex I countries the growth in CO2
emissions was mainly driven by the increase in GDP per capita and to a lesser extent by the
increase in population.
Figure 5: Kaya decomposition: Annex I countries (left panel) and non-Annex I countries
(right panel)
Source: IEA/OECD (2015: xx)
Note: TPES: total primary energy supply. ESCII: energy sector carbon intensity index.
The UNFCCC divided countries in two broad groups – Annex I and non-Annex I countries – depending
on their 1992 commitments to fight climate change. Annex I countries consist of countries that were
members of the Organisation for Economic Co-operation and Development (OECD) in 1992 plus countries
with economies in transition in 1992, such as the Russian Federation. Non-Annex I countries are the
remaining ones, mostly developing, countries that signed the convention (see Module 4 for more
information about the different groups of countries and their commitments).
1
Source: Author’s elaboration based on IEA/OECD (2015) and Kaya (1990).
2.2.2 Assessing potential future scenarios of environmental impact using the IPAT
equation
The previous section showed that population growth and increased affluence caused a
sharp increase in CO2 emissions. This section uses IPAT to glance at the future and
analyse the potential impact on emissions of expected future changes in population and
affluence. The IPAT framework can help answer a number of questions: What could
happen to total emissions if world population continues to increase? How could different
GDP per capita growth scenarios affect total emissions? What role do technological
improvements play in moderating the effects of increased population and affluence on
total emissions?
Let us start with population growth. According to 2015 world population projections
from the United Nations Department of Economic and Social Affairs (UNDESA)
Population Division (Figure 6), world population, currently at 7.35 billion, is unlikely to
stop growing this century. On the contrary, the UN estimates that there is an 80 per cent
probability that world population in 2100 will be between 10.03 billion and 12.44 billion
(with a median of 11.21 billion). How could this expected growth in population affect
emissions? To address this question we assume that GDP per capita (A) and technology
(T) stay at their 2014 levels, and we use the projected UN population data to simulate
16
CO2 emissions using IPAT. Figure 7 displays observed annual CO2 emissions from 1970
to 2014 and then adds the three IPAT-based emission scenarios using the UN population
projections. All other factors fixed, the predicted population growth could result in an
increase of 38 to 72 per cent of annual CO2 emissions between 2014 and the end of the
century.
Figure 6: United Nations 2015 world population projection
Source: Author's elaboration based on data from UNDESA (2015).
Note: The 95 (80) per cent prediction interval means that the UN estimates a 95 (80) per cent probability
that this situation will occur.
Figure 7: IPAT: The impact of projected population growth on CO2 emissions
Source: Author's elaboration based on data from UNDESA (2015), World Bank (2016), and the
Netherlands Environmental Assessment Agency (see Oliver et. al., 2015).
A related question is how different GDP per capita growth paths might affect emissions if
population and technology were to stay at their 2014 levels. To provide an answer we
rely on the baseline long-term global growth projection for 2010–2060 published by the
17
OECD (2014).12 We complement this projection (which predicts an average yearly GDP
per capita growth rate of slightly below 2.5 per cent) with a low-growth scenario (with a
yearly growth rate of only 1 per cent) and a high-growth scenario (with a yearly growth
rate of 4 per cent). Figure 8 shows observed CO2 emissions from 1971 to 2014, as well as
those under the three GDP per capita growth rate scenarios. As we see, with fixed 2014
population and technology levels, the OECD GDP per capita projection would result in
an increase of 200 per cent in CO2 emissions by 2060 with respect to 2014. The highgrowth scenario would result in an increase of more than 500 per cent, while the lowgrowth scenario would result in an increase of roughly 60 per cent.
Figure 8: IPAT: The impact of projected GDP per capita growth on CO2 emissions
Source: Author's elaboration based on data from the OECD (2014), World Bank (2016), and the
Netherlands Environmental Assessment Agency (Oliver et. al., 2015).
The analysis in Figure 7 and Figure 8 revealed that the predicted growth of population
and GDP per capita are both likely to increase CO2 emissions considerably over the next
couple of decades. The impact of the combined effects of population growth and
increased affluence on emissions will be even bigger. This raises the question of how
humanity will be able to significantly reduce CO2 emissions without limiting population
or affluence growth. Technology, the third determinant of the size of the overall impact
within the IPAT model, might be the answer. In the prior analysis, we held technology
fixed at the 2014 level. However, as population and GDP per capita are both expected to
increase significantly over the coming decades, technological improvements will have to
play an important role if humanity intends to curb CO2 emissions and thereby stabilize
CO2 concentrations in the atmosphere. To offset the effects of a growing and increasingly
affluent population, it will therefore be crucial to rely on cleaner technology that can
significantly reduce CO2 emissions per produced US dollar.
Short summary
12
See Johansson et al. (2013) for the model underlying this projection.
18
In Section 2.2, readers learned how to use the simple IPAT model to analyse forces
influencing the size of the environmental impact of an economy. Higher population levels
and increased affluence generally increase the magnitude of the environmental impact,
while technology improvements have the potential to decrease the size of that impact.
The IPAT model was used to decompose a particular environmental impact, CO2
emissions, leading to a conclusion that the observed increase in CO2 emissions was due to
an increase in population and affluence that was not offset by the effects of cleaner
production technology. Subsequently, IPAT was used to glance at the future by
establishing carbon dioxide scenarios based on various predictions of population growth
and increased affluence.
3 Sustainability of the economic system
The previous section showed how an economy affects the environment and discussed
different factors that influence the magnitude of the world economy’s environmental
impact. This section starts by investigating what could happen if the size of such an
impact were to become too large. It shows that some societies in the past collapsed
because of their devastating impact on the environment. It thus illustrates why an
economic system should be sustainable. In this context, it introduces the concept of
sustainable development and shows why trade is currently viewed by many as an
important enabler of sustainable development.
3.1 Ecocide: Lessons from history
As a subsystem of our environmental system, the economic system faces a variety of
environmental constraints that, if systematically disregarded, could potentially have
devastating effects on human societies.13 In 1798, economist Thomas Malthus published
his famous book An Essay on the Principle of Population in which his central hypothesis
was that population growth will eventually outpace food production due to diminishing
returns in agriculture. Malthus was convinced that the land’s capacity to produce food –
an environmental constraint – could not keep up with human reproduction rates. In his
view, this would inevitably lead to important food shortages, which in turn would
significantly reduce human population and force those who remain to live at subsistence
levels. Since the Industrial Revolution, we have known that Malthus’s predictions in
terms of food production were wrong because innovations enabled increased food
production to feed a growing population. Nevertheless, this does not imply that
environmental constraints have disappeared.
Economic systems that ignore environmental constraints face problems that might even
lead to a collapse of the entire system. In his book Collapse: How Societies Choose to
Environmental constraints are limits imposed on humankind by the environment. For instance, a river’s
capacity to absorb toxic waste is limited, which is an environmental constraint that restricts the amount of
toxic waste humans can discharge into the river without destroying it. If humans disregard this constraint
systematically (e.g. by pumping large quantities of toxic waste into the river), the river’s ecosystem might
be destroyed, which in turn might affect the humans and their economic activities. Another environmental
constraint, which is very important in the context of climate change, is the limited capacity of the
atmosphere to absorb greenhouse gases.
13
19
Fail or Survive, Jared Diamond (2006) analysed the role of environmental constraints in
the collapse of past societies like the Vikings, Greenland’s Norse society, Easter Island’s
Polynesian society, North America’s Anasazi society, the Mayans, and others.
Diamond (2006: 3) defines collapse as “a drastic decrease in human population size
and/or political/economic/social complexity, over a considerable area, for an extended
time.” According to him, unintended ecological suicide – ecocide – is a major
explanatory factor for the rapid decline of several past societies. Diamond identifies a
common pattern of the decline of these collapsed societies. The process started with the
growth of population and the intensification of agricultural production (through improved
technology and geographical expansion of agriculture to agriculturally marginal lands).
Unsustainable agricultural practices then led to a variety of environmental degradation
processes (e.g. deforestation, water mismanagement, erosion, overhunting, etc.) that
forced people to abandon the agriculturally marginal lands, with food shortages and
starvation often resulting. People started wars over the remaining resources and
overthrew governing elites. Population levels declined and, in parallel, the complexity of
the society in terms of economics, politics, and culture decreased. This led in some
extreme cases to a complete collapse of the society, with the entire population dying or
emigrating. Box 5 illustrates such a collapse using the example of the Mayan society.
Box 5: The ecocide of the Mayan society
The ecocide of the Mayan society is a particularly interesting example of a collapse discussed by
Diamond (2006). The Maya were one of the culturally most advanced societies of their time in
Central America. The classical age of Mayan society begun around the year 250. The Mayan
population increased exponentially and reached its peak in the 8th century and then declined
rapidly in the 9th century.
Why did the Mayan society collapse? Diamond explains the collapse using the example of the
city of Copán studied by Webster et al. (2000). Copán was a small, densely populated city in
today’s western Honduras. Fertile land was concentrated in the river valley surrounded by
relatively unfertile steep hills. Starting in the 4th century, Copán’s population started to grow
rapidly. According to Webster et al. (2000), agricultural production was intensified to satisfy
increased demand due to population growth. By the mid-6th century, the productive capacity of
the fertile valley land was exhausted and people started to cultivate on the steep hilly land around
the valley. They cut down the forests on the hill slopes that had previously protected the hills
from erosion. After some time, hill slopes eroded and the quality of soils deteriorated.
Consequently, people had to move back into the valley. The infertile acidic hill soils started to
spread into the valley and partially covered the fertile valley soils, further reducing the productive
capacity of the land. Diamond argues that the deforestation may also have begun to cause a manmade drought in the valley because the water-cycling functions of the forests were reduced as
more and more trees were cut down. During this period, the population continued to grow rapidly
while the productive capacity of the land diminished equally rapidly. This led to fights over land
and food that culminated in the destruction of the royal palace around the year 850. Population
size dropped and after 1250, there were no more signs of humans living in the Copán valley.
According to Diamond, Copán’s collapse is a good example of the mechanisms behind the
collapse of the entire Mayan society. Population growth outpaced available resources and led to
deforestation and erosion, which further reduced the size of fertile lands and potentially triggered
local man-made climate change that increased the risks of droughts. Fighting over scarce
20
resources and an inability of the governing elite to address the problems finally ended in a total
collapse of the Mayan society.
Sources: Author’s elaboration based on Diamond (2006) and Webster et al. (2000).
Diamond (2006: 7) identified several processes through which past societies transgressed
environmental constraints and damaged the environment up to the point that these
societies collapsed. These processes include overuse of renewable stock resources (e.g.
through deforestation, overhunting, overfishing) and misuse of renewable stock resources
(e.g. soil degradation, water mismanagement, adverse effects of introduced species on
native species). These processes still play a major role today. Moreover, according to
Diamond, new environmental problems have been added to the ones that threatened past
societies. Climate change is one of these new problems and is arguably the most
dangerous one for humanity in the 21st century. Humankind has been systematically
ignoring the environment’s limited capacity to absorb greenhouse gases. This
unsustainable behaviour is threatening the entire climate system of the earth and will have
important consequences for humankind unless sufficient actions are taken to reduce GHG
emissions. With the adoption of United Nations Agenda 2030, the Paris Agreement at the
21st Conference of the Parties, and relevant outcomes of other international summits, the
international community has reached agreement to collectively address the imminent and
present danger of environmental and climate change through national and global actions
in the next 15 to 30 years.
3.2 Sustainability
As Diamond (2006) showed, ignoring environmental constraints by engaging in
unsustainable economic behaviour at times resulted in the total collapse of entire
societies. Cohen (2009) argues that the sustainability question today is even more
important than ever. He advances the argument that past competition of different
organizational forms of the economic system – some more efficient than others – has
almost come to an end and resulted in today’s dominance of a single form of economic
organization: the modern market economy. Thus, according to Cohen, it is the absence of
a viable alternative to the modern market economy that makes the question of its
sustainability crucial.
There are various definitions of sustainability that differ depending on the context.14 The
Oxford English Dictionary states that “sustainability” is a derivative of the word
“sustainable,” which is defined as “able to be maintained at a certain rate or level” or
“able to be upheld or defended.” In ecology, the definition of sustainability focuses on the
continued productivity and functioning of ecosystems and often involves requirements to
protect genetic resources and biological diversity (Brown et al., 1987). In environmental
economics, sustainability is often defined with respect to the continued coexistence of the
economic and environmental systems. We follow this strand of definitions in this
material. Common and Stagl (2004: 8) define sustainability as follows: “Sustainability is
maintaining the capacity of the joint economy-environment system to continue to satisfy
14
For an overview of different usages of the concept of sustainability, see Brown et al. (1987).
21
the needs and desires of humans for a long time into the future.” Processes that are
threatening the joint economy-environment system in a way that current and future needs
and desires of humans cannot be satisfied are thus by definition called unsustainable
processes. Processes that do not threaten the joint economy-environment system are
called sustainable processes.
3.3 Sustainable development: The international awakening
The importance of a sustainable world economy has long been neglected. This changed in
1987 with the publication of the World Commission on Environment and Development
(WCED) report entitled Our Common Future (WCED, 1987). This report, now frequently
called the Brundtland report (after the chair of the commission, Gro Harlem Brundtland),
has become a milestone in terms of durably placing sustainability questions on the
international policy agenda.
The Brundtland report outlined the complex interactions between the economic and
environmental systems, stressed the importance of environmental constraints, and argued
that the current path of economic growth cannot be continued. Acknowledging the need
to eradicate poverty and raise per capita income worldwide, the report suggested an
alternative pathway of economic growth that it called “sustainable development.” The
report defined sustainable development as “development that meets the needs of the
present without compromising the ability of future generations to meet their own needs. It
contains within it two key concepts: the concept of ‘needs,’ in particular the essential
needs of the world’s poor, to which overriding priority should be given; and the idea of
limitations imposed by the state of technology and social organization on the
environment’s ability to meet present and future needs.” (WCED, 1987: 45)
Sustainable development is thus presented as the solution that would simultaneously (a)
allow for meeting the needs of the present generation through continued economic
growth, thereby raising standards of living worldwide, and (b) maintain the capacity of
the joint economy-environmental system to meet the needs of future generations.
Sustainable development is thus seen as the tool to enable us to steer the world economy
towards a sustainable path that serves both people (present and future generations) and
the planet.
Since the publication of the Brundtland report, the definition of sustainable development
has further evolved and is today considered to be a combination of three pillars:
environmental protection, economic development, and social development. Following the
2012 United Nations Conference on Environment and Development (UNCED)
conference in Rio de Janeiro, which was a direct consequence of the Brundtland report,
this definition of sustainable development was widely adopted by international
organizations (Lehtonen, 2004). Twenty years later at the Rio+20 UN Conference on
Sustainable Development, the three pillars of sustainable development were elevated to
global prominence with the agreement to develop sustainable development goals. Today,
sustainable development plays a key role in combating climate change and provides the
framework within which the Paris Agreement aims to “strengthen the global response to
the threat of climate change, in the context of sustainable development and efforts to
22
eradicate poverty” (Paris Agreement, Article 2, §1). Module 4 of this teaching material
reviews policies promoting sustainable development in the context of human-made
climate change. The module explains in detail how the current carbon-based economic
system is supposed to be transformed into a low-carbon economy.
3.4 Trade as a key component of sustainable development
The previous section argued that sustainable development is the solution to maintain the
capacity of the joint economy-environmental system for current and future generations
and simultaneously raise standards of living worldwide. In this context, the international
community – including the United Nations Conference on Trade and Development
(UNCTAD), the WTO, the United Nations Environment Programme, and various
multilateral environmental agreements – have highlighted the role that trade can play in
achieving sustainable development.
Paragraph 6 of the WTO’s Doha Ministerial Declaration stipulates that its member
countries “are convinced that the aims of upholding and safeguarding an open and nondiscriminatory multilateral trading system, and acting for the protection of the
environment and the promotion of sustainable development can and must be mutually
supportive.” The contribution of trade to sustainable development has also been
recognized during several international conferences, including the 1992 UNCED
conference in Rio de Janeiro, the 2002 World Summit on Sustainable Development in
Johannesburg, and the 2012 Rio+20 Conference. In the 2030 Agenda for Sustainable
Development, the international community recently renewed its commitment to making
trade a key enabler of sustainable development.
According to the WTO (2011), increased trade openness leads to increased resource
efficiency, higher growth rates, and higher income levels. Trade is thus seen as
supporting sustainable development objectives by affecting at least two of the sustainable
development pillars: economic development and environmental protection. By promoting
production efficiency, offering new opportunities for the sale of products, and increasing
the availability of high-quality and low-price inputs, trade is viewed as helping to reduce
poverty by stimulating economic development. Trade can also directly affect the
environmental protection pillar of sustainable development by (a) increasing the
efficiency of natural resource use due to a trade-induced increase in competition and
resulting efforts to reduce costs by reducing consumption of natural resources, thereby
supporting conservation efforts; and (b) making access to environmental goods and
services easier for all countries, thereby acting as a channel for green technology transfers
(WTO, 2011). Liberalization of trade in such eco-efficient technologies might help to
support sustainable development globally. This is certainly an important topic, given that
Section 2.2.2 showed that technologies will need to moderate the impact of expected
increases in population and affluence on CO2 emissions. Box 6 takes a closer look at
trade in environmental goods and services. Moreover, as UNCTAD (2014) points out,
trade can play an important indirect role in sustainable development by mobilizing
financial resources that might be used to finance sustainable development objectives.
Box 6: Trade in environmental goods and services
23
One of the ways in which trade can effectively support sustainable development is by facilitating
access to environmental goods and services (EGS). Trade in these goods and services is
frequently assumed to have the potential for “win-win” outcomes, generating both environmental
benefits (by disseminating environmental goods and services) and trade gains for countries
involved (UNCTAD, 2003).
The EGS sector only emerged in the 1990s as a distinct sector, but it has since grown and
changed rapidly. The relative newness of the concept makes defining the sector difficult. While
there are several national definitions that vary in scope, the OECD and Eurostat took the lead at
the international level to define the sector (UNCTAD, 2003). According to OECD/Eurostat
(1999), the EGS sector consists of “activities which produce goods and services to measure,
prevent, limit, minimize or correct environmental damage to water, air and soil, as well as
problems related to waste, noise and ecosystems.” By their nature, EGS are scattered across an
array of different economic sectors. Despite the difficulty of defining the sector, there have been
attempts at estimating the size – in other words the total value produced – of the EGS market.
Using its own definition of the sector, Environmental Business International (2012) estimated the
global size of the EGS market at US$832.2 billion in 2011, representing roughly 1.2 per cent of
global GDP. While developed countries still dominate the market in absolute terms, developing
countries from Africa, the Middle East, and Asia show the highest growth rates in that year.
Global exports of environmental goods as classified by the OECD list of environmental goods
and services increased by almost 190 per cent, from US$231 billion to US$656 billion, over
2001–2012 (Bucher et al., 2014).1 These figures would be even higher if one took trade in
environmental services into account. In 2012, trade in EGS was dominated by trade between
developed countries. During 2008–2013, eight of the 10 leading exporters and seven of the 10
leading importers of EGS were developed countries (Bucher et al., 2014). It is interesting to note,
however, that the People's Republic of China is already now the second largest exporter and
importer of EGS. Moreover, other developing countries like the Republic of Korea, Mexico,
Brazil, Malaysia, and the Russian Federation also play an increasingly important role in the
global EGS market. Generally speaking, most analysts expect that developing countries will
significantly increase their global export and import shares over the years to come (Bucher et al.,
2014). Hamwey (2005) identifies significant export strengths and potential for developing
countries in environmentally preferable products, EGS from the manufacturing and chemical
sector, and environmental services.
Given the importance of the EGS sector for the environment and its high market growth rate,
trade liberalization of the sector is considered to be an important issue. During the 2001 Doha
WTO Ministerial Conference, WTO member states codified their agreement to open negotiations
on “the reduction or, as appropriate, elimination of tariff and non-tariff barriers to environmental
goods and services” in paragraph 31 (iii) of the Doha Ministerial Declaration. The declaration
emphasizes that negotiations should enhance the mutual supportiveness of trade and the
environment in terms of both environmental benefits and trade gains for the parties involved
(UNCTAD, 2003). Lamy (2008) argues that agreement on paragraph 31 (iii) would be a direct
and immediate contribution of the WTO to fighting climate change.
Nevertheless, no WTO agreement on EGS trade liberalization has been reached to date, and
important tariff and non-tariff barriers continue to exist in this sector. Much of the environmental
goods negotiation has revolved around finding an acceptable sector definition among WTO
members, and coming up with a list of EGS for which tariffs and barriers would be reduced.
Several proposals have been submitted, most of them based on list approaches,2 but none have
gained a consensus due to conflicting interests on product coverage and negotiation modalities
24
among the negotiating parties (UNCTAD, 2009/2010). Developing countries in particular
expressed concerns related to the definitions of EGS, the potential inclusion of dual-use goods,
liberalization approaches, environmental regulation, and technology transfer.3
While multilateral negotiations seem stalled, plurilateral negotiations on EGS are taking place. In
the Asia Pacific Economic Cooperation (APEC) 2012 Vladivostok Declaration, APEC members
announced a reduction in tariffs on 54 environmental goods to a maximum of 5 per cent by 2015.
Details of the implementation of these cuts were published in early 2016. Bucher et al. (2014)
argue that this particular initiative might provide new impetus for WTO efforts in this area.
1
The OECD list of environmental goods and services covers 164 HS-6 products and includes product
categories such as pollution management, cleaner technologies, and products and resource management.
For an overview of the OECD list and a comparison with other lists of environmental goods and services,
see Steenblik (2005).
2
List approaches try to identify sets of products to be considered environmental goods and services.
Identification is mostly based on the Harmonized Commodity Description and Coding Systems (HS).
Different lists have so far been put forward, for example by Japan, APEC, OECD, and UNCTAD.
According to Ramos (2014), these lists, with the exception of the UNCTAD list, mainly reflect the export
interests of developed countries in non-agricultural trade.
3
See Ramos (2014) for a discussion and a list of studies identifying these concerns.
Source: Author.
Although trade is seen by many as an important enabler of sustainable development
because of its ability to disseminate environmental goods and services, several civil
society environmental groups have raised concerns about negative effects of trade on the
environment. These voices have been particularly loud in the 1990s and 2000s. The
following section therefore reviews evidence on this issue by looking at theoretical
models and empirical results about the effect of trade on the environment.
Short summary
In Section 3, readers learned that unsustainable behaviour played an important part in the
decline and collapse of some past societies, reminding us to take climate change
seriously, as it is this century’s biggest global environmental problem. We illustrated the
fundamental role of environmental constraints in the development of economic systems
and societies at large, and thus demonstrated the importance of a sustainable economic
system. We introduced the concept of sustainable development as a means of creating a
sustainable world economy and showed why trade plays an important role in current
sustainable development strategies.
4 Impact of trade on the environment
The previous sections of this module (a) showed that human beings can affect the climate
because the economy and the environment are interrelated and interdependent; (b)
identified the factors that determine the size of the impact an economy has on the
environment; (c) illustrated what can happen if these effects become too damaging; (d)
introduced the concept of sustainable development as a means of avoiding ecocide while
25
simultaneously helping reduce poverty; and (e) showed that trade is viewed as an
important enabler of sustainable development.
Over recent decades, especially around the start of the new millennium, numerous
environmental groups opposed further trade liberalization, fearing a negative impact on
the environment (Copeland and Taylor, 2003). The idea that trade negatively affects the
environment is also widespread among the general public. A 2007 survey conducted in
countries covering 56 per cent of the world population found that in several countries a
majority of people perceive trade to be bad for the environment. In none of the surveyed
countries did large majorities believe that trade is beneficial for the environment (The
Chicago Council on Foreign Affairs and World Public Opinion, 2007). In this context,
this section aims to provide theoretical tools and empirical insights to shed light on the
fundamental question that has been at the heart of these debates: Is international trade
good or bad for the environment? Given the overall thrust of this material, the focus will
be on climate change.
Section 4.1 introduces the topic by looking at the data on trade openness and CO2
emissions. Section 4.2 presents a conceptual framework to systematically examine how
increased trade openness affects the environment. This theoretical framework is then used
in Section 4.3 to discuss several important empirical contributions on the impact of
increased trade openness on the environment.
4.1 Trade, trade openness and the environment: A first glance at the data
Does increased trade openness have an impact on the environment? If so, is this impact
positive or negative? To help answer these questions, this section focuses on a particular
environmental impact – CO2 emissions – and looks at descriptive statistics about CO2
emissions and trade openness. While this type of analysis does not allow for making
causal statements, it is a good starting point to obtain a general idea of the nature of the
relationships at work.
When analysing the question of whether increased trade openness increases CO2
emissions, we first examine whether trade openness (measured here as exports and
imports as a percentage of GDP, and merchandise exports and imports as a percentage of
GDP) and CO2 emissions evolve in a similar manner over time.15 In other words, we
look at whether an increase (decrease) in trade openness is paralleled by an increase
(decrease) in CO2 emissions. If this is the case, we would have a first hint that trade
openness and the particular environmental impact we are looking at (CO2 emissions)
could be connected. Results of this initial descriptive analysis are shown in Figure 9.
15
It is important to distinguish between measures of trade (i.e. the absolute volume of trade flows) and
measures of trade openness. Measures of trade openness can either indicate the actual importance of trade
in the economy (these are called measures of trade openness in practice) or quantify the number and/or
importance of trade policy measures in place (measures of openness in policy). We use here two indicators
of measures of trade openness in practice. For an overview of different measures of trade and trade
openness, see UNCTAD (2010: 18–21).
26
The left panel of Figure 9 displays the evolution of world CO2 emissions and the two
indicators of world trade openness. CO2 emissions and both trade openness indicators
have been steeply rising since 1960. While total exports and imports represented 25 per
cent of world GDP in 1960, they corresponded to 59 per cent in 2014. Merchandise
exports and imports rose from 18 per cent of world GDP in 1960 to almost 49 per cent in
2014. Over the same time period, CO2 emissions increased by roughly 270 per cent. Both
trade openness indicators and CO2 emissions thus seem to evolve in a similar manner.
Annual changes in the three variables are displayed in the right panel of Figure 9. Growth
rates of CO2 emissions are positively correlated with growth rates of trade openness
(correlation coefficients equal 0.26 for total exports and imports and 0.32 for
merchandise exports and imports). This means that, on average, CO2 emissions increased
(decreased) in the same year as trade openness increased (decreased). This is a second
hint that trade openness and environmental effects could be connected. At first glance it
thus seems that increased trade openness goes hand-in-hand with increased CO2
emissions. It is important to note, however, that the results from Figure 9 do not allow for
making any causal statement on the relation between trade openness and emissions, as we
are only considering correlations. Nevertheless, these results provide a clear rationale for
a more detailed analysis, as trade openness and CO2 emissions seem to be linked, which
could potentially reflect a causal relationship.
Figure 9: Trade openness and CO2 emissions, 1960–2014
Source: Author's elaboration based on data from World Bank’s World Development Indicators database.
Given that the results from Figure 9 suggest that trade openness and CO2 emissions are
positively associated, we use cross-section data – capturing one point in time (here the
year 2011) for several countries – to identify potential underlying reasons for this positive
relationship. Following the methodology suggested by Onder (2012), we look at the
relation between trade openness (as measured by merchandise exports and imports as a
percentage of GDP) and different factors that could explain the link we found in the
previous analysis. The results are displayed in Figure 10 which contains four different
panels.
The upper left panel shows that there is a positive correlation between trade openness (as
measured by the GDP share of merchandise trade) and CO2 per capita emissions. In other
27
words, countries that trade more extensively seem to emit on average more CO2 per
person than countries that are less open to trade. This result is in line with the results we
found in Figure 9. Which factors could potentially explain this positive association? The
remaining three panels offer several possible explanations of this result.
Figure 10: Trade openness and CO2 emissions, 2011
Source: Author's elaboration based on data from the World Bank’s World Development Indicators
database. The figure reproduces Figure 2 from Onder (2012) with more recent data.
Note: Countries with oil rents higher than 30 per cent of GDP were excluded (oil rents correspond to the
difference between the value of crude oil production at world prices and total costs of production). In
addition, three outliers (Singapore, Aruba, and Hong Kong (China)) with high trade openness indicators
were also excluded. Sample sizes differ due to data availability. The sample includes 195 countries in the
top panels, 184 in the lower left panel, and 149 in the lower right panel. PPP: purchasing power parity.
A first potential explanation is provided in the upper right panel of Figure 10, which
explores the relation between trade openness and GDP per capita. We see that countries
that are more open to trade consume and produce more per person than countries with
less openness to trade. As we have seen in the previous sections, higher per capita
consumption and production (in other words, greater affluence) tends to increase the
impact on the environment. Therefore, it is possible that trade indirectly affects the
environment by increasing economic output.
A second potential explanation is found in the lower left panel of Figure 10 which
analyses the relation between trade openness and the share of value added produced in
the industrial sector. This panel shows that countries with a relatively high trade openness
ratio seem to produce a larger proportion of their value added in industrial sectors, which
are typically relatively energy-intensive. This could mean that the sectoral composition of
countries that trade frequently is biased towards energy-intensive goods, which also
28
produce more CO2 emissions. One could interpret this observation as an empirical hint
that trade affects the sectoral composition of countries, which in turn indirectly affects the
environment.
Finally, a third potential explanation is found in the lower right panel of Figure 10 which
analyses the relation between trade openness and CO2 emission intensity (CO2 emissions
per dollar produced). This panel shows that countries with higher trade openness ratios
seem to have higher average CO2 emission intensity. In other words, it is possible that
countries that are relatively more open to trade use more polluting technologies in their
production processes than countries that are less open to trade. This observation could be
an empirical hint that trade somehow affects technology and thereby indirectly influences
the environment.
The first cursory look at data in Section 4.1 reveals two important findings on whether
increased trade openness increases CO2 emissions. First, trade openness and CO2
emissions evolved in parallel over recent decades, and per capita CO2 emissions are
higher for countries with greater trade openness. Evidence thus seems to indicate that
there is at least a positive correlation between trade openness and CO2 emissions. Second,
without being able to make any causal statement, data provide preliminary evidence for at
least three different possible explanations for this positive correlation: (a) higher trade
openness seems to go hand-in-hand with increased per capita production; (b) the sectoral
composition of countries that trade frequently seems to be biased towards energyintensive goods, which produce more CO2 emissions; and (c) countries that trade
frequently seem to employ more polluting technologies than countries that trade less. The
following section will take a look at theoretical concepts that could explain these
preliminary empirical results.
4.2 Environmental impact of trade: What we can learn from theory
The analysis in Section 4.1 revealed several reasons that might explain why CO2
emissions are different for countries with different degrees of trade openness. The
theoretical framework outlined in this section aims to explain these results and serve as a
starting point for empirical work on the environmental impact of trade.
4.2.1 Scale, composition and technique effects
The current theoretical discussion on trade and the environment is framed by the concepts
introduced by Grossman and Krueger (1993) in their now-famous work analysing the
environmental impact of the North American Free Trade Agreement. The authors
distinguish three effects that economic activities such as trade can have on the
environment: the scale effect, the composition effect, and the technique effect (Table 2).
The scale effect is rather straightforward: if the overall scale of economic activity
increases, all else being equal, the environmental impact will increase. Besides the overall
scale of activities, the production mix of an economy is also important. All else being
equal, a shift in the composition of an economy in terms of the share of clean and dirty
sectors will affect the overall environmental impact of that economy. If an economy
changes its industry mix and starts to produce relatively more dirty goods, the overall
29
environmental impact increases. The opposite will happen if an economy’s production
mix shifts towards cleaner industries. This effect is called the composition effect. Finally,
technology also matters – if the scale and composition of the economy are fixed, but
production technology becomes cleaner, then the overall environmental impact of the
economy decreases, and vice versa. This is called the technique effect. Box 7 shows how
these effects can be derived formally. Each of these effects and their relation to trade are
analysed individually in the sections that follow.
Box 7: A decomposition of scale, composition and technique effects
Scale, composition and technique effects as introduced by Grossman and Krueger (1993) can
be formally derived. Let us focus again on CO2 emission. Furthermore, let us make our life
easier by assuming that an economy produces only two types of goods, dirty goods and clean
goods, and that emissions are only a by-product of the production process of dirty goods (clean
goods do not produce any emissions). We can then express total CO2 emissions of the economy
(E) as the product of total economic output (Y) times the share of dirty goods in total output (S)
times emissions per unit produced of the dirty good (A):
𝐸 = 𝑌 ∗ 𝑆 ∗ 𝐴.
Let us take logarithms and use the properties of the logarithm to obtain:
𝑙𝑛(𝐸) = 𝑙𝑛(𝑌) + 𝑙𝑛(𝑆) + 𝑙𝑛(𝐴).
And finally, let us totally differentiate the above equation, which yields:
∆𝐸
𝐸
=
∆𝑌
𝑌
+
∆𝑆
𝑆
+
∆𝐴
.
𝐴
∆𝐸
From the above, we see that the percentage change in emissions ( 𝐸 ) is equal to the percentage
∆𝑌
𝑌
change in total output ( ) plus the percentage change in the share of dirty goods in the
∆𝑆
economy ( 𝑆 ) plus the percentage change in the emissions per unit produced of the dirty good
∆𝐴
∆𝑌
( ). We see that the scale effect (a change in the overall scale of the economy ( )), the
𝐴
𝑌
composition effect (a change in the composition of the economy in terms of dirty and clean
∆𝑆
goods ( 𝑆 )), and the technique effect (a change in emissions per produced unit of the dirty good
∆𝐴
∆𝐸
( 𝐴 )) all affect overall emissions ( 𝐸 ).
Source: Author’s elaboration based on Grossman and Krueger (1993).
4.2.2 Trade-induced scale effect
It is often claimed that increased trade openness stimulates economic growth, and hence
the scale of economic activities (production, consumption, and transportation) in trading
countries.16 Consequently, if nothing else changes, these increased economic activities
16
The link between trade and growth has long been one of the key questions in economics. Numerous
theoretical models have been tested by an armada of empirical contributions. While there are many
30
will in turn have a higher total impact on the environment. (See Section 2.2.2 for an
example of how GDP growth can affect CO2 emissions.) Grossman and Krueger (1993)
called this mechanism the trade-induced scale effect. Researchers generally expect tradeinduced scale effects to have negative consequences for the environment. Since the
seminal contribution of Grossman and Krueger, numerous empirical papers (Antweiler et
al., 2001; Copeland and Taylor, 2003; Frankel and Rose, 2005; Managi et al., 2009) have
attempted to analyse the existence and size of trade-induced scale effects. Some of this
empirical evidence is reviewed in Section 4.3.
4.2.3 Trade-induced composition effect
Increased trade openness not only influences the scale of economic activities but also the
sectoral composition of trading economies. Standard trade models assume that countries
that increase their openness to international trade specialize in sectors where they have a
comparative advantage.17 Thus, opening up to trade may change a country’s sectoral
structure. The country may specialize in relatively clean sectors if it has a comparative
advantage in these sectors, or in relatively “dirty” or polluting sectors if it has a
comparative advantage in those sectors. Increased trade openness worldwide may
therefore lead to a change in production patterns across countries and thus modify the
impact of the world economy on the environment. The question then is whether tradeinduced changes in economies’ sectoral composition increase or reduce overall
environmental impacts.
From a theoretical point of view, this depends on the sources of the countries’
comparative advantage. The literature (De Melo and Mathys, 2010; Grossman and
Krueger, 1993; Managi et al., 2009) frequently distinguishes two types of trade-induced
composition effects. The first type, factor-endowment effects, arises from classical
sources of comparative advantage (factor abundance and technology differences). The
second type, pollution-haven effects, results from differences in environmental policy
stringency in different countries.
Taken in isolation, the potential net environmental impact of factor-endowment effects is
unclear. Theory predicts that developed countries with relatively high capital-labour
ratios tend to specialize in capital-intensive sectors after opening up to trade. As
production technologies in capital-intensive industries are often resource- and/or
pollution-intensive (Managi et al., 2009), this tends to increase the environmental impact
in these countries. Developing countries with rather low capital-labour ratios tend to
specialize in labour-intensive sectors after opening up to trade. This tends to decrease the
environmental impact in these countries. The net impact of factor-endowment effects on
the world environment is thus ambiguous.
theoretical arguments in favour of a positive relationship between trade and growth, one also finds various
arguments against the existence of such a positive relationship. Empirical tests of these arguments have so
far not been conclusive, sometimes showing a positive and sometimes a negative relation between trade and
growth. For an introduction to this topic, see UNCTAD (2010), Chapter 1 of Module 2.
17
For an overview of different trade models, see UNCTAD (2010), Chapter 1.
31
Theory generally predicts a negative net environmental impact of pollution-haven effects.
Pollution-haven effects tend to increase the comparative advantage of developing
countries in polluting industries because environmental regulations are often less
stringent in these countries. This can lead to shifting the production of these industries
from developed countries where industry is heavily regulated to developing countries
where there is less regulation. Such a shift would lead to an increase of the environmental
impact worldwide.
While explaining these mechanisms, theory cannot help in determining whether the
overall impact of trade-induced composition effects is positive or negative for the
environment. Everything depends on the net impact of factor-endowment effects and the
relative strength of factor-endowment and pollution-haven effects. Section 4.3 will look
at several empirical studies that have contributed to this debate.
4.2.4 Trade-induced technique effect
Trade can also influence the environment through the so-called trade-induced technique
effect, which refers to the fact that production technology may change after a country
opens up to trade. If technology changes, it is possible that the amount of resources used
or the amount of emissions generated per unit produced also change. This would in turn
affect the overall environmental impact of trade. Grossman and Krueger (1993) identify
two mechanisms through which trade can alter production technology: trade-induced
technology transfers, and changes in environmental policies. Both mechanisms are
particularly important for developing countries.
The first mechanism concerns trade-induced technology spillover effects. Firms may
bring new technologies to economies that have opened up to trade. As newer
technologies are frequently assumed to be cleaner than older ones, trade-induced
technology transfers are generally expected to have a positive impact on the environment.
The second mechanism is indirect. Environmental quality is usually assumed to be a
normal good (Antweiler et al., 2001). In economics, a good is referred to as normal when
the demand for that good increases with increased incomes. Saying that environmental
quality is a normal good means that demand for it increases when per capita income
increases. One of the reasons for such an increase of per capita income can be opening to
trade (see Section 4.2.2). Rising per capita income then increases the demand for
improved environmental quality in the countries concerned. Their citizens put pressure on
their governments to tighten environmental regulations, which in the end is beneficial for
the environment. Stricter environmental regulations might thus be an indirect
consequence of greater openness to trade.
32
Table 2: Trade-induced scale, composition and technique effects
Source: Author's elaboration based on the theoretical framework of Grossman and Krueger (1993) and on
the distinction in De Melo and Mathys (2010) between factor-endowment and pollution-haven effects.
4.3 Environmental impact of trade: What we can learn from empirical
evidence
Before the seminal work of Grossman and Krueger (1993), which outlined the theoretical
foundations that allow for systematically analysing the effects of trade openness on the
environment, few empirical contributions had been made in the field. Since then,
however, empirical research to assess the effects of trade on the environment has been
growing rapidly. Most of the research has so far focused on local pollutants (De Melo and
Mathys, 2010). Recently some authors also started to analyse the effects of trade on
deforestation or water use (see Box 8). In line with the focus of this material, this section
continues to concentrate on the empirical evidence on trade openness and CO2 emissions,
the most researched greenhouse gas to date. Our review, however, also includes a second
gas, sulphur dioxide (SO2), which is the gas responsible for acid rain. While SO2 is not a
greenhouse gas, SO2 and CO2 emissions are highly correlated, with the same energyintensive industries being the main emitters (De Melo and Mathys, 2010). It is therefore
meaningful to include SO2 in the analysis.
Antweiler et al. (2001) analyse the impact of increased trade openness on concentrations
of SO2.18 They estimate scale, composition and technique effects and show for the first
18
Note that while emissions correspond to the amount of a gas released into the atmosphere from a specific
source and over a specific time interval, a concentration is the amount of the gas in the atmosphere per
volume unit at a specific point in time.
33
time that these effects can actually be measured using available data and therefore are not
just abstract theoretical concepts. Their sample covers 43 countries over the period 1971–
1996. The empirical results reveal the existence of a trade-induced composition effect
that is lowering SO2 concentrations, i.e. SO2 concentrations decrease due to a tradeinduced change in the sectoral composition of the average economy. Moreover, they find
evidence for a scale effect (SO2 concentrations increase when GDP increases) and a
technique effect (SO2 concentrations decrease when per capita income increases).
According to their results, if an increase in trade openness generates a 1 per cent increase
in income and output, then pollution concentrations will fall by approximately 1 per cent
due to the joint impact of scale and technique effects. Overall, their results suggest that
increased trade openness seems to have a beneficial effect on reducing SO2
concentrations.
Box 8: Selected empirical evidence on the impact of trade on deforestation and water use
While Section 4.3 discusses the empirical evidence on the influence of trade on selected air
pollutants, it is important to note that trade also affects other aspects of the environment, such as
water use and deforestation. However, the empirical literature on the impact of trade on other
aspects of the environment is still scarce, and more research is needed.
The links between deforestation and trade have been empirically assessed by Frankel and Rose
(2005) and Van and Azomahou (2007). These contributions do not find a significant effect of
trade openness on deforestation and are thus inconclusive. However, in analysing the effects of
increased trade openness on deforestation for 142 countries over 1990–2003, Tsurumi and
Managi (2012) find a significant effect: their results indicate that trade-induced changes in the
sectoral composition of economies accelerated deforestation in developing countries but slowed
deforestation in developed countries.
The effects of increased trade openness on water use (the degree to which water is withdrawn and
consumed) have been analysed by Kagohashi et al. (2015). Their sample covers 43 countries over
1960–2000. Their results indicate that trade increased water use through the scale and
composition effects. However, these two effects are more than offset by the water use reducing
technique effect. Overall, their results suggest that a 1 per cent increase in trade openness reduces
water use by roughly 1 to 1.5 per cent on average. Thus trade seems to promote efficient water
use. The authors argue that their results can be explained by the role of trade in diffusing watersaving technologies and modifying industrial composition.
Source: Author.
Building on Antweiler et al. (2001), several other papers used a similar empirical
approach to estimate scale, technique and composition effects for different gases and
different countries. The findings of some of these papers are summarized in Table 3. The
first finding is that the overall effect of increased trade openness on the environment
depends on the gas. For SO2, the results from Antweiler et al. (2001) have been partially
confirmed. Overall, increased trade intensity seems to lower SO2 emissions. However,
more trade seems to increase CO2 emissions. The second finding is that the impact of
trade on emissions differs substantially depending on the countries considered. Managi et
al. (2009) find that trade decreases SO2 and CO2 emissions in OECD countries but
increases them in non-OECD countries. Finally, the third finding is that the sign and
34
magnitude of scale, composition and technique effects also depend on the combination of
the gas and the country considered: trade seems to increase emissions through scale
effects and lower them through technique effects. Depending on the gas and the country,
these two effects can offset one another. The impact of composition effects related to the
change of economic structures towards cleaner or more polluting production are gasspecific and rather small, which Cole and Eliot (2003) attribute to the simultaneous
existence of pollution-haven and factor-endowment effects that cancel each other out.
Table 3: Selected empirical results on trade and air pollution
Source: Author.
Note: The “Effects” column reports the observed impact of scale, technique and composition effects on the
concentrations or emissions of the gas. Not all papers separated the scale and technique effects – sometimes
combined scale and technique effects are reported.
In short, empirical contributions show that while trade does have an effect on the
environment, it is not possible to make a generally valid statement that this effect is
positive or negative. Sometimes, more trade seems to have a positive impact on the
environment, while other times it seems to have a detrimental effect. As results differ
from one case to another, further research is needed on a case-to-case basis to deepen our
understanding of this important question.
Short summary
Section 4 addressed the question of whether trade is harmful or beneficial to the
environment, using the example of CO2 emissions. After showing that trade openness and
CO2 emissions are correlated, the section discussed three theoretical effects through
which trade could affect the environment. Table 2 summarized the expectations with
regard to the environmental impact of each of these effects. While theory provides a
framework that enables us to conceptualize the effects of trade on the environment, it
cannot by itself make a final prediction as to whether trade is harmful or beneficial to the
environment. The section therefore reviewed selected empirical work relevant in the
context of climate change, concluding that trade seems to sometimes have a positive
impact and sometimes a negative impact on the environment.
35
5 Exercises and questions for discussion
1. Why are the economy and the environment considered to be two interdependent
systems? Describe the services the economy provides to the environment.
2. How can you classify natural resources into flow resources, non-renewable stock
resources, and renewable stock resources? Classify the following natural resources
and explain your reasoning:
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
Power of tides
Coal
Power of wind
Cattle
Petroleum
Wood
Solar radiation
Fish
Cobalt
3. Describe the general form of the IPAT identity and discuss the role of each
component (P, A, and T).
4. Download the Excel dataset from http://vi.unctad.org/tenv/files/data_exercise.xlsx.
The dataset covers 1970–2014 and contains three variables: world population (P);
world GDP per capita (A); and world CO2 emissions per world GDP (T).
(a) Using the IPAT identity, calculate world CO2 emissions for each year in the
dataset.
(b) Plot all four variables over time, and discuss the effect each driver (P, A, and
T) has on emissions (I).
(c) Suppose the world wants to stabilize its emissions at the 1970 level. Suppose
further that the world can only influence T to achieve this objective (i.e. the
evolution of the world population and world GDP per capita for the years
1970–2014 is fixed as given by the data). Calculate how much emission
intensity of technologies (T) would need to decrease each year to achieve the
objective, given the growth in population and GDP. Hint: For each year t,
calculate the technology level T̂t needed to keep emissions at the 1970 level,
given the population and GDP per capita levels in year t.
(d) Compare and discuss the observed evolution of T and the hypothetical
evolution of 𝑇̂𝑡 .
5. Using the examples provided by Diamond (2006), illustrate why economic systems
that ignore environmental constraints face problems. Can you think of other examples
not mentioned in this teaching material?
36
6. Define sustainable development. Discuss potential implications of sustainable
development for developing countries.
7. Identify key environmental goods and services produced in your country.
8. How can trade affect CO2 emissions? Distinguish, define, and discuss the scale,
composition and technique effects.
37
Annex 1: Some useful databases
Annex 2: Selected additional reading material
38
References
Antweiler W, Copeland BR, and Taylor MS (2001). Is free trade good for the
environment? American Economic Review 91(4): 877–908.
Behera SK, Singh H, and Sagar V (2013). Status of Ganges River dolphin (Platanista
gangetica gangetica) in the Ganga River basin, India: A review. Aquatic Ecosystem
Health & Management 16: 425–32.
Brown BJ, Hanson ME, Liverman DM, and Merideth RW (1987). Global sustainability:
Toward definition. Environmental Management 11: 713–19.
Bucher H, Drake-Brockman J, Kasterine A, and Sugathan M (2014). Trade in
environmental goods and services: Opportunities and challenges. International Trade
Centre Technical Paper. Geneva.
Chertow MR (2000). The IPAT equation and its variants. Journal of Industrial Ecology
4: 13–29.
Cole MA Elliott RJR (2003). Determining the Trade-Environment Composition Effect:
The Role of Capital, Labor and Environmental Regulations. Journal of Environmental
Economics and Management 46(3): 363–383.
Cohen D (2009). La prospérité du vice. Une Introduction (inquiète) à l’économie. Albin
Michel. Paris.
Common M, and Stagl S (2004). Ecological Economics – An Introduction. Cambridge
University Press. Cambridge, UK.
Copeland BR, and Taylor MS (2003). Trade and the Environment: Theory and Evidence.
Princeton University Press. Princeton, NJ and Oxford, UK.
Davis SJ, Peters GP, and Caldeira K (2011). The supply chain of CO2 emissions.
Proceedings of the National Academy of Sciences 108(45): 18554–8559.
De Melo J, and Mathys N (2010). Trade and climate change: The challenges ahead.
CEPR Discussion Paper No. 8032. Centre for Economic Policy Research. London.
Diamond J (2006). Collapse: How Societies Choose to Fail or Survive. Penguin. London.
Dihel NC (2010). Understanding trade in environmental services: Key issues and
prospects. In: International Trade in Services: New Trends and Opportunities for
Developing Countries, Cattaneo O, Engman M, Saez S, and Stern R, eds. World Bank.
Washington, DC.
39
Environmental Business International (2012). Global environmental markets.
Environmental Business Journal 25(6/7). Available at: http://ebionline.org/ebjarchives/1355-ebj-v25n06-07.
Frankel JA, and Rose AK (2005). Is trade good or bad for the environment? Sorting out
the causality. The Review of Economics and Statistics 87(1): 85–91.
Grossman G, and Krueger AB (1993). Environmental impacts of a North American Free
Trade Agreement. In: The Mexico-US Free Trade Agreement, Garber P, ed. MIT Press.
Cambridge, MA.
Hamwey R (2005). Environmental goods: Where do the dynamic trade opportunities for
developing countries lie? Cen2eco Working Paper. Centre for Economic and Ecological
Studies.
IEA/OECD (2015). CO2 Emissions from Fuel Combustion 2015. International Energy
Agency and Organization for Economic Cooperation and Development. OECD
Publishing. Paris.
IPCC (2014). Climate Change 2014: Synthesis Report. Contribution of Working Groups
I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate
Change [Core Writing Team, Pachauri RK, and Meyer LA (eds.)]. Intergovernmental
Panel on Climate Change. Geneva.
Johansson A, Guillemette Y, Murtin F, Turner D, Nicoletti G, de la Maisonneuve C,
Bagnoli P, Bousquet G, and Spinelli F (2013). Long-term growth scenarios. OECD
Economics Department Working Paper No. 1000. OECD Publishing. Paris.
Kagohashi K, Tsurumi T, and Managi S (2015). The effects of international trade on
water use. PLoS ONE 10(7).
Kaya Y (1990). Impact of carbon dioxide emission control on GNP growth: Interpretation
of proposed scenarios. Paper presented to the Intergovernmental Panel on Climate
Change Energy and Industry Subgroup, Response Strategies Working Group, Paris.
Kolstad C (2000) Environmental Economics. Oxford University Press. Oxford, UK.
Lamy P (2008). A consensual international accord on climate change is needed. Paper
presented at the Temporary Committee on Climate Change, The European Parliament.
Brussels.
Lehtonen M (2004). The environmental–social interface of sustainable development:
Capabilities, social capital, institutions. Ecological Economics 49(2): 199–214.
Managi S, Hibiki A, and Tsurumi T (2009). Does trade openness improve environmental
quality? Journal of Environmental Economics and Management 58(3): 346–63.
40
Nakicenovic N, and Swart R (eds.) (2000). IPCC Special Report on Emissions Scenarios
- A Special Report of IPCC Working Group III. Intergovernmental Panel on Climate
Change. Cambridge University Press, UK.
OECD (2014). OECD Economic Outlook No. 95 (Edition 2014/1). Organization for
Economic Cooperation and Development Economic Outlook: Statistics and Projections
(database). Available at: http://dx.doi.org/10.1787/data-00688-en.
OECD/Eurostat (1999). Environmental Goods and Services Industry Manual for the
Collection and Analysis of Data. Organization for Economic Cooperation and
Development. Paris.
Olivier JGJ, Janssens-Maenhout G, Muntean M, and Peters JHAW (2015). Trends in
global CO2 emissions - 2015 report. PBL Report 1803. PBL Netherlands Environmental
Assessment Agency and European Commission Joint Research Centre. Available at
http://edgar.jrc.ec.europa.eu/news_docs/jrc-2015-trends-in-global-co2-emissions-2015report-98184.pdf.
Onder H (2012). Trade and climate change: An analytical review of key issues. Economic
Premise No. 86. World Bank, Washington, DC.
Perman R, Ma Y, Common M, Maddison D, and McGilvray J (2011). Natural Resource
and Environmental Economics, Fourth edition, Pearson Education Limited. Essex, UK.
Ramos MP (2014). The impact of trade liberalization of environmental products on
welfare, trade, and the environment in Argentina. UNCTAD Virtual Institute Project on
Trade and Poverty. United Nations. Geneva.
Steenblik R (2005). Environmental goods: A comparison of the APEC and OECD lists.
OECD Trade and Environment Working Paper No. 2005-4. Organization for Economic
Cooperation and Development. Paris.
Strange T, and Bayley A (2008). Sustainable Development, Linking Economy, Society,
Environment. OECD Publishing. Paris.
The Chicago Council and World Public Opinion.org (2007). World public favors
globalization and trade but wants to protect environment and jobs. Available at:
http://www.worldpublicopinion.org/pipa/pdf/apr07/CCGA+_GlobTrade_article.pdf.
Tietenberg T, and Lewis L (2012). Environmental and Natural Resource Economics,
Ninth Edition. Pearson International Edition. Addison Wesley. Boston.
Tsurumi T, and Managi S (2012). The effect of trade openness on deforestation:
empirical analysis for 142 countries. Environmental Economics and Policy Studies 16:
305–24.
41
UNCTAD (2003). Trade and Environment Report 2003. United Nation. Geneva.
UNCTAD (2009/2010). Trade and Environment Report 2009/2010. United Nations.
Geneva.
UNCTAD (2010). Virtual Institute Teaching Material on Trade and Poverty, United
Nations. New York and Geneva.
UNCTAD (2014). The role of trade in financing for sustainable development. Discussion
Paper. Available at:
http://www.un.org/esa/ffd/wp-content/uploads/2014/12/10Dec14-UNCTAD-input.pdf.
UNCTAD (2015). Key statistics and trends in international trade. United Nations.
Geneva.
UNDESA (2015). World Population Prospects: The 2015 Revision. United Nations
Department of Economic and Social Affairs Population Division. New York.
Van PN, and Azomahou T (2007). Nonlinearities and heterogeneity in environmental
quality: An empirical analysis of deforestation. Journal of Development Economics 84:
291–309.
WCED (1987). Our Common Future. World Commission on Environment and
Development. Oxford University Press and United Nations. New York.
Webster D, Freter A, and Golin N (2000). Copan: The Rise and Fall of an Ancient Maya
Kingdom. Harcourt Brace Publishers. Fort Worth, TX.
World Bank (2016). Open Data Catalog. World Bank. Washington, DC. Available at:
http://datacatalog.worldbank.org/.
WTO (2010). World Trade Report: Trade in Natural Resources. World Trade
Organization. Geneva.
WTO (2011). Harnessing Trade for Sustainable Development and a Green Economy.
World Trade Organization. Geneva.
42
Module 2
The climate science behind climate change
1 Introduction
Module 1 showed that human actions can influence the climate because the economy and
the environment are interdependent. This module focuses specifically on the climate
system and the climate science behind climate change. By reviewing key points, the
module enables readers to understand how the climate system works, why climate change
occurs, and how humans induce climate change. The module highlights recent empirical
evidence pointing towards the existence of human-induced climate change, and discusses
climate-change-related impacts that can already be observed today. To conclude, the
module outlines different scenarios for the future and discusses the anticipated impacts of
climate change.
To this end, this module draws on the work of the Intergovernmental Panel on Climate
Change (IPCC) Working Group I, which deals with the physical science basis of climate
change. As mentioned in Module 1, the IPCC is the leading body collecting, reviewing,
and assessing recent state-of-the-art scientific work related to climate change. The five
IPCC assessment reports and the different IPCC special reports 19 together form the most
comprehensive and reliable source of scientific work today on climate change. The
module therefore uses some of the terminology introduced by the IPCC (see Box 9).
Box 9: The IPCC’s terminology to report findings to the public
The Intergovernmental Panel on Climate Change (IPCC) developed specific terminology to report
its findings to the public. In the fifth assessment report (IPCC, 2013a), a finding is assessed in
terms of the underlying evidence and the agreement among scientists regarding that finding. In
addition, the IPCC often assigns a level of confidence to the different findings that is based on
evidence and agreement (see Figure 11).
Figure 11: Relationship between agreement, evidence and confidence levels
Source: Mastrandera et al. (2010: 3).
To rate the amount of evidence, the IPCC uses the terms “limited,” “medium,” and “robust.” To
19
All
these
reports
are
available
on
the
IPCC
http://www.ipcc.ch/publications_and_data/publications_and_data_reports.shtml#1.
43
homepage
at:
rate the degree of agreement among scientists regarding the evidence, it uses the terms “low,”
“medium,” and “high.” Confidence levels are expressed using the terms “very low,” “low,”
“medium,” “high,” and “very high.” If the likelihood of an outcome or a result has been assessed
using statistical techniques, the IPCC reports probability values using the following terminology:
“virtually certain” (99–100 per cent probability); “very likely” (90–100 per cent probability);
“likely” (66–100 per cent probability); “about as likely as not” (33–66 per cent probability);
“unlikely” (0–33 per cent probability); “very unlikely” (0–10 per cent probability); and
“exceptionally unlikely (0–1 per cent probability). Sometimes it also uses the terms “extremely
likely” (95–100 per cent probability); “more likely than not” (>50–100 per cent probability);
“more unlikely than likely” (0–<50 per cent probability); and “extremely unlikely” (0–5 per cent
probability). For more details, see Mastrandrea et al. (2010).
Source: Author's elaboration based on Mastrandera et al. (2010).
Section 2 introduces the theoretical basis of the climate system and climate change.
Section 2.1 familiarizes the reader with the concepts of weather, climate, and climate
change, and introduces the five components of the climate system (i.e. atmosphere,
hydrosphere, cryosphere, land surface, and biosphere). Section 2.2 focuses on the planet’s
energy balance, which influences all five components of the climate system and is
therefore of crucial importance for the climate. The section stresses that the natural
greenhouse effect plays an integral role in the planet’s energy balance and is generally
responsible for the relatively warm average temperature on the surface. With the module
having outlined how the climate system works, Section 2.3 then explains that the climate
not only changes in response to factors within the climate system but also in response to
some factors that are external to the climate system. In fact, some external factors (e.g.
human activities) affect climate change drivers (e.g. greenhouse gases), which in turn
affect the energy balance of the planet. Because the earth’s energy balance affects all five
components of the climate system, external factors can thus affect the climate. Section
2.4 introduces two important concepts that allow for assessing how the planet’s energy
balance changes as a result of externally induced variations in climate change drivers:
radiative forcing and effective radiative forcing. These two important concepts allow for
quantifying the extent to which different external factors influence the climate. Finally,
Section 2.5 specifically focuses on the different ways in which humans affect the climate,
altering atmospheric concentrations of greenhouse gases and aerosols, and influencing
properties of the land surface. It concludes by assessing the relative strength of these
different human-induced perturbations of the climate system, and highlights the
importance of feedback effects.
Section 3 then provides an overview of the main changes that have occurred in the
climate system and highlights to what extent human activities have contributed to these
observed changes. To this end, the section discusses observed changes in the means of
temperature, precipitation, ice and snow cover, and sea levels. It then shows that not only
the mean state of climate variables has changed, but that the frequency of extreme climate
events has also been affected. The section concludes with a short discussion on the
impact of these observed changes on human and natural systems.
To conclude Module 2, Section 4 looks at the future of the planet’s climate system. To
this end, it introduces different scenarios describing possible pathways of climate change
44
drivers that are used by the IPCC’s climate models to project future changes in the
climate system. After discussing these scenarios, Section 4 highlights the most important
predicted changes in the main climate variables (temperature, precipitation, snow and ice
cover, and sea levels) that are likely to occur until 2100. The section ends by providing
selected examples of anticipated future risks for humans and natural systems that might
result from the simulated changes in the climate system.
At the end of this module, readers should be able to:







Distinguish the five components of the climate system;
Understand the radiative balance of the planet;
Explain how the natural greenhouse effect operates;
Define and understand the concept of radiative forcing;
Explain how economic activities can alter the climate;
List major observed human-induced changes of the climate system;
Discuss anticipated future impacts of climate change.
To support the learning process, readers will find several exercises and discussion
questions in Section 5 covering the issues introduced in Module 2. Additional reading
material can be found in Annex 2.
2 The theoretical basis of the climate system and climate change20
When speaking about climate change, it is important to clearly distinguish three distinct
but related concepts: weather, climate, and climate change. Weather can be defined as the
changing state of the atmosphere, which is characterized by temperature, precipitation,
wind, clouds, etc. (Baede et al. 2001). Weather fluctuates frequently as a result of fastchanging weather systems. Weather systems, and hence the weather, can only be
predicted with some degree of reliability for a very short period of time (one or two
weeks). They are unpredictable over longer time horizons. Climate, on the other hand, is,
loosely speaking, “long-term average weather.” IPCC (2001a: 788) defines climate as
“the statistical description in terms of mean and variability of relevant quantities [such as
temperature, precipitation, and wind] over a period of time ranging from months to
thousands or millions of years. The classical period is 30 years, as defined by the World
Meteorological Organization.” Climate not only varies from location to location
(depending on a variety of factors such as distance to the sea, latitude and longitude, the
presence of mountains, etc.) but also over time (e.g. from season to season, year to year,
century to century, etc.). Based on this definition, IPCC (2001a: 788) defines climate
change as “a statistically significant variation in either the mean state of the climate or in
its variability, persisting for an extended period (typically decades or longer).”
20
This section draws on the work of the IPCC, in particular Chapter 1 of the contribution of Working
Group I to the third IPCC assessment report (Baede et al., 2001), as well as several chapters of the fifth
assessment report (especially Boucher et al., 2013, Cubasch et al., 2013, Masson-Delmotte et al., 2013, and
Myhre et al., 2013).
45
Understanding the interactions of the variety of factors that influence the climate is
complicated. Climate, climate change, and the role of human activities in changing the
climate can only be understood if one has an understanding of the whole climate system.
The following section looks briefly at the components of the planet’s climate system and
shows how they interact.
2.1 The five components of the climate system
The climate system, schematically displayed in Figure 12, is part of the environmental
system discussed in Module 1. It consists of five main components (Baede et al., 2001):
the atmosphere, hydrosphere, cryosphere, land surface, and biosphere (in bold in Figure
12). These five components interact with each other (thin arrows in Figure 12) and are all
affected by the planet’s energy balance, which will be discussed in Section 2.2.
Figure 12: The climate system
Source: Baede et al. (2001: 88).
The most variable and rapidly changing part of the climate system is the atmosphere
(Baede et al., 2001), which marks the boundary of our environmental system as defined
in Module 1. The entire atmosphere lies within 500 km from the surface of the planet,21
with 99 per cent of its total mass within 50 km from the surface (Common and Stagl,
2004). The atmosphere can be subdivided (or stratified) into the five layers displayed in
Figure 13. The troposphere is the lowest layer of the atmosphere and extends up to
approximately 11 km from the planet’s surface. It contains most of the atmosphere’s
21
We exclude here the exosphere and consider that the atmosphere stops at the top of the thermosphere.
46
mass and plays a key role in determining the planet’s climate. Mean temperature in the
troposphere decreases with distance from the surface. In the text that follows, we will
often refer to the so-called surface-troposphere system, which encompasses the planet’s
surface and the troposphere. The next layer is called the stratosphere and extends to
roughly 50 km above the surface. Most of the incoming ultraviolet radiation is absorbed
by ozone that is concentrated in the stratosphere. The boundary between the troposphere
and the stratosphere is called tropopause. The stratosphere is followed by the mesosphere
(50–90 km above the surface) and the thermosphere (90–500 km above the surface).22
Figure 13: Layers of the atmosphere
Source: National Aeronautics and Space Administration (NASA), Climate Science Investigations, available
at: http://www.ces.fau.edu/nasa/module-2/atmosphere/earth.php.
Note: The average temperature varies with altitude and is indicated by the red line.
The atmosphere is mainly a mixture of different gases but also contains some solid and
liquid matter, namely aerosols and clouds. The composition of the atmosphere has been
changing throughout the history of the planet. Today the main bulk of the volume of the
earth’s atmosphere is composed of approximately 78.1 per cent nitrogen (N2), 20.9 per
cent oxygen (O2), and 0.93 per cent argon (Ar). Additionally, the atmosphere contains
22
Many consider the thermosphere as the boundary of the atmosphere. But strictly speaking the
thermosphere is followed by the exosphere (500–10,000 km above surface), which is considered by some
to mark the actual boundary of the atmosphere and thus the environmental system (discussed in Module 1)
with the rest of the universe.
47
several trace gases such as carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O),
and ozone (O3). These trace gases, which constitute less than 0.1 per cent of the
atmosphere’s volume, are called greenhouse gases (GHG). Water vapour (H2O) is also a
greenhouse gas whose volume in the atmosphere is volatile depending on the
hydrosphere’s hydrological cycle. Despite their small share in the total volume of the
atmosphere, greenhouse gases are of crucial importance for the earth’s climate. While the
three main gases (N2, O2, and Ar) do not absorb or reflect the infrared radiation emitted
by the planet, greenhouse gases are different – they absorb infrared radiation coming
from the planet’s surface and emit this radiation towards space and back towards the
earth’s surface, thereby increasing the temperature at the surface (see Section 2.2 for a
detailed discussion). Consequently, greenhouse gases are crucial for our planet’s climate
(Baede et al., 2001).
The hydrosphere consists of all forms of liquid water and includes oceans, rivers, and
lakes. Approximately 70 per cent of the total surface of the planet is covered by water.
Oceans alone store roughly 97 per cent of all forms of water (liquid, solid, and gas)
available on earth, while rivers and lakes store roughly 0.009 per cent (Common and
Stagl, 2004). The basic process that takes place in the hydrosphere is called the
hydrological cycle. The cycle starts with the water evaporating from the oceans, lakes,
and rivers and being released into the atmosphere, which leads to an exchange of heat
between the hydrosphere and the atmosphere. The water then returns from the
atmosphere to the surface in the form of precipitation, either directly into the oceans or
indirectly on the land from where it reaches the oceans through rivers. Water that returns
from land to oceans then in turn influences the composition and circulation of oceans.
Through this cycle, oceans not only exchange water but also heat energy, carbon dioxide,
and aerosols with the atmosphere. As oceans are able to gradually store and release large
quantities of heat, carbon dioxide, and aerosols over a long time period, they act as the
planet’s climate regulator and thus are an important source of long-term natural climate
variability (Baede et al., 2001).
The cryosphere, consisting of solid water, includes continental glaciers, snow fields, sea
ice, permafrost, and the large ice sheets of Antarctica and Greenland. The cryosphere is
relevant for the planet’s climate because of its capacity to store heat, its low capacity to
transfer heat (in other words, its low thermal conductivity), its high reflectivity of
incoming solar radiation, and its influence on ocean circulation and sea levels (Baede et
al., 2001).
The land surface, made up of soils and vegetation, encompasses all parts of the planet that
are not covered by oceans. Land surface matters for the climate because it determines
how energy from solar radiation is returned to the atmosphere. Part of the energy from
the sun is directly returned to the atmosphere as longwave infrared radiation and part of
the energy is used to evaporate water, which then returns as water vapour to the
atmosphere. The texture of the land surface, which depends on the type of soil and/or
vegetation covering the surface, also influences the atmosphere indirectly, as different
textures influence winds in different ways (Baede et al., 2001).
48
The biosphere is composed of all living organisms (also called biota). While it only
represents a very thin layer of the planet (roughly 0.4 per cent of the planet’s radius
(Common and Stagl, 2004), biota plays a key role in influencing the composition of the
atmosphere and thus the climate of the planet. Plants perform photosynthesis, a process
by which they use energy from solar radiation and specific enzymes to transform carbon
dioxide from the atmosphere and water from the hydrosphere into glucose (a
carbohydrate) and oxygen (Figure 14). During this process, they extract large amounts of
carbon from carbon dioxide present in the atmosphere and store it in the form of glucose,
and release oxygen, a by-product of photosynthesis, into the atmosphere. The biosphere
also has an impact on atmospheric concentrations of other greenhouse gases like methane
or nitrous oxide (Baede et al., 2001). Some digestive processes of animal species release
for instance methane as a by-product.
Figure 14: Photosynthesis – the chemical reaction
Source: Author.
The five components of the climate system interact in numerous ways, as schematically
illustrated by the thin arrows in Figure 12. For instance, water vapour is exchanged
between the atmosphere and the hydrosphere, carbon dioxide is constantly extracted from
the atmosphere by plants from the biosphere, and ice sheets from the cryosphere
influence the hydrosphere’s ocean circulations and levels. These are just a few examples
of the physical, chemical, and biological interactions that make the climate system
extremely complex. Some of these processes are still only partly known, and there may
be processes that are still completely unknown (Baede et al., 2001).
All five components of the system and all processes happening within the system use
energy. The balance between energy flowing into the system and energy leaving the
system is called the energy balance. Changes in the energy balance have a profound
impact on all the components and processes of the system. Given the crucial importance
of the energy balance, Section 2.2 familiarizes readers with this term and discusses the
natural greenhouse effect that is an integral part of the balance.
2.2 The earth’s energy balance and the natural greenhouse effect
The planet’s energy balance influences all five components of the climate system that
were introduced in Section 2.1.23 Understanding the energy balance and how it can be
23
In this material, we also use the term “radiative balance” as a synonym for energy balance.
49
altered is therefore of crucial importance for understanding the mechanisms behind
climate change.
Solar radiation is the energy source that powers the entire climate system. Roughly 50 per
cent of solar radiation consists of visible light, with the rest consisting mostly of infrared
and ultraviolet light (Baede et al., 2001). New satellite-based data allow for accurately
quantifying the exchange of radiative energy between the sun, the earth, and space.
However, it is more difficult to quantify energy flows within the climate system because
those flows cannot be directly measured. Consequently, it is not surprising that estimates
of the global energy balance differ considerably (Wild, 2012). IPCC (2013a) updated its
energy balance diagram in the fifth assessment report building on the newly available
estimates of Wild et al. (2013),24 which use satellite and ground-based radiation network
data combined with models from the fifth IPCC assessment report.
Figure 15 provides a schematic illustration of the planet’s energy balance as outlined by
Wild et al. (2013) and used in the fifth IPCC report. Arrows represent radiation flows,
numbers indicate best estimates of the magnitude of these flows, and numbers in
parentheses provide the uncertainty range of these magnitudes, representing present-day
climate conditions at the beginning of the 21st century. Accounting for day and night as
well as for different yearly seasons, an average amount of energy equivalent to 340 watts
per square meter (Wm-2) enters the earth’s atmosphere, and hence the environmental
system, each second.
24
In the third and fourth assessment reports, the IPCC used the energy balance diagram of Kiehl and
Trenberth (1997).
50
Figure 15: The global mean energy balance of the earth
Source: Wild et al. (2013).
Note: TOA: top of the atmosphere. Wm-2: watts per square meter.
Of these 340 Wm-2, roughly 76 Wm-2 are directly reflected back to space by clouds,
atmospheric gases, and aerosols. Another 24 Wm-2 reach the earth’s surface and are
directly reflected back to space by the surface (due to surface reflectivity, technically
referred to as surface albedo). As white light-coloured surfaces reflect more light
compared to dark-coloured surfaces, most of these 24 Wm-2 are reflected back to space
by snow fields, glaciers, ice sheets, and deserts. Of the remaining 240 Wm-2, roughly 79
Wm-2 are absorbed by the atmosphere. This leaves 161 Wm-2 that warm the planet’s land
surface and oceans (see the lower left-hand side of Figure 15). The planet’s land surface
and oceans subsequently return this energy towards the atmosphere and space as sensible
heat, water vapour, and long-wave infrared radiation.
In order to have a stable climate, there needs to be a balance between incoming solar
radiation and outgoing radiation emitted by the earth. Thus, the 240 Wm-2 of incoming
radiation absorbed by the planet’s surface and atmosphere should be returned back to
space. If this does not happen, the radiative balance of the planet is not in equilibrium. If
significantly more radiation were to enter the planet than leave, the planet would become
too hot for life; if significantly more radiation were to leave than enter, the planet would
become too cold for life. Note that our planet’s energy balance is currently not in a
complete equilibrium: Wild et al. (2013) as well as other recent studies (Hansen et al.,
2011; Murphy et al., 2009; Trenberth et al., 2009) find a small positive imbalance of the
earth’s radiative balance. In fact, instead of 240 Wm-2, only 239 Wm-2 leave the planet
(see the upper right-hand side of Figure 15).
51
The earth returns the incoming solar radiation as longwave infrared radiation,25 which is
the heat energy you can feel, for instance, emanating out from a fire. The quantity and
wavelengths of energy radiated by physical objects are specific to the temperature of the
object. Hotter objects radiate more longwave infrared radiation (in terms of magnitude
and energy) than colder objects. In order to radiate 239 Wm-2 in longwave infrared
radiation, the radiating surface should have an average temperature of roughly -19°C
(Baede et al., 2001). However, the average temperature of the earth’s surface is not -19°C
but 14°C. At this temperature, the surface alone radiates on average 397 Wm-2 (Figure
15), which is a considerable higher amount than 239 Wm-2. So how is the planet able to
“only” radiate the 239 Wm-2 required to (almost) maintain its radiative balance and have
at the same time a relatively high average temperature on the surface?
As discussed in Section 2.1, the atmosphere contains several trace gases such as water
vapour (H2O), carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), and ozone
(O3). These gases are able to absorb longwave infrared radiation from the surface of the
planet and from the atmosphere itself.26 Greenhouse gases then emit infrared radiation in
all directions. This means that they emit radiation towards space but also back towards
the surface (see the right-hand side of Figure 15). The downward-directed flux, currently
estimated at 342 Wm-2, heats up the lower layers of the atmosphere and the surface, and
thus maintains the relatively high surface temperature of 14°C. Hence, greenhouse gases
act like a “blanket” that traps heat in the lower layer of the atmosphere.27 This effect,
known as the natural greenhouse effect, results in a net transfer of infrared radiation from
warm areas near the surface to higher levels of the atmosphere (Baede et al., 2001). The
main part28 of the 239 Wm-2 of outgoing longwave radiation that are needed to balance
the incoming solar radiation is subsequently radiated back towards space from relatively
high altitudes and not directly from the surface. These areas in the higher level of the
troposphere are approximately five km above the surface at mid-latitudes and have an
average temperature of roughly -19°C (see the upper right-hand part of Figure 15). Thus,
the natural greenhouse effect is an integral part of the planet’s energy balance system that
is responsible for the relatively warm average temperature on the planet’s surface.
Having covered the components of the climate system as well as the radiative balance
that influences all these components, we can now turn our attention towards climate
change. Section 2.3 provides an overview and a classification of factors that can alter the
climate system.
In the text that follows we use the terms “longwave radiation,” “infrared radiation,” and “longwave
infrared radiation” synonymously.
26
In other words, these greenhouse gases make the atmosphere opaque (i.e. impenetrable) to a lot of the
longwave radiation emitted from the planet’s surface, but not to the incoming shortwave radiation, which
explains why much of this incoming radiation can directly reach the surface.
27
Note that clouds also act as such a “blanket.” At the same time, due to their brightness, they reflect
incoming solar radiation. As a net impact, clouds tend to have a slight cooling effect on the climate system
(Baede et al., 2001).
28
With the exception of a small share of infrared radiation that is directly radiated from the surface through
the so-called atmospheric window towards space.
25
52
2.3 Internally and externally induced climate change
Sections 2.1 and 2.2 showed that the earth’s climate is shaped by factors that are internal
to the climate system (i.e. processes within and between components of the climate
system, such as interactions between the atmosphere and oceans). It is important to
understand that in addition to these internal factors, some external factors are also able to
shape the climate. Why is this so?
Section 2.2 showed that the climate system is in its equilibrium if the net incoming solar
radiation is balanced by the outgoing longwave radiation. Such an equilibrium is marked
by a stable climate (e.g. stable mean temperature, mean precipitation, etc.). If the
radiative balance of the planet changes, the climate is likely to change as well because
through various interactions and feedback mechanisms a change in the radiative balance
affects virtually all components of the climate system. For example, a change of the
radiative balance can affect means or variances of climate variables but also other
statistics such as the occurrence of extreme events (Baede et al., 2001; see Box 10 for a
definition of extreme events). Hence, factors that are external to the climate system but
somehow influence the radiative balance of the planet can also shape the climate.
External factors can be further subdivided into natural external factors and humaninduced (i.e. anthropogenic) external factors. The most obvious example of a natural
external factor is solar activity whose variations result in a changing amount of incoming
solar radiation. Volcanic activity is another example: volcanic eruptions emit aerosol
particles into the atmosphere, which can influence the amount of incoming solar radiation
reflected back towards space. Among the human-induced external factors, human
industrial activity influences greenhouse gas concentrations in the atmosphere and
thereby affects the amount of longwave radiation that is being radiated from earth back
towards space. In short, external factors (e.g. solar activity, volcanic activities, or human
activities) have an influence on so-called climate change drivers (e.g. solar radiation,
aerosol particles, greenhouse gases),29 which in turn affect the radiative balance and
thereby shape the climate (see Figure 18 for a schematic illustration of externally induced
climate changes).
Box 10: Extreme events
Climate change does not only affect the means of climate variables such as temperature or
precipitation, but can also affect the likelihood of the occurrence of extreme weather and climate
events (Cubasch et al., 2013). Examples of such extreme events are droughts, cyclones, or heat
waves. The Intergovernmental Panel on Climate Change (IPCC) defines an extreme weather
event as an event that is “rare at a particular place and/or time of year. Definitions of ‘rare’ vary,
but an extreme weather event would normally be as rare as or rarer than the 10th or 90th
percentile of a probability density function estimated from observations” (Cubasch et al., 2013:
134). Extreme climate events can be defined as extreme weather events that persist for some time
(Cubasch et al., 2013).
Cubasch et al. (2013) show that statistical reasoning can illustrate that increases or decreases of
the frequency of extreme weather events (e.g. an increase of extremely hot days) can result from
29
Drivers of climate change are substances and processes that alter the planet’s energy balance.
53
small changes in the distribution of climate variables (e.g. an increase in the mean temperature).
For example, Figure 16 displays the probability density function of temperature. Note that
temperature is almost normally distributed (other climate variables such as precipitation are not
normally distributed but have skewed distributions). Now suppose that climate change increases
the mean temperature. As a result, the probability density function of temperature shifts to the
right (i.e. average temperature increases) as illustrated by the solid curve in Figure 16. This shift
of the average temperature affects the frequency of extreme events. On the one hand, one
observes more hot extremes; on the other, one observes fewer cold extremes. Changes in the
variance, skewness, or shape of distributions can also affect the frequency of extreme events (see
Cubasch et al., 2013: 134–35, for a more detailed discussion).
Figure 16: Extreme events - schematic presentation
Source: Author's elaboration based on Cubasch et al. (2013:134).
The climate of our planet is thus shaped by factors that are internal to the climate system
and by factors that are external to the climate system (Baede et al., 2001). This implies
that climate change, which is defined as a persistent variation in either the mean state of
the climate or in its variability (see the introduction to Section 2), can be induced
internally or externally.30 The El Niño-Southern Oscillation (ENSO), described in Box
11, is an example of an internally induced climate variability (Baede et al., 2001). The
ENSO is the result of an interaction between the atmosphere and the Pacific ocean, and
affects different climate variables such as precipitation and temperature in many parts of
the world. As this teaching material focuses on human-induced (i.e. externally induced)
climate change, we will not address internally induced climate variability further,31 and
limit ourselves to a discussion of externally induced climate change. To do so, Section
2.4 will introduce two concepts – radiative forcing and effective radiative forcing – that
allow for measuring the influence of natural and human-induced external factors on
climate change.32
30
Internally induced climate change means that factors internal to the climate system affect the mean or the
variability of the climate, while externally induced climate change means that factors external to the climate
system affect the mean or the variability of the climate.
31
See the IPCC reports listed in Annex 2 for additional readings on internally induced climate variability.
32
Note that these concepts can a priori also be used to measure the influence of most internal factors on
climate change.
54
Box 11: The El Niño-Southern Oscillation – an example of an internal interaction among
components of the climate system affecting the means and variability of different climate
variables
El Niño-Southern Oscillation (ENSO) events are naturally occurring phenomena that result from
an interaction between the atmosphere and the hydrosphere. According to the Intergovernmental
Panel on Climate Change (IPCC), “El Niño involves warming of tropical Pacific surface waters
from near the International Date Line to the west coast of South America, weakening the usually
strong sea surface temperature (SST) gradient across the equatorial Pacific, with associated
changes in ocean circulation. Its closely linked atmospheric counterpart, the Southern Oscillation
(SO), involves changes in trade winds, tropical circulation and precipitation” (Trenberth et al.,
2007: 287). Historically, the ENSO alternates between two states: El Niño and La Niña, each of
which has specific regional impacts on climate variables such as temperature or precipitation
(Figure 17). For example, surface temperature is above average during El Niño events and below
average during La Niña events in the eastern tropical Pacific region. El Niño events occur every 3
to 7 years and alternate with their counterpart La Niña (Trenberth et al., 2007).1
Figure 17: Correlations of surface temperature, precipitation and mean sea level pressure
with the Southern Oscillation Index
Source: Trenberth et al. (2007).
Note: Correlations with the Southern Oscillation Index (SOI), based on standardized Tahiti minus Darwin
sea level air pressure, for annual (May to April) means of sea level air pressure (top left), surface
temperature (top right) for 1958 to 2004, and precipitation for 1979 to 2003 (bottom left). In the SOI graph
(bottom right), red (blue) values indicate El Niño (La Niña) conditions. The graph shows the long-term
periodic fluctuation between these conditions since 1850.
The ENSO influences regional climate patterns in several parts of the globe and has a global
impact on climate variables such as surface temperature and precipitation. Figure 17 illustrates
these effects based on annual mean correlations between the climate variables and the Southern
Oscillation Index (SOI). The SOI (bottom right panel) uses observed atmospheric pressure at sea
55
level to infer the presence of El Niño and La Niña events. It is calculated as the standardized air
pressure in Tahiti (Eastern Pacific) minus the standardized air pressure in Darwin, Australia
(Western Pacific). Positive values (higher pressure in Tahiti) indicate a La Niña event; negative
values (higher pressure in Darwin) indicate an El Niño event. The bottom left panel of Figure 17
shows the correlation between SOI and precipitation: one observes, for example, a strong positive
correlation among the variables over the Western Pacific, indicating that this region experiences
above (below) average precipitation during La Niña (El Niño) events. The upper right panel
displays the correlation between SOI and surface temperature. One observes, for example, that
there is a strong negative correlation between the two variables over the eastern tropical Pacific
region, indicating that in this region, surface temperature is above (below) average during El Niño
(La Niña) events. Thus, as shown in Figure 17, internal interactions such as the El Niño-Southern
Oscillation can have significant effects on the climate.
Source: Author's elaboration based on Trenberth et al. (2007).
1
An intuitive explanation of the El Niño-Southern Oscillation can be found in geoscientist Keith Meldahl’s
video (available at https://www.youtube.com/watch?v=GTgz6ie2eSY). A more extensive explanation of
the phenomenon is provided by a Yale University open course given by Ronald Smith, a professor of
geoscience, geophysics and mechanical engineering, available at https://www.youtube.com/watch?v=bKn0CeFWtk.
2.4 Measuring the importance of factors driving climate change: radiative
forcing and effective radiative forcing
In theory, there are many ways to assess how strongly different external factors change
the climate. One could, for instance, try to directly measure the effect of a change in a
single climate change driver (e.g. greenhouse gases) that has been induced by an external
factor (e.g. human industrial activity) on different climate variables (e.g. temperature and
precipitation). However, identifying and isolating such effects is extremely difficult.
Climate scientists therefore rely on intermediate measures that quantify the influence of
external factors on climate change indirectly. The basic idea behind these measures is
rather intuitive. First, climate scientists measure how strongly the radiative balance is
affected by an externally induced variation in a climate change driver. Then they estimate
how this change of the radiative balance affects the climate (Figure 18). To do so, they
use a measure called “radiative forcing.”
Radiative forcing is the most commonly used indicator that allows for capturing how
externally induced changes in climate change drivers affect the radiative balance and
subsequently change the climate (Myhre et al. 2013). The term “forcing” indicates that
the radiative balance of the earth is forced away from its equilibrium state by an
externally induced variation in a climate change driver (Perman et al., 2011).33
Intuitively, radiative forcing measures the radiative imbalance that occurs from an
externally induced change in a climate change driver. IPCC (2001a: 795) defines
radiative forcing as “the change in the net vertical irradiance (expressed in Watts per
Climate change drivers are subsequently also called “forcing agents,” while externally induced variations
in climate change drivers are also called “external forcings” or sometimes simply “forcings.” For the sake
of simplicity, we do not adopt this terminology in this teaching material and continue to use the terms
“climate change driver” and “externally induced variations in climate change drivers.” However, the terms
“external forcings,” “forcings,” and “forcing agents” do appear in direct citations from the IPCC.
33
56
square metre: Wm−2) at the tropopause34 due to… a change in the external forcing
of the climate system, such as, for example, a change in the concentration of carbon
dioxide or the output of the Sun.”
While the radiative forcing concept is the most widely used measure to assess and
compare the size of the radiative imbalance created by externally induced variations in
climate change drivers, it has some weaknesses. The main one is that the concept keeps
all surface and tropospheric properties fixed and does not allow them to respond to the
changes induced by the variations in climate change drivers. In the fifth assessment
report, the IPCC therefore introduced a new, complementary concept called “effective
radiative forcing.” Radiative forcing and effective radiative forcing are very similar, with
the exception that effective radiative forcing allows some surface and tropospheric
properties to respond to perturbations in the short term. Effective radiative forcing is
defined as “the change in [the] net top of atmosphere downward radiative flux after
allowing for atmospheric temperatures, water vapour and clouds to adjust, but with
surface temperature or a portion of surface conditions unchanged.… Hence effective
radiative forcing includes both the effects of the forcing agent itself and the rapid
adjustments to that agent (as does radiative forcing, though stratospheric temperature is
the only adjustment for the latter)” (Myhre et al., 2013: 665). Due to the inclusion of
short-term adjustments of some surface and tropospheric properties, the effective
radiative forcing concept is believed to be a better indicator of potential temperature
responses (Myhre et al., 2013).
Radiative forcing (and effective radiative forcing) can be negative or positive. Positive
radiative forcing implies that incoming radiation is larger than outgoing radiation, leading
to an energy increase in the environmental system (i.e. a positive energy imbalance). To
rebalance the system, temperatures in the surface-troposphere system have to increase.
Negative radiative forcing implies that incoming radiation is smaller than outgoing
radiation (i.e. a negative imbalance), leading to an energy decrease in the environmental
system. To rebalance the system, temperatures in the surface-troposphere system have to
decrease.
34
The tropopause is the boundary between the troposphere and the stratosphere (see Figure 13). For
practical reasons, the tropopause is defined as the top of the atmosphere (see Ramaswamy et al., 2001, for a
detailed discussion).
57
Figure 18: Externally induced climate changes
Source: Author's elaboration based on Forster et al. (2007: 134).
Note: Non-initial radiative forcing effects have been omitted from the figure as they are not addressed by
this teaching material. Feedback effects are discussed in Section 2.5.
Before we focus entirely on human influences on the climate in Section 2.5, let us briefly
illustrate the radiative forcing concept by listing some examples of radiative forcing
induced by certain natural external factors. First, let us consider solar activity. Solar
activity, and hence solar output, is not constant but fluctuates over time at various time
scales, including centennial and millennial scales (Myhre et al., 2013). Solar output has
increased gradually during the industrial era (from 1750 up until today), which has led to
an increase in incoming solar radiation and thereby caused a small amount of positive
radiative forcing. This has had a small warming effect on the surface-troposphere system
(Forster et al., 2007). Another natural external factor, the astronomical alignment of the
sun and the earth, also varies and induces cyclical changes in radiative forcing. However,
these changes are only substantial over very long time horizons, and partially explain, for
instance, different climatic periods such as ice ages (Myhre et al., 2013). Volcanic
eruptions are another natural external factor that can, over a short period of time lasting
from several months up to a year, increase the concentration of sulphate aerosol particles
in the stratosphere that block parts of incoming solar radiation, inducing short-term
negative radiative forcing, which tends to have a cooling effect on the surfacetroposphere system (Forster et al., 2007).
While such natural external factors are important, they have played a relatively small role
during the industrial era. In its fifth assessment report, the IPCC states that “there is a
very high confidence that industrial-era natural forcing is a small fraction of the
anthropogenic forcing except for brief periods following large volcanic eruptions. In
particular, robust evidence from satellite observations of the solar irradiance and volcanic
aerosols demonstrates a near-zero (-0.1 to +0.1) Wm-2 change in the natural forcing
compared to the anthropogenic effective radiative forcing increase of 1.0 (0.7 to 1.3)
Wm-2 from 1980 to 2011. The natural forcing over the last 15 years has likely offset a
substantial fraction (at least 30 per cent) of the anthropogenic forcing” (Myhre et al.,
2013: 662).
58
2.5 Human-induced climate change
After outlining basic mechanisms that influence the climate system and introducing
concepts that allow for measuring climate change, we are now equipped with the
necessary tools to analyse human-induced climate change.
Human activities cause changes in the amounts of greenhouse gases, aerosols, and clouds
in the earth’s atmosphere. These human-induced changes in climate change drivers
influence the planet’s radiative balance and hence the climate system. Human activities
also change the land surface of the planet, which can effect, for instance, surface
reflectivity (albedo), influencing the radiative balance and thus also the climate system
(IPCC 2001b, 2007, 2013a). The sections that follow first discuss these human-induced
changes in climate change drivers separately, then assess their respective impact on the
radiative balance, and finally take a look at feedback effects that can amplify or reduce
the impacts of the different climate change drivers.
2.5.1 Human-induced greenhouse gas emissions and the enhanced greenhouse effect
The main greenhouse gases emitted by human activities are carbon dioxide (CO2),
methane (CH4), nitrous oxide (N2O), and halocarbons (Forster et al., 2007). Carbon
dioxide, methane, nitrous oxide, and many halocarbons are called “well-mixed
greenhouse gases” because they mix sufficiently in the troposphere for reliable
concentration measurements to be made from only a few remote observations (Myhre et
al., 2013).
Anthropogenic carbon dioxide emissions mainly result from human use of fossil fuels in
sectors such as transportation, energy, and cement production. Anthropogenic methane is
mostly emitted during agricultural activities and natural gas distribution, and from
landfills. Anthropogenic nitrous oxide emissions stem mostly from burning of fossil fuels
and the use of fertilizers in agricultural soil. Anthropogenic halocarbons, which include
chlorofluorocarbons (see also Box 14 in Module 3), are emitted by diverse industrial
activities and have in the past also been released by refrigeration processes. In addition to
the four main greenhouse gases, human activities also emit other pollutants such as
carbon monoxide (CO), volatile organic compounds, nitrogen oxides (NOx), and sulphur
dioxide (SO2). While these gases are negligible greenhouse gases, they indirectly
influence concentrations of other greenhouse gases such as methane or ozone through
chemical reactions (Cubasch et al., 2013).
So how strongly have humans influenced atmospheric concentrations of greenhouse
gases?
Instrumental measurements provide accurate atmospheric GHG concentrations back to
1950, while indirect measures are used for dates prior to 1950: ice core data allow for
analysing air bubbles enclosed in ice and thus provide an indirect record of past
atmospheric concentrations of well-mixed greenhouse gases (Masson-Delmotte et al.,
2013). Figure 19 combines data based on instrumental measurements and ice core data to
show that while atmospheric carbon dioxide, methane, and nitrous oxide concentrations
59
were fairly stable for more than a thousand years before the industrial revolution (starting
around 1750), they have increased rapidly since then.
Figure 19: Atmospheric carbon dioxide, methane and nitrous oxide concentrations
from year 0 to 2005
Source: Forster et al. (2007: 135).
Note: Concentration units are parts per million (ppm) or parts per billion (ppb), indicating the number of
molecules of the greenhouse gas per million or billion air molecules, respectively, in an atmospheric
sample.
Ice core data also allow us to look further back in history and track atmospheric GHG
concentrations dating several hundreds of thousands of years. The fifth IPCC assessment
report provides information covering the past 800,000 years (IPCC, 2013a). Data indicate
that pre-industrial ice core GHG concentrations stayed within natural limits. For carbon
dioxide, maximum concentrations of 300 parts per million (ppm) and minimum
concentrations of 180 ppm have been found. For methane, data indicate maximum
concentrations of 800 parts per billion (ppb) and minimum concentrations of 350 ppb.
And for nitrous oxide, ice core data show maximum concentrations of 300 ppb and
minimum concentrations of 200 ppb (Masson-Delmotte et al., 2013). The fifth IPCC
assessment report reaches the conclusion that “it is a fact that present-day (2011)
concentrations of CO2 (390.5 ppm), CH4 (1803 ppb) and N2O (324 ppm) exceed the
range of concentrations recorded in the ice core records during the past 800 ka.35 With
very high confidence, the rate of change of the observed anthropogenic WMGHG [wellmixed greenhouse gases] rise and its RF [radiative forcing] is unprecedented with respect
to the highest resolution ice core record back to 22 ka for CO2, CH4 and N2O, accounting
for the smoothing due to ice core enclosure processes. There is medium confidence that
the rate of change of the observed anthropogenic WMGHG rise is also unprecedented
with respect to the lower resolution records of the past 800 ka” (Masson-Delmotte et al.,
2013: 391).
In short, given the data and the methods to determine the origin of GHG emissions,36 the
scientific community has reached a consensus: humans have substantially altered the
35
ka is a unit of time indicating a thousand years.
For instance, one can analyse the changing isotopic composition of atmospheric CO2. By doing so, it can
be shown that the observed increase in the atmospheric CO2 concentration is of anthropogenic origin,
36
60
composition of the atmosphere and continue to do so. Since the industrial revolution,
anthropogenic GHG emissions have substantially increased GHG concentrations in the
atmosphere (IPCC 2001b, 2007, 2013a). The rate of this increase is unprecedented over
the last 800,000 years, and the concentrations are currently higher than all concentrations
recorded in ice cores over those years (IPCC, 2013a). So how does this human-induced
change of the atmosphere affect the climate?
The answer is relatively straightforward: the anthropogenic increase in GHG
concentrations amplifies the natural greenhouse effect (see Section 2.2). Increased GHG
concentrations lead to an increased rate of absorption and subsequent emissions of
infrared radiation coming from the surface. In other words, increased GHG
concentrations increase the atmosphere’s opacity to longwave radiation, but more so at
lower altitudes where air density and GHG concentrations are higher than at higher
altitudes where they are both lower. Thus the share of upward flux longwave radiation
leaving the atmosphere from higher, relative to lower altitudes, increases. As a result of
the increase in GHG concentrations, the altitudes from where earth’s radiation is emitted
towards space thus become higher. At these higher altitudes, the troposphere is colder
(Figure 13) and therefore, less energy is emitted towards space, causing a positive
radiative forcing (Baede et al., 2001). This effect is called the enhanced greenhouse
effect.
2.5.2 Human-induced changes in aerosols
In addition to increasing concentrations of greenhouse gases, human activities also
increase the amount of aerosols in the atmosphere (IPCC 2001b, 2007, and 2013a).
Atmospheric aerosols are generated either by emissions of primary particulate matter or
by formation of secondary particulate matter from gaseous precursors. These consist
mainly of inorganic compounds (e.g. sulphate, nitrate, ammonium, sea salt), organic
matter, black carbon, mineral compounds (e.g. desert dust) and primary biological aerosol
particles (Boucher et al., 2013). Humans emit aerosols through diverse industrial, energyrelated, and land-use activities (Baede et al., 2001). Unlike greenhouse gases, aerosols
remain in the atmosphere only for a short time as they are washed out by rain. Figure 20
provides an overview of key properties of the main aerosols in the troposphere, including
their main sources and their most important climate-relevant properties.
because the changing isotopic composition of atmospheric CO2 betrays the fossil origin of the increase
(Baede et al., 2001).
61
Figure 20: Key properties of main aerosols in the troposphere
Source: Boucher et al. (2013: 597).
Note: CNN: cloud condensation nuclei; IN: ice nuclei; OA: organic aerosols; POA: primary organic
aerosols; SOA: secondary organic aerosols; UV: ultraviolet.
The effects of increased amounts of aerosols on radiative forcing and hence the climate
are still not fully known (Baede et al., 2001). As the fifth IPCC assessment report states,
aerosols and clouds are still contributing the largest uncertainty to estimates of the
62
changing energy balance of the planet (Boucher et al., 2013). Two main channels of
effect have been identified. First, some aerosols (Figure 20) absorb and scatter incoming
solar radiation, and to some extent also longwave radiation from the surface. They thus
directly affect the radiative balance of the planet by sending parts of the incoming
radiation back into space (hence one speaks of the “direct effect”). This direct effect
causes a negative radiative forcing (Baede et al., 2001; Boucher et al., 2013). Second,
aerosols interact with clouds. According to Boucher et al. (2013), some aerosols (Figure
20) act as cloud condensation nuclei (CCN) and ice nuclei (IN), around which cloud
droplets and ice crystals form. Aerosols thus play an important role in cloud formation
and are often described as “cloud seeds.” By interacting with clouds, aerosols indirectly
influence cloud albedo and the lifetimes of clouds (this effect is often referred to as the
“indirect effect”). This has an impact on the reflection of incoming shortwave radiation as
well as the absorption of outgoing longwave radiation by the atmosphere and thus on the
planet’s energy balance.
2.5.3 Human-induced land-use change
Humans not only affect the climate by emitting GHG emissions and aerosols, but also by
changing the surface characteristics of land. This so-called “land-use change” can result
from various human activities including agriculture, irrigation, deforestation,
urbanization, and traffic, and it influences physical and/or biological properties of the
land surface (Baede et al., 2001). Hurtt et al. (2006) estimate that between 42 and 68 per
cent of the total land surface was affected by human activities over the 1700–2000 period.
By changing the land surface, humans directly and indirectly alter the planet’s energy
balance, water cycles, carbon cycles, and heat fluxes (Cubasch et al., 2013; Myhre et al.,
2013). Changes in physical properties of the land surface can influence the reflectivity of
the surface (land albedo) and hence affect the amount of incoming solar radiation
reflected towards space, thereby directly influencing the energy balance of the planet
(Baede et al., 2001). Irrigation and other water-intensive activities can affect the water
cycle and thus indirectly affect the energy balance. Changes in biological properties can
also have important consequences, especially in terms of GHG emissions. If humans
convert forests into cultivable land, they cut or burn the existing forests. They thereby
destroy carbon sinks, which reduces carbon storage, releases carbon dioxide into the
atmosphere, and changes surface albedo (Cubasch et al., 2013). The combined effect of
these physical and biological changes is complex and difficult to assess.
2.5.4 Human-induced radiative forcing
Sections 2.5.1 to 2.5.3 explained different ways in which humans influence climate
change drivers, and thereby affect the radiative balance of the earth and thus change the
climate. This leads to the question of the importance of the different effects as well as the
overall effect humans have on the climate. The previously introduced radiative forcing
and effective radiative forcing concepts are very useful to this end. IPCC (1996, 2001,
2007, 2013a) estimated mean radiative forcing of each of the climate change drivers that
are affected by humans. As knowledge about the climate system and climate change is
63
constantly growing, these numbers have been revised several times. Figure A1 in Annex
1 provides an overview of the estimates of the different assessment reports.
Figure 21 displays the IPCC (2013a) estimates of radiative forcing (hatched) and
effective radiative forcing (solid) of different climate change drivers over the 1750–2011
period. Uncertainties are displayed by dotted and solid lines that indicate 5 to 95 per cent
confidence intervals. They vary widely depending on the climate change driver under
consideration. The largest positive radiative forcings are due to carbon dioxide and other
well-mixed greenhouse gases. Together, their human-induced concentration increases are
responsible for an estimated radiative forcing of 2.83 Wm-2. Human-induced increases of
tropospheric ozone (where ozone acts as a greenhouse gas) caused an effective radiative
forcing of 0.4 Wm-2. Ozone increases in the stratosphere (where they block parts of
incoming solar radiation) caused a negative radiative forcing of -0.1 Wm-2. Changes in
surface albedo caused by land-use change are estimated to cause negative effective
radiative forcing (-0.15 Wm-2), while black carbon particles on snow and ice cause
positive effective radiative forcing (0.04 Wm-2). The greatest uncertainty is associated
with aerosols. While both aerosol-radiation and aerosol-cloud interactions are estimated
to cause negative radiative forcing, confidence intervals are very large.
Figure 21: Radiative forcing of the climate between 1750 and 2011
Source: Myhre et al. (2013: 697).
Note: This figure is a bar chart for radiative forcing (hatched) and effective radiative forcing (solid) for the
period 1750–2011. Uncertainties (5 to 95 per cent confidence range) are given for radiative forcing (dotted
lines) and effective radiative forcing (solid lines). WMGHG: well-mixed greenhouse gases.
According to IPCC findings, total anthropogenic radiative forcing is virtually certain to
be positive and is estimated to be 2.3 Wm-2. The probability of negative anthropogenic
radiative forcing is estimated to be smaller than 0.1 per cent and thus exceptionally
unlikely (Myhre et al., 2013). This confirms that human activities contribute to increase
the level of solar energy stored in the climate system, which results in the warming of the
64
lower atmosphere and the earth’s surface. Compared to the human influence, natural
factors had only a low influence on the planet’s energy balance and thus on climate
change (Figure 21). The IPCC (Myhre et al., 2013: 661) therefore concludes that “it is
unequivocal that anthropogenic increases in the well-mixed greenhouse gases [WMGHG]
have substantially enhanced the greenhouse effect, and the resulting forcing continues to
increase. Aerosols partially offset the forcing of the WMGHGs and dominate the
uncertainty associated with the total anthropogenic driving of climate change.”
2.5.5 Feedback effects and non-linearity
While we have seen that anthropogenic radiative forcing has been positive and has
increased during the industrial era, hence warming the climate system, we have not yet
discussed what specific impacts this radiative forcing has had on the climate system (see
Section 3 for a discussion of observed impacts, and Section 4 for a discussion of
anticipated impacts). A priori we know that positive (negative) radiative forcing increases
(decreases) the temperature of the climate system.
However, this relationship is complicated because of the existence of so-called feedback
effects that can either amplify or diminish the effect that specific radiative forcings have
on variables such as temperature or precipitation (IPCC 2001, 2007, 2013a). Feedback
effects that amplify the effect of driver-induced radiative forcing are called “positive
feedbacks,” while those that reduce the effect of driver-induced radiative forcing are
called “negative feedbacks” (Le Treut et al., 2007). Due to these internal feedback
mechanisms, climate variables like temperature will in general not react in a linear way to
changes in climate change drivers (IPCC 2001, 2007, 2013a).37 While a detailed
discussion of all known feedback mechanisms is out of the scope of this teaching
material, Figure 22 provides some examples that have been listed in the first chapter of
the fifth IPCC report (Cubasch et al., 2013).
Figure 22 schematically displays some key feedback mechanisms related to increases in
carbon dioxide concentrations that, as we have seen, cause positive radiative forcing and
hence tend to warm the surface-troposphere system, all else being equal. Some of these
feedback mechanisms are positive (i.e. they additionally warm the system), some are
negative (i.e. they cool the system), and some can be positive or negative. For example,
the so-called water-vapour feedback is a positive feedback: as the planet gets warmer due
to the stronger greenhouse gas effect, more water evaporates. This leads to an increase of
water vapour in the atmosphere, where it acts as a strong greenhouse gas, further
enhancing the greenhouse effect. Snow and ice albedo are at the root of another positive
feedback mechanism: increasing temperatures cause additional snow and ice to melt,
which changes the size of the land surface. White reflecting surfaces are transformed into
darker, absorbing surfaces. This lowers the planet’s reflectivity of incoming solar
radiation and thus increases the amount of energy the planet’s surface absorbs, leading to
additional warming. If the average temperature of the planet increases, the planet starts to
radiate more infrared radiation, sending more energy back towards the atmosphere and
ultimately space. This infrared radiation mechanism is an example of a negative feedback
37
The example in the next paragraph illustrates this non-linearity.
65
mechanism, which is however almost negligible for a small temperature change of the
order of 1-4°C.
Figure 22: Positive and negative feedback mechanisms
Source: Cubasch et al. (2013:128).
Note: Climate feedbacks related to increasing CO2 and rising temperature include negative feedbacks (-),
positive feedbacks (+) and positive or negative feedbacks (±). The smaller box highlights the large
difference in timescales for the various feedbacks. GHG: greenhouse gas.
Short summary
Section 2 reviewed basic elements of climate science. It showed that the climate system
has five components: atmosphere, hydrosphere, cryosphere, land surface, and biosphere.
These components interact in complex ways. All of them depend on the planet’s energy
balance (or radiative balance), i.e. the difference between the energy that flows into the
climate system and the energy that flows out of the climate system. For a stable climate,
incoming energy should equal outgoing energy. Section 2 explained that climate can
change due to variations in factors that are either internal or external to the climate
system. External factors such as human activities can affect the climate by influencing
climate change drivers such as atmospheric greenhouse gas concentrations. Externally
induced changes in climate change drivers then change the energy balance of the planet
and thus affect the climate. The radiative forcing concept introduced in Section 2
measures how strongly external factors influence the radiative balance. Positive radiative
66
forcing indicates that more energy is flowing into the system than out of the system,
leading to higher temperatures. Negative radiative forcing indicates that less energy is
flowing into the system than out of the system, leading to lower temperatures. Section 2
then showed that human activities influence several climate change drivers (atmospheric
greenhouse gases, aerosol concentrations, reflectivity of the planet’s surface) and have
thereby caused overall positive radiative forcing since the start of the industrial era,
which is heating up the planet. Section 2 concluded by highlighting the importance of
feedback mechanisms that can amplify or diminish the effects that positive or negative
radiative forcing has on different climate variables such as temperature or precipitation.
3 Observed changes in the climate system
Section 2 introduced the climate system and explained how it can be affected by human
activities. This section provides a short summary of human-induced changes of the
climate system that have been observed and assessed by the scientific community. To do
so, we focus on observed changes as reported in the fifth IPCC assessment report.
IPCC reports observed changes in the climate based on research undertaken using a
variety of different instrumental measurements. These measurements are either on-site
measurements, which measure variables such as temperature or precipitation directly on
site, or off-site measurements. Off-site measurements (also called “remote sensing”)
gather data from so-called remote sensing platforms,38 which measure climate variables
from a certain distance (i.e. “off-site”). Instrumental data are complemented with
paleoclimate reconstructions (e.g. the ice core data used to analyse carbon dioxide
concentrations mentioned in Section 2.5), which allow for extending the covered period
considerably.
Based on an extensive review of available research, IPCC (2013b: 4) finds that the
“warming of the climate system is unequivocal, and since the 1950s, many of the
observed changes are unprecedented over decades to millennia. The atmosphere and
ocean have warmed, the amounts of snow and ice have diminished, sea level has risen,
and the concentrations of greenhouse gases have increased.” In the text that follows, we
take a closer look at the most important climate perturbations and responses (see the
scheme in Figure 18) – namely, observed changes in mean temperature, precipitation, ice
cover, sea levels, and occurrence of extreme events. The section concludes with a short
overview of the impacts of these observed changes on human and natural systems.
3.1 Observed changes in temperature
Climate change has resulted in an increase in the annual average global surface
temperature. Data from several independent datasets concur that the combined mean land
and ocean surface temperatures increased by 0.85°C over the 1880–2012 period. The
magnitude of this increase lies with 90 per cent probability within the range of 0.65°C to
1.06°C. Over the 1951–2012 period, this increase is estimated to have been 0.72 °C, lying
with 90 per cent probability within the range of 0.49°C to 0.89°C (Hartmann et al., 2013).
38
Satellites are examples of remote sensing platforms.
67
The longest available dataset with data sufficiently detailed to allow for calculating
regional trends shows that almost the entire planet experienced an increase of average
surface temperature over the 1901–2012 period, as illustrated by Figure 23. It is
important to note that while the long-term warming trend is highly robust, short-term
temperature trends vary due to natural variability and are thus sensitive to the start and
end year of the period under observation. This leads to a high inter-annual and decadal
variation in short-term warming trends (Hartmann et al., 2013). As an example, the IPCC
states that the rate of warming from 1998 to 2012 was slower than the overall rate from
1951 to 2012. The IPCC explains this lower rate by the fact that the 1998–2012 period
started with a strong El Niño event, which led to a relatively high mean temperature in
the first year of the measurement period.
Figure 23: Observed surface temperature changes from 1901 to 2012
Source: IPCC (2013b: 6) based on Hartmann et al. (2013).
Mean temperature increases can be observed not only at the planet’s surface, but also in
oceans and in the troposphere. The IPCC states that it is virtually certain that the average
temperature in the upper ocean (0 to 700 meters below the surface) increased over the
1971–2012 period, and it is likely that the temperature also increased during the 1870–
1971 period (Rhein et al., 2013). Ocean temperature also likely increased in lower ocean
segments during the 1975–2009 period (700 to 2,000 meters below the surface) and the
1992–2005 period (3,000 meters below the surface to bottom of the ocean). Rhein et al.
(2013) estimate that oceans have absorbed by far the most of the climate system’s energy
increase (i.e. 93 per cent of the energy accumulated during 1971–2010). It is also
virtually certain that the troposphere has warmed since the mid-20th century (Hartmann
et al., 2013).
The IPCC states that it is extremely likely that human-induced changes in the climate
system have been the dominant cause of these observed temperature increases. More
specifically, it states that it is extremely likely that more than 50 per cent of the increase
68
in mean surface temperature is attributable to human-induced increases in GHG
concentrations. Human-induced variations in atmospheric GHG concentrations alone
likely increased average surface temperature by 0.5°C to 1.3°C between 1951 and 2010.
The joint effect of other human-induced variations in climate change drivers (e.g.
aerosols) on mean surface temperature is likely to be within -0.6°C to 0.1°C, while
naturally induced changes only account for -0.1°C to 0.1°C. Human activities have also
very likely substantially contributed to the observed warming of upper ocean levels and
the troposphere (Bindoff et al., 2013).
3.2 Observed changes in precipitation
Climate change also affects precipitation levels. While the fourth IPCC assessment report
concluded that global precipitation increased north of 30°N between 1900 and 2005 and
has decreased in the tropics since 1971, the fifth assessment report relativizes these
findings (Hartmann et al., 2013). It states that even if all available long-term datasets
point towards an increase in global mean precipitation over the 1901–2008 period, the
order of magnitude of this increase varies widely depending on the data source. The IPCC
thus attributes only a low confidence level to evidence indicating global precipitation
increases over land surfaces prior to 1950 and a medium confidence level to evidence
indicating precipitation increases after 1950.
Changes in precipitation seem to differ widely by region. Figure 24 displays precipitation
increases in middle and higher latitudes in the northern and southern hemisphere.
Hartmann et al. (2013) have medium confidence in the evidence suggesting that
precipitation increased in mid-latitudes of the northern hemisphere for the period before
1950, and high confidence for the period after 1950. For all other regions, however,
confidence in the evidence of precipitation increases is low due to data quality.
Figure 24: Observed change in annual precipitation over the land surface
Source: IPCC (2013b: 8) based on Hartmann et al. (2013).
Note: mm yr-1: millimeters per year.
Despite this uncertainty, the IPCC has medium confidence that human actions have
contributed to these observed changes in mean precipitation since 1950. The IPCC also
69
has medium confidence that since 1973, human activities have affected atmospheric
humidity, which is another important variable of the hydrological cycle (Bindoff et al.,
2013).
3.3 Observed changes in ice and snow cover
Ice and snow cover are also affected by climate change. The IPCC reports a very likely
decrease of 3.5 to 4.1 per cent per decade in the extent of annual Arctic sea ice over the
1979–2012 period (in fact, the extent decreased in every season and in every decade after
1979 – see panel b in Figure 25). There is also high confidence that perennial and multiyear Arctic sea ice shrank over the same period and that the average winter sea ice
thickness decreased between 1980 and 2012. Both the Greenland and Antarctic ice sheets
have also been losing mass at an accelerated speed (Vaughan et al., 2013). Finally, there
is very high confidence that the average snow cover in the northern hemisphere has
decreased since the 1950s (see panel a in Figure 25).
These observed changes in the cryosphere are also at least partly human-driven. The
IPCC estimates that it is (a) very likely that human actions have contributed to the sea ice
loss in the Arctic since 1979; (b) likely that human actions have contributed to the ice
surface melting of Greenland since 1993; (c) likely that human actions have contributed
to the retreat of glaciers since the 1960s; and (d) likely that human actions have
contributed to the snow cover reductions in the northern hemisphere (Bindoff et al.,
2013).
3.4 Observed changes in sea levels
Global sea levels also respond to variation in climate change drivers. The IPCC has high
confidence in findings suggesting that the rate of sea level rise since the 1850s is higher
than the mean increase over the last 2,000 years. As panel c of Figure 25 shows, global
sea levels increased by almost 0.2 meters over the 1901 to 2010 period. Most of these
increases since the 1970s are attributable to the aforementioned reductions in glaciers and
to the thermal expansion of oceans, which are due to the mean temperature increases in
oceans (Rhein et al., 2013). Hence, the IPCC concludes that it is very likely that humans
have also substantially contributed to the observed increase of sea levels since the 1970s
(Bindoff et al., 2013).
70
Figure 25: Selected observed changes in snow cover, ice extent and sea level
Source: IPCC (2013b: 10).
3.5 Observed changes in extreme events
Besides affecting mean states of variables such as temperature, precipitation, snow cover,
ice cover, and sea levels, variations in climate change drivers have also altered the
probability of occurrence of extreme events (see Box 10 for a definition of extreme
events), such as droughts, floods, heat waves, cyclones, or wildfires. Table 4 summarizes
the IPCC’s evidence of changes in the occurrence of these types of extreme events, and
provides an overview of its assessment of the role of humans in driving these observed
changes.
71
Table 4: Changes in the probability of occurrence of extreme events
Source: Author's elaboration based on IPCC (2013b: 7).
3.6 Impacts on natural and human systems
Sections 3.1–3.5 showed that human-induced variations in climate change drivers such as
atmospheric greenhouse gas or aerosol concentrations have already affected the climate
system to a considerable extent. These observed changes in temperature, precipitation, ice
and snow, sea levels, and occurrence of extreme events have in turn had several impacts
on natural and human systems on a global scale (see Figure 26 for a schematic overview).
The IPCC’s evidence is the strongest for climate change impacts on natural systems
(Field et al., 2014). The IPCC has medium confidence that the observed changes in
precipitation and snow and ice cover have affected hydrological systems and the quantity
and quality of water resources. There is high confidence that climate change has affected
various terrestrial, freshwater, and marine species. Some of these species have changed
their geographical range, seasonal activities, and migration routes in response to a
changing climate (Field et al., 2014). The population sizes of species have also been
affected by climate change: the IPCC reports, for instance, that climate change has
increased tree mortality in some regions and contributed to the extinction of some animal
species in Central America (Field et al., 2014).
72
Figure 26: Observed impacts attributed to climate change
Source: Field et al. (2014: 42).
While evidence is the strongest for natural systems, some impacts on humans that are
attributable to a changing climate have also been reported. Overall, crop yields have been
negatively affected. The IPCC has high confidence in evidence suggesting that negative
impacts on crop yields have been more frequent than positive impacts on crop yields
(Field et al., 2014). In particular, wheat and maize yields seem to have been reduced as a
result of a changing climate (medium confidence). While economic losses due to extreme
weather events have increased on a global scale, the IPCC has only low confidence in
evidence suggesting that these observed economic losses are directly attributable to
climate change (Field et al., 2014). Evidence of negative health impacts attributable to
climate change is spare but growing. The IPCC has medium confidence in findings
suggesting that heat-related mortality has increased and cold-related mortality has
decreased regionally as a result of mean temperature increases. Medium confidence is
also attributed to findings indicating that the distribution of some waterborne illnesses has
changed due to climate change (Field et al., 2014).
Evidence suggests that climate change has already negatively affected human and natural
systems. Furthermore, the IPCC expects that “the likelihood of severe, pervasive and
irreversible impacts on people and ecosystems” will increase if humans continue to emit
large quantities of GHG emissions and thereby change the climate system (Field et al.,
2014: 62). Section 4 briefly discusses these possible future impacts of climate change.
73
Short summary
Section 3 reviewed changes in the climate system that have been observed and reported
by the IPCC. It highlighted that, in response to human activities, mean temperature
increased, precipitation patterns changed, ice and snow cover diminished, and sea levels
rose. The section also pointed out that the frequency of some extreme events such as the
number of hot days increased as a result of human activities. The section concluded by
reviewing the key impacts of these observed changes on human and natural systems.
4 Anticipated changes in the climate system and potential impacts of climate
change
Section 3 outlined the changes in the climate system that have been observed over past
decades and explained that the main part of these changes is attributable to human
actions. To inform policymakers about possible future changes in the climate system and
to assess potential future impacts on humankind and natural systems, the IPCC relies on
model predictions. This section introduces readers briefly to the different scenarios used
by climate scientists to predict future changes in the climate system, outlines the most
important anticipated changes, and lists key expected impacts on human and natural
systems.
In the fifth assessment report, IPCC reviews and assesses the results of numerous studies
that use climate models to simulate future changes within the climate system. Climate
models are mathematical models that describe key aspects and processes of the climate
system. To simulate changes in the climate system, these models need information on
how climate change drivers such as atmospheric greenhouse gas and aerosol
concentrations will evolve over the coming decades. As this information cannot yet be
observed, researchers have to rely on plausible scenarios of the future. This means that
they have to rely on information generated using plausible assumptions on how human
activities will evolve and what consequences they will have on climate change drivers.
To facilitate worldwide climate research, the IPCC provides such scenarios, which are
called Representative Concentration Pathways (RCPs).39
Each RCP is based on different assumptions about future economic activities, population
growth, energy sources, and other factors, and contains corresponding values of estimated
future greenhouse gas and aerosol emission trajectories and concentrations until 2100.
The IPCC provides four main scenarios, which are called RCP2.6, RCP4.5, RCP6.0, and
RCP8.5. The numbers indicate the projected radiative forcing by the end of the 21st
century (van Vuuren et al., 2011), e.g. RCP8.5 projects radiative forcing of 8.5Wm-2 by
2100.
While a detailed discussion of all the assumptions behind the RCPs is out of the scope of
this teaching material, we will provide a short overview of the main characteristics of
each scenario. RCP2.6 assumes that drastic climate policy interventions manage to reduce
39
For an overview of RCPs, see van Vuuren et al. (2011). For an introductory guide to RCPs, see Wayne
(2013).
74
GHG emissions such that they peak before 2020, leading in 2100 to a slight reduction in
today’s emission levels and atmospheric GHG concentrations that result in radiative
forcing of 2.6 Wm-2 (Wayne, 2013). RCP2.6 can thus be viewed as the IPCC’s best-case
scenario. Directly opposed to RCP2.6 is RCP8.5, which can be viewed as the IPCC’s
worst-case scenario. RCP8.5 assumes a world with no additional climate change policies
and high population growth (Wayne, 2013). In this scenario, greenhouse gas emissions
continue to grow over the entire century and result in radiative forcing of 8.5 Wm -2.
RCP4.5 and RCP6.0 are located between these two “extreme” scenarios. In both of them,
GHG emissions would peak during the century (however, considerably later than in
RCP2.6), leading to radiative forcing of 4.5 Wm-2 and 6.0 Wm-2 by 2100, respectively.
The solid lines starting from 2012 in Figure 27 illustrate the projected pathways of carbon
dioxide emissions for all four scenarios. Note that IPCC (2014) estimates that if humans
were to continue to live as they are today, they would end up in a scenario between
RCP6.0 and RCP8.5.
Figure 27: Carbon dioxide emission trajectories according to the four representative
concentration pathways
Source: IPCC (2014: 9).
Note: GTCO2/yr: Gigatonnes of carbon dioxide per year; RCP: representative concentration pathways.
Numerous studies have used these four RCP scenarios in their climate models to predict
future changes in the climate system. Based on these studies, the IPCC anticipates major
changes within the climate system that we summarize below separately for different
climate variables (temperature, precipitation, snow and ice cover, and sea level).
Simulations based on all four scenarios predict an increase of mean temperature over the
21st century, as indicated by Figure 28. While the magnitudes of the rise in global mean
surface temperature are comparable for all four scenarios over the 2016–2035 period,
mean temperature estimates differ widely for the 2035–2100 period, depending on the
scenario (IPCC, 2014). Compared to the mean of pre-industrial temperature levels (1861–
1880), RCP8.5 is estimated to lead to an increase of more than 4°C, RCP6.0 to an
increase of roughly 3°C, RCP4.5 to an increase of roughly 2.5°C, and RCP2.6 to an
75
increase below 2°C.40 As explained in Modules 1 and 4 of this teaching material, a mean
temperature increase below 2°C is the main target of the Paris Agreement. This implies
that “limiting total human-induced warming (accounting for both CO2 and other human
influences on climate) to less than 2°C relative to the period 1861–1880 with a
probability of >66 per cent would require total CO2 emissions from all anthropogenic
sources since 1870 to be limited to about 2900 Gt” (IPCC, 2014: 63).
Figure 28: Predicted increases in mean surface temperature
Source: IPCC (2014: 63).
Note: Dots indicate decadal averages, with selected decades labelled. The two overlapping dark-orange and
grey surfaces indicate the spread of total human-induced as well as carbon-dioxide-induced warming
obtained by different models and scenarios. GTCO2: Gigatonnes of carbon dioxide; GtC: Gigatonnes of
carbon.
Unlike temperature, no clear global trend is identifiable for precipitation. Some scenarios
(RCP8.5) project an increase in annual mean precipitation for high latitude regions, for
the equatorial Pacific region, and for some mid-latitude regions, but a decrease in other
mid-latitude and subtropical regions (IPCC, 2014). The IPCC also expects an increase in
extreme precipitation events in some regions of the globe.
All simulations predict decreases in Arctic sea ice (RCP8.5-based simulations even
predict a nearly ice-free Arctic ocean in the summer season before the 2050s), nearsurface permafrost (with expected decreases of 37 per cent in RCP2.6 up to 81 per cent in
RCP8.5), and decreasing glacier volume (with expected decreases of 15 to 55 per cent in
40
For the sake of clarity, the spread of these estimates has been omitted. However, the total spread (using
all models and scenarios) of total human-induced warming and carbon dioxide-induced warming is
displayed in Figure 28.
76
RCP2.6 up to 35 to 85 per cent in RCP8.5, excluding glaciers in the Arctic and
Greenland, and Antarctic ice sheets) (IPCC, 2014).
Finally, all RCP scenario-based simulations predict ongoing increases in sea level over
the entire 21st century. IPCC (2014) estimates that by 2081–2100, sea levels will have
risen from 0.26 to 0.55 meters (RCP2.6) up to 0.45 to 0.82 meters (RCP8.5) compared to
their 1986–2005 levels.
All these anticipated changes in the climate system will have impacts on human and
natural systems. Generally speaking, IPCC (2014: 64) expects that climate change “will
amplify existing risks and create new risks for natural and human systems. Risks are
unevenly distributed and are generally greater for disadvantaged people and communities
in countries at all levels of development. Increasing magnitudes of warming increase the
likelihood of severe, pervasive, and irreversible impacts for people, species, and
ecosystems. Continued high emissions would lead to mostly negative impacts for
biodiversity, ecosystem services, and economic development, and would amplify risks for
livelihoods and for food and human security.”
Some of these risks are global, including (a) increased risk of ecosystem and biodiversity
losses; (b) increased risk of food and water insecurity; (c) increased risk of loss of rural
livelihoods and income, especially affecting poor segments of the population; (d)
increased risk to human health and of disrupting livelihoods due to extreme events and
sea level rise; and (e) increased systemic risk as the anticipated increase in the frequency
of extreme weather events could lead to breakdowns of infrastructure networks (IPCC,
2014). Other anticipated risks vary locally, as shown in Figure 29. IPCC (2014: 73)
emphasizes that many “aspects of climate change and its associated impacts will continue
for centuries, even if anthropogenic emissions of greenhouse gases are stopped. The risks
of abrupt or irreversible changes increase as the magnitude of the warming increases.”
77
Figure 29: Key anticipated risks per region
Source: IPCC (2014: 14).
Short summary
Section 4 introduced readers to different scenarios used by the IPCC to predict future
changes in the climate system. The section highlighted that it is likely that future changes
in the climate system (such as increases in temperature) will occur as a result of human
activities. It then concluded by showing that these anticipated changes in the climate
system will have important negative impacts on human and natural systems.
5 Exercises and discussion questions
1. What is the difference between weather and climate?
2. Name and define the five components of the climate system. Discuss two
interactions among selected components.
3. Discuss the global energy balance of the planet by using Figure 15 as a basis for
your discussion. Why is the global energy balance of crucial importance for the
planet’s climate system?
78
4. Why is the average surface temperature 14°C and not -19°C? Define the natural
greenhouse effect.
5. List two internal factors that can affect the climate.
6. List four external factors that can change the climate and explain how they can do
so.
7. Define radiative forcing. Why is this concept a useful tool to measure the
influence of external factors on the planet’s climate?
8. Discuss the different ways in which human activities can affect the climate of the
planet.
9. What human-induced changes in climate variables can already be observed today?
Discuss changes that affect your country.
10. What is the role of scenarios in the IPCC’s simulation of future climate change
impacts?
11. List five anticipated impacts of climate change. Discuss which changes will be
most important for your country.
79
Annex 1
Figure A1: Estimates of global mean radiative forcing provided by the different
IPCC assessment reports
Source: Myhre et al. (2013: 696).
Note: ERF: Effective radiative forcing; SAR: Second IPCC Assessment Report; TAR: Third IPCC
Assessment Report; AR4: Fourth IPCC Assessment Report; AR5: Fifth IPCC Assessment Report.
80
Annex 2
81
References
Baede APM, Ahlonsou E, Ding Y, and Schimel D (2001). The climate system: an
overview. In: Climate Change 2001: The Scientific Basis. Contribution of Working
Group I to the Third Assessment Report of the Intergovernmental Panel on Climate
Change, Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Dai X, and
Maskell K, Johnson CA, eds. Cambridge University Press. Cambridge, UK and New
York.
Bindoff NL, Stott PA, AchutaRao KM, Allen MR, Gillett N, Gutzler D, Hansingo K,
Hegerl G, Hu Y, Jain S, Mokhov II, Overland J, Perlwitz J, Sebbari R, and Zhang X
(2013). Detection and attribution of climate change: from global to regional. In: Climate
Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change, Stocker TF, Qin
D, Plattner GK, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, and Midgley
PM, eds. Cambridge University Press. Cambridge, UK and New York.
Boucher O, Randall D, Artaxo P, Bretherton C, Feingold G, Forster P, Kerminen VM,
Kondo Y, Liao H, Lohmann U, Rasch P, Satheesh SK, Sherwood S, Stevens B, and
Zhang XY (2013). Clouds and aerosols. In: Climate Change 2013: The Physical Science
Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change, Stocker TF, Qin D, Plattner GK, Tignor M,
Allen SK, Boschung J, Nauels A, Xia Y, Bex V, and Midgley PM, eds. Cambridge
University Press. Cambridge, UK and New York.
Common M and Stagl S (2004). Ecological Economics – An Introduction. Cambridge
University Press. Cambridge, UK.
Cubasch U, Wuebbles D, Chen D, Facchini MC, Frame D, Mahowald N, and Winther JG
(2013). Introduction. In: Climate Change 2013: The Physical Science Basis. Contribution
of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, Stocker TF, Qin D, Plattner GK, Tignor M, Allen SK, Boschung J,
Nauels A, Xia Y, Bex V, and Midgley PM, eds. Cambridge University Press. Cambridge,
UK and New York.
Field CB, Barros VR, Mach KJ, Mastrandrea MD, van Aalst M, Adger WN, Arent DJ,
Barnett J, Betts R, Bilir TE, Birkmann J, Carmin J, Chadee DD, Challinor AJ, Chatterjee
M, Cramer W, Davidson DJ, Estrada YO, Gattuso JP, Hijioka Y, Hoegh-Guldberg O,
Huang HQ, Insarov GE, Jones RN, Kovats RS, Romero-Lankao P, Larsen JN, Losada IJ,
Marengo JA, McLean RF, Mearns LO, Mechler R, Morton JF, Niang I, Oki T, Olwoch
JM, Opondo M, Poloczanska ES, Pörtner HO, Redsteer MH, Reisinger A, Revi A,
Schmidt DN, Shaw MR, Solecki W, Stone DA, Stone JMR, Strzepek KM, Suarez AG,
Tschakert P, Valentini R, Vicuña S, Villamizar A, Vincent KE, Warren R, White LL,
Wilbanks TJ, Wong PJ, and Yohe GW (2014). Technical summary. In: 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, Field CB Barros VR, Dokken DJ, Mach
82
KJ, Mastrandrea MD, Bilir TE, Chatterjee M, Ebi KL, Estrada YO, Genova RC, Girma
B, Kissel ES, Levy AS, MacCracken S, Mastrandrea PR, and White LL, eds. Cambridge
University Press, UK and New York.
Forster P, Ramaswamy V, Artaxo P, Berntsen T, Betts R, Fahey DW, Haywood J, Lean J,
Lowe DC, Myhre G, Nganga J, Prinn R, Raga G, Schulz M, and Van Dorland R (2007).
Changes in atmospheric constituents and in radiative forcing. In: Climate Change 2007:
The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment
Report of the Intergovernmental Panel on Climate Change, Solomon S, Qin D, Manning
M, Chen Z, Marquis M, Averyt KB, Tignor M, and Miller HL, eds. Cambridge
University Press. Cambridge, UK and New York.
Hansen JM, Sato M, Kharecha P, and von Schuckmann K (2011). Earth’s energy
imbalance and implications. Atmospheric Chemistry and Physics 11: 13421–449.
Hartmann DL, Klein Tank AMK, Rusticucci M, Alexander LV, Brönnimann S, Charabi
Y, Dentener FJ, Dlugokencky EJ, Easterling DR, Kaplan A, Soden BJ, Thorne PW, Wild
M, and Zhai PM (2013). Observations: atmosphere and surface. In: Climate Change
2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change, Stocker TF, Qin
D, Plattner GK, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, and Midgley
PM, eds. Cambridge University Press. Cambridge, UK and New York.
Hurtt GC, Frolking S, Fearon MG, Moore B, Shevliakova E, Malyshev S, Pacala S, and
Houghton RA (2006). The underpinnings of land-use history: three centuries of global
gridded land-use transitions, wood-harvest activity, and resulting secondary lands. Global
Change Biology 12: 1208–229.
IPCC (1996). Climate Change 1995: The Science of Climate Change. Contribution of
Working Group I to the Second Assessment Report of the Intergovernmental Panel on
Climate Change. Cambridge University Press. Cambridge, UK and New York.
IPCC (2001a). Appendix I: Glossary. In: Climate Change 2001: The Scientific Basis.
Contribution of Working Group I to the Third Assessment Report of the
Intergovernmental Panel on Climate Change, Houghton JT, Ding Y, Griggs DJ, Noguer
M, van der Linden PJ, Dai X, Maskell K, and Johnson CA, eds. Cambridge University
Press. Cambridge, UK and New York.
IPCC (2001b). Climate Change 2001: The Scientific Basis. Contribution of Working
Group I to the Third Assessment Report of the Intergovernmental Panel on Climate
Change, Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Dai X, Maskell
K, and Johnson CA, eds. Cambridge University Press. Cambridge, UK and New York.
IPCC (2007). Climate Change 2007: The Physical Science Basis. Contribution of
Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on
Climate Change, Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB,
83
Tignor M, and Miller HL, eds. Cambridge University Press. Cambridge, UK and New
York.
IPCC (2013a). Climate Change 2013: The Physical Science Basis. Contribution of
Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, Stocker TF, Qin D, Plattner GK, Tignor M, Allen SK, Boschung J,
Nauels A, Xia Y, Bex V, and Midgley PM, eds. Cambridge University Press. Cambridge,
UK and New York.
IPCC (2013b). Summary for policymakers. In: Climate Change 2013: The Physical
Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate, Stocker TF, Qin D, Plattner GK, Tignor M, Allen
SK, Boschung J, Nauels A, Xia Y, Bex V, and Midgley PM, eds. Cambridge University
Press. Cambridge, UK and New York.
IPCC (2014). Climate Change 2014: Synthesis Report. Contribution of Working Groups
I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate
Change, Pachauri RK and Meyer LA, eds. Intergovernmental Panel on Climate Change.
Geneva.
Kiehl JT, and Trenberth KE (1997). Earth’s annual global mean energy budget. Bulletin
of the American Meteorological Society 78(2): 197–208.
Le Treut H, Somerville R, Cubasch U, Ding Y, Mauritzen C, Mokssit A, Peterson T, and
Prather M (2007). Historical overview of climate change. In: Climate Change 2007: The
Physical Science Basis. Contribution of Working Group I to the Fourth Assessment
Report of the Intergovernmental Panel on Climate Change, Solomon S, Qin D, Manning
M, Chen Z, Marquis M, Averyt KB, Tignor M, and Miller HL, eds. Cambridge
University Press. Cambridge, UK and New York.
Masson-Delmotte V, Schulz M, Abe-Ouchi A, Beer J, Ganopolskia , González Rouco JF,
Jansen E, Lambeck K, Luterbacher J, Naish T, Osborn T, Otto-Bliesner B, Quinn T,
Ramesh R, Rojas M, Shao X, and Timmermann A (2013). Information from paleoclimate
archives. In: Climate Change 2013: The Physical Science Basis. Contribution of Working
Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate
Change, Stocker TF, Qin D, Plattner GK, Tignor M, Allen SK, Boschung J, Nauels A,
Xia Y, Bex V, and Midgley PM, eds. Cambridge University Press. Cambridge, UK and
New York.
Mastrandrea MD, Field CB, Stocker TF, Edenhofer O, Ebi KL, Frame DJ, Held H,
Kriegler E, Mach KJ, Matschoss PR, Plattner GK, Yohe GW, and Zwiers FW (2010).
Guidance note for lead authors of the IPCC Fifth Assessment Report on Consistent
Treatment of Uncertainties. Intergovernmental Panel on Climate Change. Geneva.
84
Murphy D, Solomon S, Portmann R, Rosenlof K, Forster P, and Wong T (2009). An
observationally based energy balance for the earth since 1950. Journal of Geophysical
Research: Atmospheres 114(D17107).
Myhre G, Shindell D, Bréon FM, Collins W, Fuglestvedt J, Huang J, Koch D, Lamarque
JF, Lee D, Mendoza B, Nakajima T, Robock A, Stephens G, Takemura T, and Zhang H
(2013). Anthropogenic and natural radiative forcing. In: Climate Change 2013: The
Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report
of the Intergovernmental Panel on Climate Change, Stocker TF, Qin D, Plattner GK,
Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, and Midgley PM, eds.
Cambridge University Press. Cambridge, UK and New York.
Perman R, Ma Y, Common M, Maddison D, and McGilvray J (2011). Natural Resource
and Environmental Economics. Fourth edition. Pearson Education Limited. Essex, UK.
Ramaswamy W, Boucher O, Haigh J, Hauglustaine D, Haywood J, Myhre G, Nakajima
T, Shi GY, and Solomon S (2001). Radiative Forcing of Climate Change. In: Climate
Change 2001: The Scientific Basis. Contribution of Working Group I to the Third
Assessment Report of the Intergovernmental Panel on Climate Change, Houghton JT,
Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Dai X, Maskell K, and Johnson CA,
eds. Cambridge University Press. Cambridge, UK and New York.
Rhein M, Rintoul SR, Aoki S, Campos E, Chambers D, Feely RA, Gulev S, Johnson GC,
Josey SA, Kostianoy A, Mauritzen C, Roemmich D, Talley LD, and Wang F (2013).
Observations: ocean. In: Climate Change 2013: The Physical Science Basis. Contribution
of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, Stocker TF, Qin D, Plattner GK, Tignor M, Allen SK, Boschung J,
Nauels A, Xia Y, Bex V, and Midgley PM, eds. Cambridge University Press. Cambridge,
UK and New York.
Trenberth KE, Fasullo JT, and Kiehl J (2009). Earth’s global energy budget. Bulletin of
the American Meteorological Society 90: 311–24.
Trenberth, KE, Jones PD, Ambenje P, Bojariu R, Easterling D, Klein Tank A, Parker D,
Rahimzadeh F, Renwick JA, Rusticucci M, Soden B, and Zhai P (2007). Observations:
surface and atmospheric climate change. In: Climate Change 2007: The Physical Science
Basis. Contribution of Working Group I to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change, Solomon S, Qin D, Manning M, Chen Z,
Marquis M, Averyt KB, Tignor M, and Miller HL, eds. Cambridge University Press.
Cambridge, UK and New York.
Van Vuuren DP, Edmonds J, Kainuma M, Riahi K, Thomson A, Hibbard K, Hurtt GC,
Kram T, Krey V, Lamarque JF, Masui T, Meinshausen M, Nakicenovic N, Smith SJ, and
Rose SK (2011). The representative concentration pathways: An overview. Climatic
Change 109: 5–31.
85
Vaughan DG, Comiso JC, Allison I, Carrasco J, Kaser G, Kwok R, Mote P, Murray T,
Paul F, Ren J, Rignot E, Solomina O, Steffen K, and Zhang T (2013). Observations:
cryosphere. In: Climate Change 2013: The Physical Science Basis. Contribution of
Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, Stocker TF, Qin D, Plattner GK, Tignor M, Allen SK, Boschung J,
Nauels A, Xia Y, Bex V, and Midgley PM, eds. Cambridge University Press. Cambridge,
UK and New York.
Wayne G (2013). The Beginner’s Guide to Representative Concentration Pathways.
Sceptical Science. Available at: http://www.skepticalscience.com/rcp.php?t=3.
Wild M (2012). New directions: a facelift for the picture of the global energy balance.
Atmospheric Environment 55: 366–67.
Wild M, Folini D, Schär C, Loeb N, Dutton EG, and König-Langlo G (2013). The global
energy balance from a surface perspective. Climate Dynamics 40: 3107–134.
86
Module 3
The economics of climate change
1 Introduction
Module 1 showed that economic activities can influence the climate because the economy
and the environment are interdependent. Module 2 explained the climate science behind
climate change and showed how human activities contribute to change the climate. This
module shows how economic theory explains the occurrence of transboundary
environmental problems, such as climate change, and clarifies why it can be so difficult
to implement solutions to these problems even though those solutions are well known.
Neoclassical economic theory stipulates that markets allocate resources in an optimal, i.e.
efficient, way. At the same time, however, economic behaviour leads to important
negative effects on the environment such as climate change. If markets allocate resources
efficiently, how is it possible that severe environmental problems like climate change
exist? Sections 2 and 3 of this module aim to answer this question.
Section 2 reviews the basic concepts of economic theory on markets and optimal
allocations. It introduces the standard theoretical model describing market economies and
the concept of Pareto efficiency used to evaluate whether resources are allocated
efficiently. It shows that, in theory, fully competitive markets allocate resources in a
Pareto-efficient way and thus optimally from a social point of view. However, this
theoretical result only holds up provided that certain assumptions are satisfied; if these
assumptions are not met, markets fail and generate outcomes that are not socially optimal.
This is what is called a “market failure.”
Section 3 shows that climate change is the result of such a market failure occurring
because two assumptions are not met. First, contrary to the assumption of all
commodities being private goods, the atmosphere is an open-access resource (see Section
3.1). As such, there are no legal constraints on its use, and therefore any country can
release any amount of greenhouse gases into the atmosphere at any time. Open-access
resources cannot be “efficiently” exploited in market economies because countries tend to
overuse them. Second, GHG emissions are negative externalities (see Section 3.2).
Negative because they detract from social welfare and externalities because they are
external to any accounting within the economic system. Economic agents do not take
negative externalities into account when making their decisions, and consequently overemit greenhouse gases into the atmosphere. Due to the combination of these two factors,
markets fail to generate a Pareto-optimal situation. This gives rise to climate change,
considered to be the biggest market failure in human history.
Since markets have failed, policy interventions are needed to mitigate climate change.
Even though it has been known for some time how policy interventions could mitigate
climate change, the international community has been struggling for decades to solve the
issue. Section 4 investigates how economic theory explains the inability of governments
and societies to mitigate climate change. It shows that climate change mitigation is a
87
global “public good” that has non-rival and non-excludable benefits (see Section 3.1). As
no country can single-handedly take actions to eliminate the threat of climate change,
international cooperation is needed. However, such cooperation is extremely difficult to
achieve when dealing with a global public good. To illustrate this difficulty, the section
introduces game theory concepts that are used to analyse strategic interaction among
countries. It shows that the main cause of the failure of collective action is that countries
have strong incentives to free-ride and not contribute to reducing emissions. This leads to
an outcome that is not socially optimal. The section concludes by drawing several lessons
from game theory models that allow us to identify ways to foster cooperation among
countries.
At the end of this module, readers should be able to:






State the implications of the first welfare theorem;
Understand the conditions under which the first welfare theorem does not hold
and define market failures;
Explain why the atmosphere is considered to be an open-access resource;
Explain why greenhouse gas emissions are negative externalities;
Discuss why climate change is considered to be the biggest market failure in
human history;
Use game theory tools to analyse how economists explain the fact that the
international community has been struggling to implement climate change
mitigation policies.
To support the learning process, readers will find several exercises and discussion
questions in Section 5 covering the issues introduced in Module 3. Additional reading
material can be found in Annex 1.
2 The competitive markets model and Pareto efficiency
Neoclassical economic theory stipulates that markets allocate resources efficiently. If this
is indeed the case, then why do markets sometimes fail to do so, producing highly
inefficient outcomes leading to environmental problems such as climate change? To
answer this question we first have to understand why economic theory concludes that
markets produce efficient outcomes, in particular, understand (a) how economists
conceptualize market economies, and (b) how economists evaluate the efficiency of
market economies. This section clarifies these two points and shows that, given certain
conditions, economic theory predicts that markets indeed produce efficient outcomes.
Section 3 subsequently shows that in the context of climate change, some of these
conditions are violated, and therefore markets fail to achieve efficient outcomes.
2.1 A simple model of market economies
Economic processes are inherently complex and difficult to understand. Economists thus
rely on theoretical models of the economy to improve their understanding of these
processes. Economic models are theoretical constructs that aim to represent different
88
aspects of economic processes in a simplified way. They therefore rely on assumptions
that might not necessarily hold in reality. Nevertheless, the models are very useful tools
because they shed light on complex economic mechanisms and their relationships with
the environment. Market economies, currently the dominant form of economic
organization (Cohen, 2009), have been extensively studied and modelled. Economists
have developed a standard model of a competitive market economy,41 which we will now
briefly introduce and which will serve as a starting point to understand how economic
theory explains the occurrence of climate change.
In this standard model, the economy is assumed to be composed of economic agents
(firms and households) acting purely in their own self-interest – i.e. trying to optimize
their behaviour to achieve the outcome that is best for them. In other words, firms try to
maximize their profits while households try to maximize their utility (i.e. consume a mix
of goods and services that makes them as satisfied as they can be on a given budget). The
model assumes that there are many different markets, one for each commodity. Each
competitive market connects demand (how much households are willing to pay for the
commodity) and supply (how much firms want to be paid to produce the commodity),
thereby determining how much of the specific commodity is produced. In such a
competitive market, the price of a commodity adjusts until demand equals supply. This is
called equilibrium. Put differently, a market is in equilibrium if the sum of the quantities
of commodities that households want to buy at current prices equals the quantities of
commodities that firms want to produce at current prices (see Box 12 for a graphical
illustration of an equilibrium in a single competitive market).
Box 12: Equilibrium in a single competitive market
Figure 30 displays a single competitive market for a normal good and shows the demand and
supply curve for this particular good. The demand curve relates the quantity demanded by
consumers to the market price. It is downward sloping, meaning that the cheaper the good, the
more of it consumers are willing to buy. The supply curve relates the quantity supplied to the
price and is upward sloping, meaning that the higher the price firms can charge, the more they are
willing to supply. The market is in equilibrium at price P* and corresponding quantity Q*, as this
is the only point where demand equals supply. Note that given the supply and demand curves, this
is the only equilibrium in this particular market. For any other price, supply does not equal
demand. To see this, look for instance at price Pl: at this price, consumers demand Qdl, but
producers are only willing to supply Qsl. This situation is clearly not an equilibrium, as the
quantity demanded is higher than the quantity supplied.
41
Going into the details of the microeconomic foundations of the competitive market model is beyond the
scope of this material. Interested readers may refer to standard microeconomic textbooks such as Varian
(2010). Additional readings can also be found in Annex 1.
89
Figure 30: Equilibrium in a single competitive market
Source: Author.
If all firms maximize their profits, all consumers maximize their utility given their budget
constraints, and all individual commodity markets are in equilibrium, then the resulting
allocation of commodities and the corresponding set of prices is called a competitive
equilibrium. Put differently, in this state, the whole market economy, and not only a
single commodity market, is in equilibrium.
While the model allows us to understand how market economies allocate resources, it
remains silent about how efficient those economies are in allocating these resources. This
is important to know, however, as resources are scarce and allocating them improperly is
costly. Economists have tried to answer this question by using the competitive market
model and a very specific concept of efficiency – Pareto efficiency – introduced in the
next section.
2.2 Pareto efficiency
Various conceptual tools are used to evaluate different allocations of resources. The most
commonly used tool is called Pareto efficiency,42 named after the 19th century Italian
economist Vilfredo Pareto. An allocation of resources in which one can find a way to
make one or more persons better off without making another person worse off is called a
Pareto-inefficient allocation. If there is no way to make one or more persons better off
without making another person worse off an allocation is called a Pareto-efficient
allocation.
Pareto-inefficient allocations are a priori not desirable. After all, if you could improve the
situation of at least one person without deteriorating the situation of another person, i.e. if
42
Pareto efficiency is sometimes also referred to as economic efficiency or Pareto optimality.
90
you could “Pareto-improve” a situation, why not do it? When all such Pareto
improvements have been made and no more are available, the allocation of resources is
said to be Pareto-optimal.
While the Pareto-efficiency criterion is useful as a benchmark for evaluating whether or
not an allocation is efficient, it has its limits. If two allocations are Pareto-efficient, the
criterion is unable to evaluate which of the two allocations is preferable. Amartya Sen
(1970: 22) made this point explicit by writing that an economy may be Pareto-efficient
“even when some people are rolling in luxury and others are near starvation, as long as
the starvers cannot be made better off without cutting into the pleasures of the rich... [A]n
economy can be Pareto optimal and still be perfectly disgusting.” Thus, Pareto efficiency
does not allow us to identify allocations that optimize justice or equality. The criterion
can only determine whether a society is allocating resources wastefully or not (MasColell et al., 1995) and whether allocations can be changed in a way that unambiguously
improves the welfare of a society.
2.3 Competitive market equilibria and Pareto efficiency
Despite its limitations, Pareto efficiency allows for evaluating whether market
economies, as modelled by the competitive market model, allocate resources efficiently.
According to neoclassical economic theory, this is indeed the case. The first welfare
theorem states that, as long as some conditions are satisfied, competitive markets’
equilibria are Pareto-efficient allocations.43 The first welfare theorem thus formalizes
Adam Smith’s conjecture in The Wealth of Nations in 1776 of an “invisible hand,” and is
one of the most central theoretical results in economics. It implies that markets composed
of economic agents acting purely in their own self-interest will achieve an allocation of
resources that constitutes a social optimum from which nobody’s position can be
improved without harming someone else. Box 13 illustrates this through the example of
production in a single market.
43
Lerner (1934, 1944) was the first economist to (graphically) prove the existence of the first welfare
theorem. Lerner (1944) also outlined the foundations of the second welfare theorem, stating that any
Pareto-efficient equilibrium can be achieved by a market after lump-sum transfers of the initial wealth
endowment, a theorem that was first formally proved by Arrow (1951).
91
Box 13: A Pareto-efficient level of production in a single market
Let us take an intuitive look at why market equilibria can be considered Pareto-efficient, using the
example of production output in a single market. Figure 31 displays supply and demand curves of
a normal good in a competitive market. As we have already seen, competitive markets determine
the total quantity produced by taking into account how much consumers are willing to pay for the
good and how much suppliers have to be paid to produce the good. In an equilibrium, demand
and supply are balanced so that the willingness of consumers to pay an amount P* for an extra
unit is equal to the willingness of suppliers to be paid the amount P* to supply an extra unit.
Why is the resulting equilibrium production amount considered to be Pareto-efficient? Suppose
that instead of producing the equilibrium quantity Q*, suppliers would produce less, say, Ql. Then
some producer would be willing to produce an additional unit and sell it at a price slightly above
Psl, but well below the price that someone would be willing to pay for an extra unit (which would
be a price slightly below Pdl). This would clearly be a win-win scenario, as by producing and
exchanging that additional unit, at least these two persons would be better off. The same holds for
all quantities below Q*. Thus as long as less than Q* is produced, at least one person could be
made better off by increasing total production. Would producing a quantity greater then Q* also
be a Pareto improvement? The answer is no. As the price any supplier asks for an extra unit
produced would be higher than any consumer’s willingness to pay for that unit, nobody would
buy this extra unit, and thus no person would be better off. At the same time, the supplier
producing the extra unit that cannot be sold would be even worse off. Therefore, only the market
equilibrium allocation Q* with the corresponding price P* can be considered to be a Paretoefficient allocation. This is an intuitive illustration of the claim that competitive markets
determine a Pareto-efficient amount of output.
Figure 31: A Pareto-efficient level of output in a single competitive market
Source: Author's elaboration based on a similar discussion in Varian (2010: 310–12).
The first welfare theorem depends, however, on a series of assumptions that Perman et al.
(2011: 103) call “institutional arrangements.” These arrangements have to be satisfied or
the first welfare theorem does not hold and competitive markets do not produce socially
92
optimal situations. These assumptions require that (a) markets exist for all goods and
services; (b) these markets are perfectly competitive (no agent can influence the price);
(c) all agents have full information on current and future prices; (d) private property
rights exist for all resources and commodities; (e) there are no externalities; and (f) all
commodities are private goods.44,45 In reality, these assumptions are often violated
(Sandler, 2004). Consequently, markets fail to achieve Pareto-efficient allocations and
economists speak of market failures, which can potentially have devastating
consequences and lead to situations that are not socially optimal. As we will see in
Section 3, climate change is the result of such a market failure. Some have even argued
that climate change is the biggest market failure in human history (Stern, 2007).
Short summary
Section 2 reviewed basic elements of economic theory that are needed to understand how
economic theory explains climate change. The section started by introducing the standard
model of a competitive economy and discussed the concept of market equilibrium. It then
briefly introduced the Pareto-efficiency criterion used to evaluate different allocations of
resources. Finally, it showed that markets allocate resources in a Pareto-efficient way
provided that certain conditions are satisfied, but fail to do so if these conditions are not
met. These so-called market failures are at the heart of economists’ explanations of the
occurrence of climate change and are analysed in greater detail in the following sections.
3 Climate change: The greatest market failure in human history
Section 2 of this module showed that, according to the first welfare theorem of
neoclassical economic theory, markets composed of economic agents acting purely in
their own self-interest allocate resources in a Pareto-efficient way, i.e. produce a socially
optimal situation in which no one’s welfare can be improved without harming someone
else’s welfare. In this ideal theoretical world, human-induced climate change would not
occur.
However, current empirical evidence points towards the existence of such humaninduced climate change (IPCC, 2014a). Economists maintain that this empirical finding is
due to the violation of two underlying assumptions of the first welfare theorem. The first
of these assumptions, assumption (f), is that all goods have to be private goods. But since
the atmosphere is an open-access resource (see Section 3.1), this assumption is violated.
As we will see, open-access resources are very different from “normal” private goods,
and markets cannot Pareto-efficiently handle the former. The second violated assumption,
44
In addition to the assumptions highlighted in this section, there is an additional technical assumption on
consumer preferences called local nonsatiation of preferences, which also has to be satisfied for the first
welfare theorem to hold (Mas-Colell et al., 1995: 549–50).
45
Note that, strictly speaking, assumptions (e) and (f) are already implied by assumption (d). If enforceable
private property rights for all resources and commodities do exist, then all benefits and/or costs accrue to
the agent holding the property right for the commodity or resource. This implies that there would not be
any externalities and only private goods. However, we follow Perman et al. (2011) and list these
assumptions separately for the sake of clarity, as we will discuss assumptions (e) and (f) extensively in
sections 3.1 and 3.2.
93
assumption (e), concerns the absence of any externality. As we will see in Section 3.2,
GHG emissions are an example of a negative externality that again leads to a Paretoinefficient outcome.
Because these two assumptions are violated, the first welfare theorem does not hold up,
markets fail, and economic agents over-emit greenhouse gases, leading to growing GHG
accumulation in the atmosphere that exacerbates climate change. In Sections 3.1 and 3.2,
we discuss both violations in detail and show why climate change is considered to be the
result of a major market failure.
3.1 The atmosphere: An open-access resource
Theory tells us that markets achieve Pareto-optimal allocations if they deal with private
goods, but tend to fail if market systems involve non-private goods. This section will
show that the atmosphere, in its function as a GHG repository, is not a private good. The
section will then help us understand the consequences of this fact.
Goods can be classified as private or public using two criteria that characterize the
benefits of their consumption: rivalry and excludability (Sandler, 2004; Perman et al.,
2011).46 Benefits of consumption are rival when a unit of a good can only be consumed
by one individual, meaning that one individual’s consumption is at the expense of
somebody else’s consumption. Benefits of consumption are excludable when one can
prohibit another person from consuming the good. Pure private goods are defined as
having completely rival and excludable benefits. To illustrate the concept, let us look at a
particular pure private good: bread. If an individual eats a loaf of bread, the loaf is
destroyed and cannot be eaten by anybody else. Thus, the benefits of consuming a loaf of
bread are completely rival. At the same time, consuming bread is completely excludable,
as bread comes in separable units and one can identify who owns a specific unit and thus
has the right to consume it. The right to consume the loaf can be traded or gifted, but
everything else is considered stealing and is punishable by law. This means that any
individual not owning the right to consume the loaf can be excluded from eating a
particular loaf of bread; in other words, the benefits of consuming the loaf are excludable.
Bread, similar to most goods we consume in our daily life, is therefore a pure private
good.
The opposite of pure private goods are pure public goods. These types of goods were first
described in 1954 by the American economist Paul Samuelson (1954). The benefits of
pure public goods are non-rival and non-excludable. Non-excludability means that
nobody can be excluded from consuming the good, and consumption entitlements can be
neither identified nor traded or gifted. Non-rivalry means that an individual can consume
a particular unit of the good without affecting the possibility of another individual
consuming the same unit. Let us illustrate this with a classical example of a pure public
good: national defence. Once a certain level of defence is provided by the military, no
citizen who is living within the country can be excluded from being protected by the
Note that some scholars (e.g. Tietenberg and Lewis, 2012) use the term “divisibility” as an equivalent of
“rivalry.”
46
94
military (in other words, from consuming the benefits of the provided defence level).
Hence, the benefits of national defence are non-excludable. If a citizen consumes the
provided defence level, he or she does not diminish in the slightest the possibility of
another citizen consuming the same level of defence. Thus, the benefits of national
defence are also non-rival.
The economics governing pure private and pure public goods are thus very different.
Within these two extreme cases, the literature identifies a variety of goods with different
degrees of rivalry and excludability, or in other words, with different degrees of
publicness (Perman et al., 2011).47 For our purpose, we will not discuss all these different
types of goods,48 but rather concentrate on one particular type – open-access resources –
to which the atmosphere belongs.
Open-access resources are part of a larger class of goods called common-pool resources.
They are defined by Tietenberg and Lewis (2012: 629) as “common-pool resources with
unrestricted access.” Because scholars are still struggling to unambiguously define
common-pool resources, we will follow Ostrom (2008: 11) who defines them as
“sufficiently large that it is difficult, but not impossible, to define recognized users and
exclude other users altogether. Further, each person’s use of such resources subtracts
benefits that others might enjoy.” Thus, the benefits of consuming common-pool
resources are rival (which distinguishes them from pure public goods) but excludable
only with difficulties. If no individual or group owns a common-pool resource or is able
to exercise full control over such a resource, the resource is called an open-access
resource. Examples of such open-access resources are fisheries, forests, or the
atmosphere.
To be specific, the atmosphere is an open-access resource with respect to its function as a
repository of GHG emissions. Everybody can use the atmosphere to dump their GHG
emissions free of charge, and it is difficult (although not impossible) to restrain this
usage. Thus, the benefits of using the atmosphere’s function as a GHG repository are
only partly excludable. Moreover, if firms and individuals use the atmosphere as a GHG
repository, other firms and individuals cannot do the same without causing serious
negative environmental consequences like climate change. In that sense, the benefits of
using the atmosphere as a GHG repository are rival.
Considering that the atmosphere is an open-access resource and not a pure private good,
economic theory tells us that it cannot be efficiently exploited by a market economy. If
market economies exploit open-access resources such as the atmosphere, they tend to
overuse these resources. The resulting market failure then leads to outcomes that are not
socially optimal. The main reason for this overuse is the open access to the resource. As
no individual or group is able to control who is dumping greenhouse gases into the
atmosphere, economic actors adopt a “use it or lose it” mentality and use the GHG
It is important to understand that the attribute “public” refers to the publicness property of rivalry and
excludability, not to the way a good is produced (by public or private agents). It is possible that a public
good is privately produced and/or a private good is publicly produced.
48
See Sanders (2004) for an overview on the different types of these goods.
47
95
repository function of the atmosphere on a “first come, first serve” basis, thereby
collectively overusing the resource (Tietenberg and Lewis, 2012) by dumping massive
amounts of greenhouse gases into the atmosphere. This problem is widely known as the
“tragedy of the commons,”49 and is at the heart of economists’ explanation of the
occurrence of climate change.
As we will see in Section 3.2, the overuse induced by the open-access problem is made
worse by what economists call negative externalities: economic agents make their
decisions to exploit the atmosphere’s GHG repository function solely on the basis of their
own benefits and costs. By doing so, they ignore the negative impact their GHG
emissions have on others. This leads to a situation in which no agent has an incentive to
reduce its GHG dumping activities and to conserve the atmosphere’s capacity to absorb
greenhouse gases.
3.2 Greenhouse gas emissions: A negative externality problem
Section 3.1 showed that market economies overuse the GHG repository function of the
atmosphere because they cannot efficiently manage open-access resources. In addition to
the open-access problem, there is a second related factor contributing to the overuse of
the atmosphere’s GHG repository function: emitting greenhouse gases is a negative
externality problem. This is an additional reason why markets fail and hence climate
change occurs.
An externality exists if actions by one economic agent unintentionally affect other agents.
To be more precise, according to Perman et al. (2011: 121), an externality occurs if “the
production or consumption decisions of one agent have an impact on the utility or profit
of another agent in an unintended way, and when no compensation/payment is made by
the generator of the impact to the affected party.” A priori, externalities can be positive or
negative.50 Whereas positive externalities have a positive unintended effect on other
agents, negative externalities have a negative unintended effect on others. Vaccinations
are examples of positive externalities. They protect the vaccinated person as intended, but
also have a positive unintended external effect, lowering the probability that other nonvaccinated persons catch the disease. GHG emissions are a typical example of negative
externalities. While they are the by-product of an intended production process, they have
a negative unintended effect on others by contributing to climate change.
Markets fail in the presence of externalities because agents do not take these effects into
account when making their decisions. The reason why they do not take them into account
is that, by definition, these effects are unintended and hence, there is no monetary reward
Actually, as noted by Common and Stagl (2004), the expression “tragedy of the commons” – introduced
in a famous article by Garrett Hardin in 1968 – is misleading because the problem is not related to common
property rights, but to the open access to the common property. Thus, as Common and Stagl suggest, the
tragedy should have been labeled “the open-access tragedy.”
50
Note that there are various ways to classify externalities. Besides classifying externalities into positive
and negative, one can also classify them by the economic activity in which they originate (production or
consumption) and by the economic activity that they affect (production or consumption). See Perman et al.
(2011) for an overview on classifications of different externalities.
49
96
or penalty to the agent generating the externality. Thus, in the case of positive
externalities, generating the positive effect, in itself, is not sufficiently encouraged as
there is no reward for doing so. In the case of negative externalities the opposite happens:
as there is no monetary punishment for generating negative effects, agents are not
sufficiently discouraged from generating them. Consequently, markets produce too much
of the negative externality, thereby exacerbating the negative external effect. As we will
see, this is exactly what happens in the case of GHG emissions.
Human beings produce greenhouse gases as by-products of production processes (e.g. by
burning fossil fuels to produce energy, by raising cattle, and through a very wide variety
of other activities). Emitting large quantities of greenhouse gases into the atmosphere
causes climate change and thus indirectly imposes costs on present and future
generations. These costs are not borne by the emitting firms or consumers of goods
whose production emits greenhouse gases, but rather by all people across the planet, and
by future generations. Thus, when making a production and/or consumption decision
involving GHG emissions, economic agents only take their personal costs and benefits
into account, ignoring the additional costs they impose on others. Consequently, they
emit more than what is socially optimal. To clarify this, let us look at Figure 32 for a
graphical illustration of the case of production decisions involving greenhouse gases.51
Figure 32 displays a market for a given good from the point of view of a specific firm
that emits greenhouse gases during its production process. The x axis displays the
quantity produced and the y axis the price of the good. The demand curve represents
consumers’ demand for the good. As usual, the lower the price, the more of the good
people want to buy. Note that the demand curve also represents the firm’s private
marginal benefits (PMB). A firm’s private marginal benefit corresponds, loosely
speaking, to the price the firm can charge per unit produced. For simplicity, we shall
assume here that there are no consumption-related externalities, and thus private marginal
benefits equal social marginal benefits (SMB).52
51
The following example of a negative production externality closely follows the discussion in Common
and Stagl (2004: 327–29).
52
Consumption-based externalities would drive a wedge between the marginal benefits perceived by a firm
(PMB) and the marginal benefits perceived by society (SMB).
97
Figure 32: Greenhouse gas emissions: A negative production externality problem
Source: Author's elaboration based on Common and Stagl (2004: 327).
Figure 32 also displays a second curve labeled PMC, which stands for the firm’s private
marginal costs. This curve indicates, loosely speaking, the costs the firm incurs for
producing the last unit of its total production output. These costs are also used by the
firm’s managers to decide how much the firm will produce. In order to maximize the
firm’s profits, the manager will equate the firm’s PMCs with the firm’s PMBs
(represented by the demand curve). Thus, the firm will produce the quantity QM, selling
each unit at the price PM, as this is the only point where PMC equals PMB. The problem
is that this allocation is not Pareto-efficient and thus not optimal from society’s point of
view. Why is this the case?
As mentioned earlier, the firm emits greenhouse gases during the production process and
thus contributes to climate change. The costs associated with emitting greenhouse gases
(i.e. the costs of climate change) are, however, not borne by the firm, but by the rest of
the society.53 From the point of view of the society, these are additional costs associated
with the production of the good that have to be taken into account. Therefore, the social
marginal costs of producing the good (SMCs) are higher than the firm’s private marginal
costs (PMCs). The difference between the social marginal cost and private marginal cost
is known as the external marginal cost and is given by EMC = SMC-PMC. In our
example, the external marginal costs correspond to the climate change-related costs
occurring as a result of greenhouse gases emitted during the production of the good. The
Pareto-efficient point of production would thus not be the production level the firm
chooses, but the pair QSO and PSO, where social marginal benefits equal social marginal
costs. We thus see in Figure 32 that the firm produces too much of the good (QM instead
of QSO) and sells the good at a price that is too low (PM instead of PSO) because it ignores
the climate change-related costs it imposes on others. The existence of the negative
production externality therefore leads to a market failure due to which the firm emits too
53
To be precise, it is possible that the firm bears some of the costs of climate change, but the bulk of these
costs accrue either to other individuals on the planet or to future generations.
98
much greenhouse gases into the atmosphere compared to a Pareto-efficient situation. This
example can be generalized and extended to also include negative consumption
externalities, such as the pollution created by using cars.
In conclusion, as GHG emissions are negative externalities and the atmosphere can be
used as an open-access resource to dump them, markets fail to achieve Pareto-optimal
situations. Instead of optimally allocating resources, market economies dump
unprecedented quantities of GHG emissions into the planet’s atmosphere, and are thereby
changing the climate of the earth. This massive market failure represents a danger for
present and future generations and as such has been identified by Stern (2007) as the
biggest market failure in the history of humankind.
Short summary
Sections 2 and 3 showed that while markets theoretically achieve Pareto-efficient
outcomes, they fail to do so if some conditions are not satisfied. The occurrence of
climate change is an example of such a failure, identified by some influential authors as
the biggest market failure in human history. As shown in Section 3, this failure has two
causes. First, the atmosphere’s function as a GHG repository is an open-access resource
and is therefore overused by economic actors because, unlike pure private goods, openaccess resources cannot be efficiently dealt with by market economies. Second,
greenhouse gas emissions are negative externalities. Economic actors do not directly bear
the climate change-related costs associated with the emissions that they can dump free of
charge into the atmosphere; consequently, they emit too much greenhouse gases.
Together, these two reasons explain why markets fail and, hence, why climate change
occurs.
4 Why is it so hard to solve the climate change problem?
As shown in Section 3, climate change occurs because markets fail and do not achieve
Pareto-optimal allocations. As we have seen, rational economic agents, pursuing their
own self-interest in an uncoordinated way, overexploit the atmosphere’s GHG repository
function and ignore the costs they impose on others. Consequently, they emit too much
greenhouse gases into the atmosphere and cause climate change. Therefore, to solve the
problem one cannot simply rely on a market system, as it is precisely this system that
fails and which is at the root of the problem.
Climate change can only be fixed by a policy intervention that corrects the market failure.
A priori, the solution is simple: if humans emitted less greenhouse gas into the
atmosphere, they could stop or at least considerably slow climate change. If each
country’s politicians decided to cut GHG emissions, climate change could be mitigated or
even avoided altogether. Thus, where is the problem? Why is humankind still struggling
to implement this solution on a global scale? Since Mancur Olson’s now-famous work in
1965, The Logic of Collective Action, economists have been working on a theory to
explain this puzzling situation. Section 4 will show how economic theory explains the
fact that the international community has been struggling for decades to solve the climate
99
change problem despite the fact that the apparently simple solution has long been known.
As we will see, much of the problem is related to the fact that solving global
environmental problems like climate change requires cooperation among countries,
which is not necessarily easy to achieve.
4.1 Mitigating climate change: A global public good
Mitigating climate changes requires reducing anthropogenic GHG emissions (IPCC,
2014b). Economists conceptualize climate change mitigation as a good that can be
supplied and consumed. To supply the good, efforts have to be taken to reduce GHG
emissions. Consumption of the good is passive: everybody can reap the benefits of
climate change mitigation by living in a world that is not affected by the potentially
catastrophic effects of a changing climate (IPCC, 2014a). However, climate change
mitigation is not a “normal” good, i.e. it is not a pure private good like most of the
commodities we consume in our daily lives.54 From an economist’s point of view, climate
change mitigation is a so-called global public good (Ferroni et al., 2002; IPCC, 2001;
Perman et al., 2011; Sandler, 2004). The public-good nature of climate change mitigation
is at the heart of economists’ explanation for the inability of the international community
to solve the climate change problem. As we will see in the sections that follow, supplying
global public goods in sufficient quantities is difficult and involves several problems that
are hard to overcome.
Section 3.1 introduced pure public goods and defined them as those having nonexcludable and non-rival benefits. As we explained, non-excludability means that nobody
can be excluded from consuming the good, i.e. once the goods are supplied, their benefits
are instantly available to everybody. Non-rivalry means that a particular unit of the good
can be consumed by an individual or group without affecting the possibility of another
individual or group consuming the same unit. Sandler (2004) compares this property of
global public goods to magic, stating that, like magic, a unit of a public good can be
consumed by any number of individuals without any degradation in quality or quantity of
that unit.
Climate change mitigation has exactly these two properties and is thus considered to be a
public good. If a country decides to reduce its GHG emissions, the benefits are instantly
available to all humans worldwide. Cutting a country’s GHG emissions directly lowers
the global GHG concentration in the atmosphere and thereby slows climate change
compared to a business-as-usual scenario. Thus all countries gain from such a reduction.
For the country that cuts its emissions, it is impossible to limit the benefits of its GHG
reductions only to itself. The benefits of climate change mitigation are therefore nonexcludable. Moreover, the benefits of climate change mitigation are also non-rival. The
benefit a country obtains from reducing its GHG emissions does not reduce in the
slightest the benefits other countries obtain from this reduction. Thus, consuming the
benefits of GHG reductions does not affect the possibility of other nations enjoying these
benefits. As the benefits are both non-rival and non-excludable, GHG reductions are
considered to be pure public goods. Moreover, given that the benefits of a country’s GHG
54
See Section 3.1 for a definition of pure private goods.
100
reduction can be consumed by all countries on the globe, the literature has labelled
climate change mitigation a “global public good” (Ferroni et al., 2002; Perman et al.,
2011; Sandler, 2004).
Supplying global public goods such as climate change mitigation is a challenge. As
shown in Section 3, markets cannot efficiently supply this type of good because they
cannot deal with non-private goods. Policymakers therefore need to intervene and correct
this market failure. But also nations struggle to supply global public goods. No country
can single-handedly solve the problem, as no single country can by itself change the
composition of the world’s atmosphere and mitigate climate change (IPCC, 2001).
Countries therefore need to collaborate in order to mitigate climate change. As we will
see, many of the problems associated with supplying global public goods come from the
fact that the costs and benefits for one country do not only depend on that country’s
actions, but also on the actions – or lack thereof – of other countries. This makes
supplying (global) public goods what is called a “collective action problem” (Sandler,
2004). This type of problems often result in collective action failures that occur when
rational decisions by individual countries result in an inefficient outcome that a single
country cannot change (Sandler, 2004). As we will show, the non-excludability of climate
change mitigation has a devastating implication for the supply of such a good: instead of
supplying the good by reducing GHG emissions, countries might just wait for other
countries to supply the good and then, as they cannot be excluded from receiving the
benefits, benefit from the result. This is a behaviour that the literature calls “free-riding.”
As every country can make the same reasoning, it is quite possible that at the end of the
day no country will actually be reducing its emissions.
Section 4.2 introduces basic theoretical elements of game theory, a theoretical tool that
allows for systematically analysing the problems associated with the strategic interactions
between countries trying to mitigate climate change.
4.2 Basic game theory notions and concepts55
In order to understand why countries struggle to reduce their GHG emissions, or in other
words, why countries struggle to supply the global public good of climate change
mitigation, we have to introduce basic game theory tools. Using these tools will enable
readers to understand the nature of the problem at stake and grasp different important
concepts like free-riding. Game theory is a theoretical apparatus that allows for studying
any form of strategic interaction among individuals, groups, or countries (Varian, 2010).
Widely used to study political negotiations and economic behaviour, it is particularly
useful to analyse the issues involved in collective action problems such as climate change
mitigation. Before we start with the analysis of climate change mitigation, we will
introduce basic game theory notions and concepts that allow for analysing strategic
interactions among countries.56
55
Note that this short review only covers the essential concepts and notions needed for the analysis of
climate change mitigation. Annex 1 provides additional readings on game theory.
56
The presentation of basic notions of game theory in this section draws on Chapter 28 of Varian (2010).
101
Strategic interactions involve at least two individuals or groups called “players.” A priori,
any strategic interaction can involve many players and each player can have a large set of
different strategies. An example of a strategic interaction would be climate change
mitigation negotiations among different countries (the players), with each country
pursuing its own strategy. A country’s strategy could be, for instance, to reduce GHG
emissions or to not reduce GHG emissions. A formal representation of such strategic
interactions is called a “game.” Games are the game theorist’s tool to formally capture
situations in which individuals or groups interact in a context where their welfare not
only depends on their own actions, but also on the actions of others (Mas-Colell et al.,
1995). To introduce the basic game theory notions and concepts we limit ourselves for
now to strategic interactions with only two players. This kind of game can be represented
in what is called a “payoff matrix.” To illustrate how to represent a game in a payoff
matrix, we use an example by Varian (2011: 522).
Suppose you have two players (player A and player B) playing a game. Each player has a
sheet of paper and has to write one out of two words on his or her sheet. Player A can
either write “Top” or “Bottom.” Simultaneously, player B can choose to write “Left” or
“Right.” Player A does not see what player B is writing and vice versa. Each player will
obtain a payoff that depends not only on his or her actions, but also on the actions of the
other player. And each player has been informed about the potential payoffs he or she
could get before playing the game. In the game theory terminology, the players’ options
to play the game are called “strategies.” Thus in this game, each player has two strategies:
player A’s strategies are “Top” and “Bottom,” and player B’s strategies are “Left” and
“Right.”57 After both players have written down their word, the sheets of paper are
examined and the players receive their payoffs. These payoffs are listed in a payoff
matrix. The payoff matrix of this simple game is displayed in Figure 33. It shows the
payoffs to each player for each combination of strategies, and it is known to each player
before they start to play the game.
Actually, this is not completely accurate. Formally, strategies are defined to be a “complete contingent
plan, or decision rule, that specifies how the player will act in every possible distinguishable circumstance
in which she might be called upon to move” (Mas-Colell et al., 1995: 228). However, as the game we are
currently looking at consists of only one move (writing down the word) and the players have to write down
the word simultaneously, and because we exclude the possibility of mixed strategies, the choices “Top” and
“Bottom” are the two strategies of player A, while the choices “Left” or “Right” are the two strategies of
player B.
57
102
Figure 33: Payoff matrix of a two-player game with a dominant strategy
equilibrium
Source: Author's elaboration based on Varian (2010: 523).
As an example, if player A plays the strategy “Top” and player B plays the strategy
“Right,” we examine the upper right field of the payoff matrix that contains the payoffs
for this combination of strategies. If the players play in this way, player A will receive a
payoff of 1 (the first entry in the upper right box), while player B will receive a payoff of
2 (the second entry in the upper right box). Similarly, if player A plays “Bottom” and
player B plays “Left,” we see in the lower left box of the matrix that player A will receive
a payoff of 4 while player B receives a payoff of 2.
Having understood how one reads a payoff matrix, we can now ask ourselves what
happens if these two players actually play this game. Player A has a strong incentive to
always play “Bottom,” because the payoffs from playing “Bottom” (4 and 2) are always
higher compared to the corresponding payoffs of playing “Top” (2 and 1). Player B has a
strong incentive to always play “Left,” as the payoffs from playing “Left” (3 and 2) are
always higher than the corresponding payoffs of playing “Right” (2 and 1). Thus, we
expect that in equilibrium we end up in the lower left box of the matrix, as player A will
always play “Bottom” and player B will always play “Left.”
We have been able to quickly find the solution to this game because each player has what
is called a “dominant strategy,” which is a player’s best strategy regardless of what the
other player plays (Mas-Colell et al., 1995).58 Here, player A’s “Bottom” strategy
dominates his or her “Top” strategy, as the player will get a higher payoff playing
“Bottom” no matter what player B chooses. On the other hand, player B’s dominant
strategy is playing “Left” because no matter what player A plays, he or she will get a
higher payoff if he or she plays “Right.” The resulting equilibrium is therefore called a
“dominant strategy equilibrium.”
As Varian (2010) points out, dominant strategy equilibria are simple to find, but do not
happen often. Let us look at a game with the same rules but a different payoff structure
displayed in Figure 34. As one can see, this game has no dominant strategies. If player A
Note that this kind of strategy is called “strictly dominant” as opposed to “weakly dominant,” but for the
sake of simplicity we will not discuss this difference as it is not needed to understand our main point. For
more information see the additional readings listed in Annex 1.
58
103
chooses “Top,” then player B wants to choose “Left,” but if player A chooses “Bottom,”
then player B wants to choose “Right.” Thus there is no dominant strategy for player B
(or for player A, as you can easily verify) and therefore also no dominant strategy
equilibrium.
Figure 34: Payoff matrix of a two-player game with two Nash equilibria
Source: Author's elaboration based on Varian (2010:524).
In real-life strategic interactions, players frequently do not have dominant strategies.
Game theorists therefore developed alternative concepts that allow for analysing strategic
interactions. One of these concepts, called the Nash equilibrium, was introduced in 1951
by John Nash, an American economist and Nobel Prize laureate.59 Instead of requiring
that a player’s strategy be optimal for all possible choices of the other player, this concept
only requires that a player’s strategy be optimal given optimal strategies of the other
player (Varian, 2010). In other words, in a Nash equilibrium, each player’s strategy
choice is the best strategy that the player can choose, given the strategy that is actually
played by the other player. In that sense, Nash equilibria are weaker solution concepts
than dominant strategy equilibria: each dominant strategy equilibrium is also a Nash
equilibrium, but not each Nash equilibrium is a dominant strategy equilibrium.
In Figure 34, “Top”/”Left” would be such a Nash equilibrium: if player B plays “Left,”
the best strategy player A can choose is to play “Top.” At the same time, if player A
plays “Top,” the best strategy player B can choose is “Left.” Thus, “Top”/”Left” is a
Nash equilibrium.60 Of course, neither of the two players knows what the other player
will play before they make their move, but each player might have some idea about what
the other player’s choice will be. A Nash equilibrium can thus be interpreted as “a pair of
expectations about each person’s choice such that, when the other person’s choice is
revealed, neither individual wants to change his behavior” (Varian, 2010: 525). For
instance, if player A expects player B to choose “Left,” player A will choose “Top,” and
if player B expects player A to choose “Top,” player B will choose “Left” and the game
ends in the Nash equilibrium “Top”/”Left.” From this Nash equilibrium no player would
want to unilaterally change his or her strategy if he or she had the opportunity, as the
The game theory work of Nash was popularized in Ron Howard’s 2001 biographical movie “A Beautiful
Mind.”
60
As you can verify, using similar reasoning, “Bottom”/”Right” is a second Nash equilibrium in this game.
59
104
player could only lose. While the concept of the Nash equilibrium has several problems,61
it is a very useful concept to analyse strategic interactions.
Having covered the basic notions and concepts of game theory, we now have the
necessary tools to analyse why countries struggle to reduce GHG emissions and solve the
global climate change problem. Climate change mitigation involves a strategic interaction
among countries, as a coordinated effort by different countries is needed to mitigate
climate change. Using the game theory tools introduced above, we will analyse this
strategic interaction, showing what goes wrong and highlighting the reasons for the
failure of the international community to mitigate climate change so far.
4.3 Climate change mitigation: A game theory perspective
We now turn our attention towards the analysis of the decade-long failure of the
international community to mitigate climate change. To do so we use game theory tools
to model a strategic interaction among countries trying to supply the global public good
of climate change mitigation.
4.3.1 Climate change mitigation in a world composed of two countries
We start our investigation of this strategic interaction by making a simplifying
assumption. We suppose that the world is composed of two identical countries that try to
mitigate climate change: country A and country B. Each individual country can play two
strategies: it can choose to either reduce its GHG emissions by 20 per cent, or not to
reduce its GHG emissions at all.
If the country chooses not to reduce its emissions, it does not bear any costs but also does
not receive any benefit. If the country chooses to reduce its emissions, it has to pay a cost
of 12 in order to finance the GHG reduction. At the same time, the emission reduction
yields a benefit of 10 for the country, as it reduces the negative effects of climate change.
While these cost and benefit numbers have been chosen arbitrarily, they reflect an
important property of climate change mitigation. As the climate is the product of the
behaviour of all nations on the planet, no single country can change the composition of
the atmosphere alone and mitigate climate change single-handedly (IPCC, 2001). The
benefits of a single country reducing GHG emissions will therefore be relatively small
compared to the costs the country incurs in doing so. Hence, the cost of financing the
61
One of the problems is that it is possible that a game has more than one Nash equilibrium, as illustrated
by the game in Figure 34. Another problem is that some games might not have a Nash equilibrium if one
considers only so-called “pure strategies.” Pure strategies are the ones we have looked at so far and which
consist of choosing an option once and for all. Instead of choosing an option and playing this option with
100 per cent probability, players can also randomize their choices. If they do so, game theorists speak of
“mixed strategies” or “randomized strategies.” When players play mixed strategies they assign probabilities
to each of their choices and play the choices according to the probabilities assigned (Varian, 2010). If one
takes mixed strategies into account, one can show that Nash equilibria exist for a fairly broad range of
games, making the concept very popular (see Chapter 8 in Mas-Colell et al., 1995, for a discussion on this
topic). We will not further discuss mixed strategies in this teaching material as we do not need them to
analyse climate change mitigation.
105
GHG reduction for a single country (12) has been chosen to be bigger than the benefit
(10) the country obtains from single-handedly reducing its emissions.
As mentioned in Section 4.1, climate change mitigation is a global public good, meaning
that the benefits of abating are non-rival and non-excludable. This implies that if country
A chooses to abate – paying the cost of 12 and receiving the benefit of 10 – country B
also receives a benefit of 10 without doing anything whatsoever. This is a crucial point
one has to understand. Country B receives a benefit from the emission reductions made
by country A, because country A’s emission reductions lower global concentrations of
atmospheric emissions. Country A’s efforts thereby slow climate change, which
positively affects country A and country B. This leads to the payoff matrix displayed in
Figure 35.62
Figure 35: Reducing greenhouse gas emissions in a world composed of two countries
Source: Author.
As can be seen in Figure 35, if both countries decide not to reduce their emissions,
nothing will happen: country A and country B both obtain a payoff of 0 as there are
neither mitigation costs nor mitigation benefits associated with inaction. If country A
decides to reduce its emissions while country B decides not to reduce, country A obtains
a payoff of -2 (the benefits it produces (10) minus its costs of doing so (12)) while
country B receives a payoff of 10. Similarly, if country B decides to reduce while country
A decides not to reduce, country B receives a payoff of -2 and country A receives a
payoff of 10. Finally, if both countries decide to reduce, each country obtains a payoff of
8 (country A receives the benefit produced by itself (10) plus the benefit produced by
country B (10) and pays its reduction costs (12); the same goes for country B).
In game theory, this game structure is called a “prisoner’s dilemma” and is probably the best-known
application of game theory. This game was introduced by Merrill Flood and Melvin Dresher in 1952.
Albert Tucker used a fictive interaction between two prisoners as an illustrating anecdote, which gave the
game its name (Poundstone, 1992). Varian (2010: 527) outlines Tucker’s original story as follows: “[T]wo
prisoners who were partners in a crime were being questioned in separate rooms. Each prisoner had a
choice of confessing to the crime, and thereby implicating the other, or denying that he had participated in
the crime. If only one prisoner confessed, then he would go free, and the authorities would throw the book
at the other prisoner, requiring him to spend 6 months in prison. If both prisoners denied being involved,
then both would be held for 1 month on a technicality, and if both prisoners confessed they would both be
held for 3 months.” This game has a similar payoff matrix as in Figure 35, leading both rational prisoners to
confess, while they would have been better off denying the crime.
62
106
Clearly, from the point of view of the whole planet, the best outcome of the game would
be that both countries reduce their emissions and thereby contribute to mitigating climate
change. This would yield an overall benefit of 16 (8+8), which is the highest possible
payoff for the whole planet in the game.63 Thus, “Reduce”/”Reduce” is a so-called “social
optimum.” But in this game, the social optimum is not the likely outcome. Or stated
differently, “Reduce”/”Reduce” is not a Nash equilibrium because the best choice
country A could make given that country B reduces its emissions is not to reduce its
emissions. In that case, country A would receive a payoff of 10, which is greater than the
payoff of 8 the country would obtain by also reducing its emissions. The exact same
reasoning applies for country B. This free-riding behaviour is the main problem
associated with the provision of public goods like climate change mitigation. Actually,
the best payoff a country could hope for in the whole game (10) is obtained if the country
free-rides, that is, if it lets the other country take action and benefits from these actions
without doing anything by itself.
So what is the likely outcome of the game? One can easily see that country A’s dominant
strategy in this game is not to reduce its emissions: country A will be better off by not
reducing, no matter the choice of country B. The same holds true for country B: for both
choices of country A, country B obtains a higher payoff if it does not reduce its
emissions. Hence “Not reduce”/”Not reduce” is the dominant strategy equilibrium and at
the same time also the Nash equilibrium of the game. At this equilibrium, neither country
A nor country B has an incentive to alter its choice unilaterally if given the opportunity. If
a country were to unilaterally alter its choice in the Nash equilibrium, it would obtain a
payoff of -2 instead of 0. The game thus predicts that no country will reduce its
emissions, as both prefer to free-ride. At the same time, the whole planet would clearly be
better off if both countries simultaneously decided to reduce their emissions and thereby
mitigate climate change.
Although the game is very simple, it perfectly illustrates the problem underlying all
collective action failures: “individual rationality is not sufficient for collective rationality.
That is, individuals [or countries] who abide by the tenets of rationality may make
choices from which the collective is left in an inferior position” (Sandler, 2004: 18).
Instead of contributing to mitigating climate change by reducing emissions, countries
have strong incentives to free-ride; climate change will thus not be mitigated.
4.3.2 Climate change mitigation in a world composed of n countries64
So far we have analysed a strategic interaction in a world composed of two countries.
Given that there are considerably more than two countries in the world, we relax this
restrictive assumption and generalize the collective action problem to incorporate any
number of countries.
63
The other options yield planetary payoffs of 0 (0+0), 8 (-2+10), and 8 (10-2), respectively.
This section draws on the presentation of the n country prisoner’s dilemma in Sandler (2004) and Perman
et al. (2011).
64
107
To do so, suppose the planet consists of a total of n identical countries. Each country can
choose between two strategies: to reduce GHG emissions by 20 per cent or not to reduce
GHG emissions. Suppose further that for a given country i, the cost of reducing its GHG
emissions by 20 per cent equals ci. By reducing its greenhouse gases by 20 per cent, the
country produces a non-rival and non-excludable benefit of bi (whereas ci>bi), benefiting
it and each of the other n-1 nations. Table 1 displays the payoffs of this game from the
perspective of country i. The rows show the two strategies of country i while the columns
indicate the actions of the other n-1 countries. The top row displays the free-rider payoffs
for country i: if one other country reduces its emissions, country i receives a benefit of bi,
if two other countries reduce their emissions, country i obtains 2bi , etc. The bottom row
shows the payoffs for country i when it does reduce its emissions: if no other country
contributes, country i obtains a payoff of bi-ci, if one other country contributes, country i
receives 2bi-ci, etc.
Table 5: Reducing greenhouse gas emissions in a world of n countries
Source: Author's elaboration based on Figure 2.2 in Sandler (2004) and Table 9.1 in Perman et al. (2011).
As one can see, even with n countries, country i’s dominant strategy is to free-ride
because the payoffs in the upper row exceed the payoffs in the bottom row by ci-bi, no
matter how many other countries contribute. As the same holds for all other countries, the
unique Nash equilibrium of the game is that nobody reduces its emissions. Given that
nobody has an incentive to take action, the aggregated planetary payoff in the Nash
equilibrium is 0.
The Nash equilibrium is clearly not a socially optimal outcome. The socially optimal
outcome of the game occurs when all countries limit their emissions, leading to a
maximized planetary payoff of n2bi-nci. Again, we want to emphasize that this social
optimum is not a stable solution. Even if all other n-1 countries contribute, country i
always has an incentive to free-ride because it will obtain (n-1)bi, which is greater than
nbi-ci. As the free-rider payoffs are higher for all countries no matter how many other
countries contribute, the world ends up in the unique Nash equilibrium with nobody
reducing its emissions.65
65
To illustrate these results with numbers, assume that the costs and benefits are the same as in the twocountry example (ci = 12 and bi = 10) and assume further that the world is composed of the 194 United
Nations member countries (n = 194). Then, the planetary payoff in the social optimum would equal
1942*10-194*12=374,032, with each individual country obtaining a payoff of 194*10-12=1,928. But the
socially optimal outcome is not an equilibrium, because nation i (and thus also all other nations) would
have an incentive to free-ride (the country’s payoff would be greater if it free-rides and all the other n-1
108
Hence, the modified game structure does not result in a different prediction. Even in a
world composed of many countries, rational countries would still free-ride and not
contribute to mitigating climate change. While the game theory approach reveals the
essential nature of the problem (free-riding), things are even more complicated in reality.
Countries are not identical as we assumed in the games above: they vary in terms of size,
wealth and technology. Costs and benefits of mitigation also differ widely among
countries and are only imperfectly known (Common and Stagl, 2004; Perman et al.,
2011; Sandler, 2004; Stern, 2007). Countries are not equally affected by the
consequences of climate change, thus they would not equally benefit from mitigation
efforts. Mitigation costs and opportunities also differ between countries. Finally, there is
the issue of international and intergenerational justice affecting negotiations about the
financing of climate change mitigation (see the discussion on the principle of common
but differentiated responsibilities in Module 4). All these factors further complicate the
collective action problem and influence the likely outcome (or prognoses) of strategic
interactions.66
In addition to factors related to the global public good nature of climate change
mitigation and their game theory implications, there are other factors that can influence
the outcome of collective action. Box 14 illustrates this issue by showing the vastly
different outcomes of two, at first sight similar, collective action problems – mitigation of
ozone-shield depletion and mitigation of climate change.
Box 14: Different collective action outcomes for identical collective action problems
Sandler (2004) highlights that seemingly identical collective action problems like the provision of
a global public good can have vastly different collective action outcomes. To illustrate his point,
he compares the collective action outcome of two pollution problems. One the one hand, the
international community has been rather efficient in mitigating stratospheric ozone shield
depletion caused by chlorofluorocarbons (CFCs) and bromide-based substances. On the other
hand, it is still struggling with mitigating climate change caused by the accumulation of
greenhouse gases in the atmosphere. Mitigating CFCs and mitigating greenhouse gases are both
global public goods. Moreover, both problems are pollution problems and their solution requires
international cooperation to reduce emissions. In the CFC case, humankind succeeded, while
efforts to reduce GHG emissions have not been successful to date. Sandler explains this
difference in the outcomes of collective action by the differences listed in
Table 6. All these factors make finding a collective action solution more difficult in the case of
climate change as compared to ozone-shield depletion.
countries contribute: (194-1)*10=1,930>1,928). Thus the world ends up in the unique Nash equilibrium
with a planetary payoff of 0, which is clearly less than the social optimum.
66
More complicated game structures can incorporate some of these additional factors. See Annex 1 for
additional reading material on this issue.
109
Table 6: Factors affecting outcomes of collective actions
Source: Table 10.4 in Sandler (2004), updated by the author.
Consequently, as Sandler (2004: 213) states, “the contrast between global collective action and
inaction hinges on factors that go beyond the non-rivalry and non-excludability of these global
public good’s benefits…. Knowledge of just the properties of a public good is not always
sufficient to provide a prognosis for collective action.”
Source: Author's elaboration based on Sandler (2004).
4.3.3 Climate change mitigation: Lessons from game theory models
Section 4 started by asking how economic theory explains the fact that the international
community has been struggling for decades to solve the climate change problem, even
though the apparently simple solution has been known for a long time. Sections 4.1, 4.3.1
and 4.3.2 provided the theoretical answer to this question. The two games modelling the
strategic interaction between countries showed that, due to the global public good nature
of mitigating climate change, the payoff countries obtain from free-riding is higher than
the payoff they would obtain from reducing their emissions. Rationally behaving
countries thus have strong incentives to free-ride and not reduce their emissions. This
results in a massive collective action failure.
110
In addition to explaining this outcome, game theory also provides certain insights on how
this outcome might be changed. Clearly, payoffs from acting alone matter. If they were
positive, each country would have an incentive to act alone and reduce emissions. This
would lead to a situation where the Nash equilibrium corresponds to the social optimum.
Stated differently, if the payoffs from acting alone were positive, the social optimum
would be incentive-compatible: individual countries would have incentives to reduce
emissions single-handedly because of the positive payoff, leading automatically to a
socially optimal situation.
Such an outcome could be reached through a binding international agreement between
countries (Common and Stagl, 2004; Perman et al., 2011; Sandler, 2004; Tietenberg and
Lewis, 2012). The countries would have to agree that they all reduce their emissions.
However, as each country has a strong incentive to sign on to such a treaty and
subsequently free-ride to maximize its payoff, the agreement would need to contain
penalties. In theory, any penalty larger than ci-bi would do the job. With this penalty, the
payoff of acting alone (bi-ci) would be larger than the payoff of not acting (0-p, whereas
p>ci-bi is the penalty) and the Nash equilibrium would correspond to the social optimum.
However, putting in place such a penalty system is not straightforward, as the history of
climate change negotiations has shown (see Module 4).67 At least equally difficult would
be enforcement of such a penalty system. Countries could only be forced to pay penalties
if there were a third party with the power to force countries to pay (Perman et al., 2011).
As today’s world consists of sovereign states, no such enforcer exists on the global level.
International environmental agreements therefore rely on voluntary actions of countries
and should thus be designed in a self-enforcing way. The theoretical literature (see
Perman et al., 2011, for a short review) on international environmental agreements
consequently offers rather pessimistic views regarding their efficiency: treaties seem to
largely codify actions countries intended to take anyway, and achieve little when the
number of affected countries is large. Moreover, as climate change mitigation illustrates,
effective cooperation seems hardest to achieve when the stakes are high and when the
cooperation is most needed (Barret, 1994). Reviews of international environmental
agreements show that effective cooperation requires that (a) the treaty yield positive net
benefits for all participating countries, (b) the parties reach a consensus on the design of
the treaty, and, most importantly, (c) the treaty can be enforced by the participating
countries (Finus, 2003).
Besides self-enforcing international agreements, game theory suggests a number of
alternative mechanisms to enforce cooperation.68 Cooperation can be fostered if countries
are able to make credible commitments to reduce emissions regardless of what other
countries are doing. Transfers and other side payments could increase the number of
cooperating countries. Furthermore, the co-benefits of emission reductions, such as those
achieved by linking development and climate change policy issues, might alter the
payoffs for countries and thus change the nature of the game. Moreover, theory indicates
that the prospects for cooperation are less pessimistic when games are repeated. The two
games we introduced above are “one-shot games” played only once. If the same game is
67
68
Putting in place a penalty system is itself a pure public good problem (Sandler, 2004).
See Chapter 9.3 in Perman et al. (2011) for an overview.
111
played repeatedly over an undefined time horizon, and if countries communicate, a
variety of different options emerge. As will be seen in Module 4, the Paris Agreement has
been partially built on these mechanisms.
Short summary
How does economic theory explain why the international community has been struggling
for decades to solve the climate change problem, even though the apparently simple
solution has been known for a long time? To answer this question, Section 4 showed that
climate change mitigation is a global public good with non-rival and non-excludable
benefits. As no country is able to change the composition of the atmosphere singlehandedly, international cooperation is needed. The section introduced game theory
concepts that were subsequently used to analyse the strategic interaction between
countries. Two games illustrated the main cause of the collective action failure: that
countries have strong incentives to free-ride and not contribute to emission reductions
because the benefits of mitigating climate change are non-rival and non-excludable. This
leads to an outcome that is not socially optimal. The section concluded by drawing
several lessons from game theory models that would allow for fostering cooperation
between countries.
5 Exercises and questions for discussion
1. Define Pareto efficiency and discuss the advantages and disadvantages of the concept.
2. Describe the first welfare theorem. Why is this theorem a fundamental result in
economics? What assumptions underlying the first welfare theorem are violated in the
case of climate change?
3. What criteria are used to classify goods as private and public goods? List two
examples of a private and a public good and discuss why these goods are considered
private/public by using the criteria you defined above.
4. What are open-access goods? Why is the atmosphere considered to be an open-access
good? What happens if markets deal with open-access goods such as the atmosphere?
5. Define externalities and provide an example of a positive and a negative externality.
How could governments solve a negative externality problem? Illustrate your
reasoning using the example of GHG emissions.
6. Consider a coal plant that releases CO2 emissions into the atmosphere while
producing units of energy. Suppose that the private marginal cost of the plant is given
by PMC = 50+0.25*Q, where PMC is the private marginal cost in US dollars per unit
produced and Q is units of energy produced. Because the firm releases CO2 emissions
into the atmosphere, it imposes an external cost on society that equals US$2 per unit
of energy produced. Suppose further that the private marginal benefit (PMC) – which
112
equals the social marginal benefit (SMC) – per unit of energy produced is given by
PMB = SMB = 100-0.25*Q.
(a) Derive the social marginal cost (SMC) function in US dollars per unit
produced.
(b) Draw a diagram illustrating the private marginal cost function, the social
marginal cost function, and the private marginal benefit function.
(c) Find the profit-maximizing energy output of the coal plant and the
corresponding price per unit of energy produced.
(d) Find the socially optimal energy output and the corresponding price.
(e) Compare the private equilibrium with the socially optimal outcome and
discuss underlying externality issues.
7. Suppose player A and player B play a game with the following payoff matrix:
Player B
Left
Right
Top
1,1
-1,-1
Bottom
-1,-1
1,1
Player A
(a) Define dominant strategies. Does player A have a dominant strategy? Does
player B have a dominant strategy?
(b) How many pure-strategy Nash equilibria can you find in the game?
(c) Can you find a real-world example that could be described by such a payoff
matrix?
8. How does economic theory explain why the international community has been
struggling for decades to solve the climate change problem? To analyse this question,
suppose that the world is composed of two countries and discuss the payoff matrix of
the mitigation game. What is the likely outcome of this game? Explain the lessons
that can be learned from this simple game.
113
Annex 1
114
References
Arrow K (1951). An extension of the basic theorems of welfare economics. In:
Proceedings of the Second Berkeley Symposium of Mathematical Statistics and
Probability, Neyman J, ed. University of California Press. Berkeley and Los Angeles.
Barrett S (1994). Self-enforcing international environmental agreements. Oxford
Economic Papers 46: 874–94.
Cohen D (2009). La prospérité du vice. Une Introduction (inquiète) à l’économie. Albin
Michel. Paris.
Common M, and Stagl S (2004). Ecological Economics – An Introduction. Cambridge
University Press. Cambridge, UK.
Ferroni M, Mody A, Morrissey O, Willem te Velde D, Hewitt A, Barrett S, and Sandler T
(2002). International public goods: incentives, measurement, and financing. World Bank.
Washington, DC.
Finus M (2003). Stability and design of international environmental agreements: The case
of transboundary pollution. In: The International Yearbook of Environmental and
Resource Economics 2003/2004: A Survey of Current Issues, Folmer H and Tietenberg T,
eds. Cheltenham: Edward Elgar: 82–158.
IPCC (2001). Climate Change 2001: Mitigation of Climate Change. Contribution of
Working Group III to the Intergovernmental Panel on Climate Change Third Assessment
Report (TAR). Cambridge University Press. Cambridge, UK and New York.
IPCC (2014a). Climate Change 2014: Synthesis Report. Contribution of Working Groups
I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate
Change, Pachauri RK, and Meyer LA, eds. Cambridge University Press. Cambridge, UK
and New York.
IPCC (2014b) Climate Change 2014: Mitigation of Climate Change. Contribution of
Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, Edenhofer O, Pichs-Madruga R, Sokona Y, Farahani E, Kadner S,
Seyboth K, Adler A, Baum I, Brunner S, Eickemeier P, Kriemann K, Savolainen J,
Schlömer S, von Stechow C, Zwickel T, and Minx JC, eds. Cambridge University Press.
Cambridge, UK, and New York.
Kolstad CD (2000). Environmental Economics. Oxford University Press. New York.
Lerner AP (1934). The concept of monopoly and the measurement of monopoly power.
Review of Economic Studies 1: 157–75.
Lerner AP (1944). The Economics of Control. Principles of Welfare Economics.
Macmillan. New York.
115
Mas-Colell A, Whinston MD, and Green JR (1995). Microeconomic Theory. Oxford
University Press. New York.
Nash J (1951). Non-cooperative games. The Annals of Mathematics 54(2): 286–95.
Olson M (1965). The Logic of Collective Action: Public Goods and the Theory of
Groups. Harvard University Press. Cambridge, MA.
Ostrom E (2008). The challenge of common-pool resources. Environment 50: 9–20.
Perman R, Ma Y, Common M, Maddison D, and McGilvray J (2011). Natural Resource
and Environmental Economics, Fourth Edition. Pearson Education Limited. Essex, UK.
Pindyck R, and Rubinfeld D (2012). Microeconomics, Eighth Edition. Pearson Series in
Economics. Prentice Hall. Upper Saddle River, NJ.
Poundstone W (1992). Prisoner’s Dilemma. Doubleday. New York.
Samuelson PA (1954). The pure theory of public expenditure. Review of Economics and
Statistics 36: 387–89.
Sandler T (2004). Global Collective Action. Cambridge University Press. New York.
Sen A (1970). Collective Choice and Social Welfare. Holden Day. San Francisco.
Stern N (2007). The Economics of Climate Change. The Stern Review. Cambridge
University Press. New York.
Straffin P (2004). Game Theory and Strategy, Fifth Edition. The Mathematical
Association of America. Washington, DC.
Tietenberg T, and Lewis L (2012). Environmental and Natural Resource Economics,
Ninth Edition. Pearson International Edition. Addison Wesley. Boston.
Varian HR (2010). Intermediate Microeconomics: A Modern Approach, Ninth Edition.
W.W. Norton & Company. New York.
116
Module 4
The politics of climate change – towards a low-carbon world
1 Introduction
Modules 1 and 2 outlined the links between the environment and the economy, explained
how human activities change the climate, and discussed the observed and anticipated
impacts of climate change. Module 3 showed that climate change is the result of a large
market failure that can only be corrected by policy interventions at the global level. Such
a correction is not easy to achieve, however, because the world is composed of sovereign
countries. These countries need to coordinate their efforts to transform their economies
into low-carbon economies and thus limit greenhouse gas (GHG) emissions. However, as
shown in Module 3, such coordination represents a collective action problem that is
difficult to overcome. For the past 25 years, the international community has not found a
convincing solution, as global GHG emissions continue to increase. Nevertheless, at the
United Nations Climate Change Conference of the Parties in Paris (COP21) in 2015, the
international climate change policy framework took a new and promising direction. This
module provides an in-depth analysis of climate change policies and the international
politics of climate change.69
Section 2 discusses the different policy instruments and technological solutions that could
enable us to limit climate change. Section 2.1 focuses on policy measures that aim to
stabilize the concentrations of greenhouse gases in the atmosphere. Such stabilization
requires either reducing the flow of emissions reaching the atmosphere, or extracting the
already-emitted greenhouse gases from the atmosphere. Section 2.1.1 discusses policies
that aim to directly or indirectly decarbonize our economic systems by pricing CO2,
increasing energy efficiency, and substituting fossil fuels with alternative sources of
energy.70 Section 2.1.2 explains how policy could promote technological solutions to help
capture and store already-emitted CO2 emissions before they reach the atmosphere.
Section 2.1.3 discusses the role of policies in promoting carbon geoengineering, a
technology that could directly extract CO2 out of the atmosphere. Unlike the other policy
options to stabilize GHG concentrations, carbon geoengineering would not require
transforming economies into low-carbon economies. After having discussed the different
policy and technological options to stabilize GHG concentrations, Section 2.2 focuses on
solar geoengineering, a technology that allows for reflecting additional incoming solar
radiation back into space and thus stabilizing temperatures on the planet. While most of
the policy measures discussed in Section 2 are subject to the collective action problem
explained in Module 3, carbon geoengineering and solar geoengineering are not: they
could be implemented by only a few nations. However, these technologies are not yet
fully developed, their implementation may entail considerable environmental risks, and
carbon geoengineering is currently expected to be very expensive.
69
To do so, it relies on, among other sources, the work of the Intergovernmental Panel on Climate Change
(IPCC) and on Barrett et al. (2015), who present a collection of studies and reviews on climate change
policy.
70
While the section focuses on policies capable of reducing CO 2, note that many of the policies discussed
simultaneously also reduce other energy-related greenhouse gases like nitrous oxide.
117
In parallel with policies aimed at limiting climate change, which are discussed in Section
2, societies also need to undertake actions to adapt human and natural systems so that
they are better prepared for the anticipated impacts of climate change. These policies are
described in Section 3. Adaptation needs differ from one place to another because the
anticipated impacts of climate change differ locally. The first step of any comprehensive
adaptation policy thus should be to assess local risks and vulnerabilities in order to
identify adaptation needs. Only then can specific adaptation options be selected and
implemented. The section also presents a schematic overview of adaptation policies
drawing on the IPCC’s classification of adaptation options, and provides examples of
adaptation measures implemented by developing countries.
After having described the different climate change limitation and adaptation policy
options, Section 4 focuses on the international climate change policy architecture. The
section reviews key international climate change policy developments and explains that
of the policy instruments outlined in Sections 2 and 3, international climate change policy
negotiations have so far focused mostly on those policy instruments that aim to limit the
net flow of emissions. As explained in Module 3, these policy instruments are subject to a
severe collective action problem. The section shows that up until now, international
policy negotiations have been based on a top-down approach that has been unable to
overcome this collective action problem. At COP21, the international climate change
policy framework considerably changed with the adoption of the Paris Agreement.
Unlike the past 25 years of international climate change policy, the Paris Agreement is
based on a bottom-up approach. The section ends by discussing several key issues of the
Paris Agreement, with a particular focus on the situation of developing countries.
At the end of this module, readers should be able to:







Distinguish the two fundamental options to limit climate change;
Describe different policy instruments to stabilize atmospheric GHG
concentrations;
Analyse advantages and disadvantages of the different policy instruments to
stabilize atmospheric GHG concentrations;
Understand how geoengineering could limit climate change;
Evaluate the importance of climate adaptation policies;
Describe the international climate change policy architecture;
Evaluate the links between the Paris Agreement and development.
To support the learning process, readers will find several exercises and discussion
questions in Section 5 covering the issues introduced in Module 4. Useful data sources
and additional reading material can be found in Annexes 1 and 2.
2 Policy options and technological solutions to limit climate change
As explained in Section 4 of Module 2, the IPCC anticipates that, compared to the preindustrial era, mean surface temperature on the planet will increase by roughly 4°C by the
end of the 21st century if the world does not implement additional policies to limit
118
climate change. This human-induced temperature increase is expected to trigger “severe,
pervasive and irreversible impacts on people and ecosystems” (Field et al., 2015: 62; see
also Module 2). If climate change policies could limit the warming to less than 2°C above
pre-industrial temperatures, these risks would be substantially reduced (IPCC 2014a,
2014b). Stabilizing temperatures at less than 2°C above pre-industrial levels is thus a
crucial policy objective that is currently on the agenda of almost all countries (see Section
4).
To limit warming to 2°C and thereby avoid the main bulk of the negative effects of
climate change, humankind has essentially two options (Edenhofer et al., 2015). The first
is to stabilize concentrations of greenhouse gases in the atmosphere (see Section 2.1). The
second is to offset the expected temperature increase by increasing the amount of
reflected incoming solar radiation using solar geoengineering technologies (see Section
2.2). The following sections discuss different policy instruments that could be used to
implement both these options.
2.1 Policy instruments and technologies to stabilize concentrations of greenhouse
gases in the atmosphere
IPCC (2014b) estimates that the atmospheric concentration of greenhouse gases in 2100
must be stabilized at below 450 parts per million (ppm) CO2-equivalent (CO2-eq)71 if the
increase in mean surface temperatures is to stay below 2°C. GHG concentrations can be
stabilized by either reducing the flow of GHG emissions towards the atmosphere or by
removing those gases from the atmosphere.
While all four major greenhouse gases contribute to the warming of the planet, stabilizing
CO2 concentration is of crucial importance, because climate perturbations from
accumulated fossil-fuel-based CO2 emissions last for thousands of years (Archer et al.,
2009). Recall as well that the fifth IPCC assessment report found an almost linear
relationship between total cumulative CO2 emissions emitted since the start of the
industrial era and global mean surface warming (see Section 4 of Module 2). This implies
that the larger the total sum of emitted CO2 emissions (and hence the larger atmospheric
CO2 concentrations), the higher the mean temperature will be in the 21st century. Due to
the importance of CO2 in limiting climate change, this section mostly focuses on policies
to stabilize the concentration of CO2 in the atmosphere.72
CO2 concentrations can be stabilized in three complementary ways (Barrett and MorenoCruz, 2015): (a) gradually reducing anthropogenic CO2 emissions to zero (see Section
2.1.1); (b) capturing and storing CO2 emissions before they reach the atmosphere (see
Section 2.1.2); and/or (c) reducing concentrations of CO2 in the atmosphere through
direct removal of CO2 from the atmosphere (see Section 2.1.3).
71
CO2-equivalent concentration is a measure to compare radiative forcing of a mix of different greenhouse
gases. It indicates the concentration of CO2 that would cause the same radiative forcing as a given mixture
of CO2 and other greenhouse gases (IPCC, 2014c).
72
Note, however, that several of these policies simultaneously address emissions of other greenhouse gases
resulting from fossil fuel combustion.
119
The first two options affect the flow of emissions to the atmosphere by either reducing
the amount of emitted CO2 or by preventing the emitted CO2 from reaching the
atmosphere. Policy instruments available for this purpose are listed in Table 7 and further
discussed in Sections 2.1.1 and 2.1.2. To meet the 2°C target with a probability of at least
66 per cent – taking all other human influences on the climate into account - IPCC
(2014b) estimates that the above policies need to keep total cumulative CO2 emissions
since the start of the industrial era below 2,900 gigatonnes of CO2 (GtCO2) by 2100. By
2014, roughly 2,000 GtCO2 had already been emitted. Hence, if humans intend to reach
the 2°C target, they have roughly 900 GtCO2 left to emit in the future, all else being
equal. This does not leave too much space for further emissions, as 900 GtCO2 is
equivalent to less than 25 years of emissions at the 2014 level (Edenhofer et al., 2015).
Policies aimed at emission reduction and emission capture and storage thus jointly need
to limit future CO2 emissions to 900 GtCO2. Unfortunately, these policy options are
subject to the collective action problem discussed in Module 3. This makes the prospects
for transforming current carbon-based economies into low-carbon economies and thereby
stabilizing atmospheric CO2 concentrations below the 450 ppm CO2-eq concentration
level relatively bleak.
The third option is different. It does not reduce the flow of emissions towards the
atmosphere as the first two options do, but lowers atmospheric CO2 concentrations by
directly removing the already emitted CO2 from the atmosphere. As such, this third
option – called carbon geoengineering – has the potential to stabilize CO2 concentrations
even if humans continue to emit CO2 emissions as they do today. The most promising
carbon geoengineering technology – industrial air capture (see Table 7) – will be
discussed in Section 2.1.3. As we will see, collective action prospects are entirely
different for this option.
Table 7: Selected policy options to stabilize concentrations of carbon dioxide in the
atmosphere
Source: Author.
120
2.1.1 Policy instruments to reduce anthropogenic carbon dioxide emissions
We first look at policy instruments capable of reducing CO2 emissions that have been at
the centre of the international policy debate on climate change mitigation (see Section 4)
and have been well known for several decades (see also Module 3). Generally speaking,
there are three policy instrument categories directed towards reducing anthropogenic CO2
emissions. The first contains policy instruments that put a price on carbon dioxide, the
second consists of policy instruments that conserve energy, and the third contains policy
instruments to directly decarbonize the energy system by substituting energy from fossil
fuels with alternative energy sources.
2.1.1.1 Policy instruments that price carbon dioxide emissions
Module 3 showed that GHG emissions are a typical example of a negative externality.
Economic actors do not directly bear the climate-change-related costs associated with the
emissions, which they can dump free of charge into the atmosphere. Consequently, they
produce too much GHG emissions. This externality problem can be partly corrected by
putting a price on carbon. If economic actors have to pay a price for emitting CO2, they
directly bear (part) of the climate change-related costs associated with their emissions. As
emitting CO2 would then no longer be free of charge, economic actors would emit less
emissions compared to a scenario without a price on carbon. Pricing carbon would reduce
emissions not only directly by affecting the decisions of economic actors, but also
indirectly by making investments in cleaner technologies and research and development
(R&D) of new technologies more attractive. It is thus not surprising that the necessity of
putting a price on carbon has been clear for a long time (Barrett et al., 2015). Several
policy instruments can be used to price carbon and thereby help decarbonize economies.
The most important ones are carbon taxes, cap and trade systems, removal of fossil fuel
subsidies, and fuel taxation.
Carbon taxes directly tax the amount of CO2 emitted. The level of the tax is often
expressed per tonne of CO2 (e.g. US$50 per tonne of CO2). Carbon can be taxed at
various points of the energy supply chain. Usually one distinguishes between so-called
upstream and downstream carbon taxes, although combinations of the two approaches are
also possible (Wang and Murisic, 2015). Upstream carbon taxes impose a charge on
producers or importers of raw materials such as coal, oil, or natural gas that contain CO2
(Mansur, 2012). The levied charge is designed to be proportional to the amount of carbon
contained in the raw material. As upstream carbon taxes directly tax fossil fuel producers
or importers, the levies are subsequently passed forward and affect the economy-wide
prices of electricity, petroleum products, and energy-intensive goods. Downstream
carbon taxes impose a charge on direct sources of CO2 emissions such as cars, power
plants, energy-intensive industries, etc. (Mansur, 2012). The charge levied is proportional
to the amount of CO2 emitted by the source of the emission. A fully inclusive downstream
carbon tax thus also affects all prices of commodities whose production or consumption
processes involve the release of CO2 emissions. One of the differences between the two
types of carbon taxes is the amount of affected agents and hence the administrative
simplicity: upstream carbon taxes are often imposed on relatively few firms (the fossil
121
fuel producers and importers), while downstream carbon taxes, if they are fully inclusive,
affect many different agents.
Economists consider carbon taxes the most cost-efficient policy instrument to reduce CO2
emissions. In their recent review on different carbon pricing policy instruments, Sterner
and Köhlin (2015: 252) note that a carbon tax is “generally more efficient than direct
regulation of technology, products, and behaviour, as it affects consumption and
production levels as well as technologies, it covers all industries and production and
provides dynamic incentives for innovation and further emissions reductions. In addition,
the tax revenue can be used to facilitate the transition toward renewable energy, cover
administrative and implementation costs, or lower taxes on labour.… Furthermore, a tax
is easy to incorporate in the existing administration.” Besides these direct effects, carbon
taxes can also generate considerable national co-benefits. One example of such a cobenefit of carbon taxes is improvement in health: carbon taxes decrease the use of coal
and thus have the potential to reduce deaths related to air pollution (Parry et al., 2014).
While carbon taxes are an appealing policy instrument from an economic point of view,
they face severe political hurdles at the national level. Among them are (a) the presence
of strong fossil fuel lobbies actively opposing implementation of carbon taxes; (b)
pressure from the public to not implement carbon taxes in order to avoid increases in
commodity prices; (c) a general perception that taxes reduce consumption and production
and thereby induce welfare losses; (d) the fact that carbon taxes, unlike other policy
instruments, clearly identify winners and losers of the policy, leading to stronger
opposition; and (e) potential institutional preferences favouring other carbon pricing
instruments (Sterner and Köhlin, 2015).
Despite these political hurdles, some countries have already implemented carbon taxes.
Figure 36 presents an overview of countries that have implemented carbon taxes or cap
and trade systems. Following Finland in 1990, several Northern European countries
implemented carbon taxes in the early 1990s. By 2016, 16 countries and one subnational
entity (the Canadian province of British Columbia) had implemented a carbon tax.73
While most of these countries are developed ones, two developing countries (Mexico and
South Africa) have implemented a carbon tax and a third one (Chile) plans to implement
such a tax in 2017. However, only eight countries (Sweden, Finland, Switzerland,
Norway, Denmark, Ireland, Slovenia, and France) and the province of British Columbia
have a tax rate that is higher than US$10 per tonne of CO2 (World Bank, 2015). To put
this figure into perspective, most simulations suggest that a global average carbon price
of US$80 to US$120 per tonne of CO2 would be consistent with the 2°C target (World
Bank, 2015). Sweden is currently the only country with a carbon price in this range, as it
imposes a carbon tax of US$130 per tonne. This tax is by far the highest in the world –
the second highest carbon price (US$65 per tonne) is in Switzerland. Hence, besides the
fact that carbon taxes are only imposed on a fraction of global CO2 emissions, there is
also a large gap between the required and the actual price of carbon. Having said that, the
73
Finland, Poland, Sweden, Norway, Denmark, Latvia, Slovenia, Estonia, Switzerland, Ireland, Iceland,
Japan, France, Mexico, Portugal, and South Africa. Note that a 17th country (Chile) plans to implement a
carbon tax in 2017.
122
Swedish experience does clearly illustrate that carbon taxes are effective in reducing CO2
emissions, as shown in Box 15.
Figure 36: Overview of implemented or scheduled carbon pricing policy
instruments
Source: World Bank (2015: 11).
Note: The Regional Greenhouse Gas Initiative (RGGI) is a cooperative cap and trade effort among the US
states of Connecticut, Delaware, Maine, Maryland, Massachusetts, New Hampshire, New York, Rhode
Island, and Vermont.
123
Box 15: The results of the Swedish carbon tax
Sweden’s carbon tax of US$130 per tonne (as of 1 August 2015) is the highest carbon tax
worldwide to date. It particularly affects the Swedish transport sector, as the country taxes
gasoline and diesel strictly in proportion to carbon emissions. It also applies to commercial use,
residential heating, and partly to the industrial sector, which is, however, subject to a reduced tax
rate (Sterner and Köhlin, 2015).
Given that Sweden also uses other climate change policies, it is difficult to precisely isolate the
impact of the Swedish carbon tax on emissions. Nevertheless, Sterner and Köhlin (2015) note that
the tax has achieved a cost-effective and efficient emission reduction over recent decades and has
enabled the country to meet the commitments made in the Kyoto Protocol. Since 1990, Swedish
CO2 emissions have declined by 22 per cent while the country’s economy has been growing.
Sterner and Köhlin thus note that the country has successfully managed to decouple domestic
CO2 emissions and economic growth and thus made strides towards a low carbon-economy. As
illustrated in Figure 37, Sweden’s CO2 emissions per US dollar of GDP are about one-third of the
world average and have been constantly declining for more than 40 years. Note, however, that
this figure does not account for potential carbon leakage effects of the Swedish carbon tax
(relocation of polluting industries towards countries with less stringent climate change policies
and hence lower costs of emitting CO2).
Figure 37: Carbon dioxide emissions per unit of GDP
Source: Sterner and Köhlin (2015: 254).
Source: Author's elaboration based on Sterner and Köhlin (2015).
Note: OECD: Organisation for Economic Co-operation and Development.
A second policy instrument to price CO2 emissions is known as cap and trade (CAT)
systems (also called emissions trading systems - ETS). Unlike carbon taxes, CAT
systems do not directly regulate the price of CO2, but rather its quantity. Regulators
decide on an upper limit of emissions that a country, region, or specific industry can emit
(hence the “cap” in “cap and trade”), and then allocate permits to emit CO2 emissions to
different firms. Each permit allows the firm to emit a certain quantity (often a tonne) of
CO2. Each firm initially receives a certain quantity of permits, allowing it to emit a
124
certain quantity of emissions. If the quantity emitted is smaller than the quantity of
permits the firm possesses, the firm can sell its excess permits. If the quantity emitted is
larger than the quantity of permits the firm possesses, the firm has to buy additional
permits. In this way, a CAP system creates a market for CO2 emissions (hence the “trade”
in “cap and trade”). The basic idea is to generate incentives for firms to reduce their
emissions, as they can sell excess permits and make money out of emitting less. At the
same time, the system penalizes firms that pollute too much, as they have to spend money
to buy additional permits.
Regulators can allocate permits through auctions, free allocation, or a hybrid system that
combines free allocation and auctions (Sterner and Köhlin, 2015). If the regulating
authority decides to allocate permits by auction, it can generate extra revenue for the
government and does not need to devise a specific mechanism for the allocation of
permits to firms. If permits are allocated freely, no government revenue is raised, as firms
do not have to pay to obtain the initial permits (“free allocation”). In that case, the
regulating authority must determine how to distribute the permits. It can, for instance,
allocate permits proportional to firm output or based on historical emission levels of
firms. Regulators can adapt the cap over time, lowering the total quantity of emissions
allowed to be emitted by issuing fewer permits, and thereby tighten the CAT system.
While CAT systems can in theory achieve any amount of emission reduction in a costeffective way, they face considerable political obstacles, making stringent
implementation difficult. Industry lobbying and fears that carbon prices might skyrocket
and slow economic development often lead to a lax design of CAT systems with
extremely high caps, resulting in only few emission reductions.
Several CAT systems are currently in place and more are scheduled for implementation
(see Figure 36). The People's Republic of China, for instance, has already implemented
several regional pilot CAT systems. Based on these pilots, it is planning to launch a
national CAT system in 2017 that is expected to cover major industry and energy sectors
(World Bank, 2015). The largest CAT system currently in place is the European Union’s
(EU) ETS, which was launched in 2005 and covers approximately 45 per cent of the
EU’s CO2 emissions (Sterner and Köhlin, 2005). The success of the programme has so far
been mixed at best: most studies find evidence of only small if any emission reductions
(Anderson and Di Maria, 2010; Bel and Joseph, 2015; Ellerman et al., 2010; Georgiev et
al., 2011). The main reason for these mixed results lies in the design of the European
system, which has led to issuance of too many permits during the first two phases of the
programme (2005–2007 and 2008–2012). This oversupply of emission allowances
resulted in an extremely low price of CO2 emissions, reaching even zero euros in 2007.
The third phase of the programme, which started in 2013, aimed to correct these initial
design flaws. The EU replaced the previous system of national caps with a single EUwide cap that will be reduced annually by 1.75 per cent, made auctioning (instead of free
allocation) the default rule to allocate permits, and included more sectors and gases in the
system (European Commission, 2015). The European Commission is confident that due
125
to these design changes, the system will achieve substantial emission reductions in the
years to come.74
Carbon taxes and CAP systems are the main policy instruments to directly price CO2
emissions. Together, these instruments were applied in roughly 40 countries (including
10 developing and transition countries) and 20 subnational regions in 2015 (World Bank,
2015; see also Figure 36). In many cases both instruments were applied in a
complementary way in the same country. The instruments put a price on roughly half of
the emissions emitted in these countries and regions, which represent about 12 per cent of
global emissions (World Bank, 2015). As long as the international carbon pricing
landscape remains this fragmented, carbon leakage may be a major problem for all
carbon pricing policy instruments (see Box 16). Based on (rather spare) existing
empirical evidence, however, the World Bank (2015) notes that carbon leakage has so far
not materialized on a significant scale and that it affects mainly sectors that are both
emissions- and trade-intensive.
Box 16: Carbon leakage, competitiveness, and carbon border tax adjustments
Domestic carbon pricing is a major policy instrument, allowing countries to reduce their
domestic CO2 emissions. In a world marked by a fragmented carbon pricing landscape, costs of
emitting CO2 differ widely among countries. This cost difference raises two major concerns
(Flannery, 2016). First, countries that put a high price on CO2 worry about the international
competitiveness of their firms, as high carbon taxes are assumed to increase relative production
costs and hence relative prices (especially in energy-intensive sectors). Second, countries
worry that the impact their climate change policies have on global emissions might be limited
by carbon leakage. Carbon leakage refers to the possibility that firms relocate their production
or future investments towards countries with less stringent climate change policies and thereby
shift their emissions to these countries.
While carbon leakage does not yet seem to have materialized on a significant scale (World
Bank, 2015), and while results from empirical economic analysis suggest that the overall
macroeconomic impact on competitiveness, investment, and employment is small and
statistically insignificant compared to other factors (Aldy, 2016), policymakers worry about
these two possible effects. One possibility to mitigate them is so-called carbon border
adjustments (BAs). Kortum and Weisbach (2016: 2) define border adjustments as “taxes or
other prices on imports and rebates on exports based on ‘embedded carbon,’ the additional
emissions of carbon dioxide caused by production of a good. For imports, they can be thought
of as the carbon tax that would have been imposed had the good been produced domestically
(but using the production process and fuel that was actually used abroad). They therefore
reduce the advantage of producing abroad and selling domestically because the carbon price is
the same regardless of where production takes place. For exports, BAs are a rebate of the tax
that was previously paid when the exported good was produced. By rebating previously paid
taxes or carbon prices on exports, BAs reduce the disadvantage of producing domestically
when selling into foreign markets.”
Such border tax adjustments can limit potential leakage and competitiveness effects, but in turn
they pose several difficulties and risks. BAs are administratively complex instruments that are
74
The European Commission (2015) states that since the start of the third phase, emissions have fallen by 5
per cent and are expected to fall by 21 per cent in 2020 (compared to 2005) and 43 per cent in 2030
(compared to 2005).
126
difficult to design and implement (Flannery, 2016). A major challenge of designing BAs is to
avoid conflict with World Trade Organization rules. Moreover, their effects on the welfare of
the country imposing them are also difficult to estimate and not necessarily positive (Flannery,
2016).
Source: Author’s elaboration based on Flannery (2016).
While carbon taxes and CAP systems are the main policy instruments to tax carbon
directly, two other policy instruments can achieve this goal indirectly. The first is
removing fossil fuel subsidies that encourage consumption of fossil-fuel-based energy,
discourage investments in clean energy sources and energy efficiency, and impose large
costs on governments (Sterner and Köhlin, 2015). Phasing out these types of subsidies is
an administratively simple way of making carbon-based fuels more expensive, thereby
indirectly pricing CO2 and contributing to lower CO2 emissions. The second policy
instrument, directly linked, is fossil fuel taxation. While fossil fuel taxes do not tax fossil
fuels in proportion to their carbon content, they still provide an effective way to indirectly
tax CO2 emissions in the transport sector. As such, they lower total consumption of fossil
fuels as people adapt their behaviour (a fuel price increase of 1 per cent is estimated to
lead to a reduction in fuel consumption of 0.7 per cent in the long run), create incentives
for investments in fuel-saving technologies, and favour the development of energyefficient cars (Sterner and Köhlin, 2015). Fossil fuel taxes such as taxes on petrol and
diesel have been around for a very long time, are frequently applied, and have been wellstudied. Reviewing various studies, Sterner (2007), finds that fuel taxation is the policy
instrument that has had the highest impact on global carbon emissions.
2.1.1.2 Policy instruments that save energy
Policy instruments that increase the energy efficiency of economies can also contribute to
reducing anthropogenic CO2 emissions, especially if they are implemented together with
carbon pricing instruments. By increasingly relying on energy-efficient technologies and
thereby using less energy from fossil fuels, CO2 emissions per US dollar of GDP can be
substantially reduced. The emission reduction potential of energy conservation policies is
large. For example, the International Energy Agency (IEA) estimated that if its 25 energy
efficiency recommendations were implemented globally, up to 7.6 GtCO2 per year
(roughly 21 per cent of global CO2 emissions emitted in 2014) could be saved by 2030
(IEA, 2011).
IEA (2011) identified seven priority areas where increases in energy efficiency can
substantially affect CO2 emissions. These areas include cross-sectoral energy efficiency
increases, and sectoral energy efficiency increases in buildings, appliances and
equipment, transport, lighting, industry, and energy utilities. A wide variety of policy
instruments are available to increase energy efficiency in these areas, including the
following (World Energy Council, 2013):

Institutional approaches (e.g. setting up energy-efficiency agencies, energyefficiency laws, or national energy-efficiency programmes);
127




Product regulations (e.g. energy labels and/or minimum energy-efficiency
standards for cars, buildings, domestic appliances and motors; or bans of specific
energy-intensive products);
Consumer regulations (e.g. mandatory energy audits for selected customers,
energy-saving quotas, mandatory energy-consumption reporting, and energysaving plans);
Financial and fiscal measures (e.g. energy-efficiency funds, subsidies for energy
audits by sector, subsidies or soft loans for energy-efficiency investment and
equipment by sector, tax credits or deductions on cars, appliances, and buildings;
accelerated depreciation for industry, tertiary or transport sectors; or tax
reductions by type of equipment such as appliances, cars, lamps, etc.);
Cross-cutting measures (e.g. voluntary agreements, mandatory professional
training, or promotion of energy-saving companies).
While energy-saving policy instruments can reduce anthropogenic CO2 emissions, their
potential to save energy and thus reduce CO2 emissions might be reduced by so-called
rebound effects. Rebound effects can be illustrated by the following example: if you buy
a more fuel-efficient car you will be driving more often, as you will have to pay less for
fuel (Gillingham et al., 2013). Such behavioural changes in reaction to increased
availability of energy-efficient technologies might offset parts of the energy-saving
potential. However, the literature finds that rebound effects are too small to offset the
energy-saving effects of energy-efficiency policy instruments (Gillingham et al. 2013).
2.1.1.3 Policy instruments that substitute energy from fossil fuels with alternative energy
sources
These instruments aim to reduce the use of energy from fossil fuels and thus facilitate the
decarbonization of the global energy system. IPCC (2014a) estimated that by 2100, the
share of low-carbon energy sources (e.g. renewable energy sources or nuclear power) in
total energy should increase from the current roughly 20 per cent to over 80 per cent. If
this is not achieved, it is unlikely that the atmospheric GHG concentration in 2100 will
stay below the level compatible with the 2°C temperature target. Fundamental progress in
new and more efficient energy technologies is needed for such a transformation, as lowcarbon energy technologies are currently not cost-competitive with fossil-fuel-based
technologies when applied on a large scale (Toman, 2015).
While carbon pricing and some energy-saving policy instruments alter incentives of
economic agents and thus indirectly promote investments in a low-carbon economy, the
energy-substitution policy instruments try to directly promote the replacement of fossilfuel-based energy with low-carbon energy. To decarbonize the energy system, policies
need to stimulate the development and implementation of new low-carbon technologies.
A priori, policies can promote two technological options: nuclear energy technologies
and renewable energy technologies. While next-generation nuclear reactors might replace
a share of fossil-fuel-based energy and address the cost and safety issues undermining
current generation reactors (Toman, 2015), we focus in this section on the promotion of
renewable energy technologies.
128
Renewable energy technologies – hydropower, wind and solar energy, tidal and wave
energy, ocean and geothermal energy, and biomass energy – are expected to play a key
role in decarbonizing the energy system (Bossetti, 2015). Renewables not only have the
potential to substantially contribute to a low-carbon world, but might also generate
important co-benefits, including (a) increased energy security due to more widely
diversified energy sources; (b) reduced local pollution as fossil-fuel-based energy carriers
such as coal are replaced; (c) promotion of green growth; and (d) new possibilities for
development in many regions of the world, as renewables such as solar power are more
evenly distributed globally than fossil fuels (Bossetti, 2015). For example, the African
Development Bank (2015) estimates that Africa has enormous potential for renewable
energy, and that its capacity for renewable energy generation could reach more than
10,000 gigawatts (GW) for solar energy, 109 GW for wind energy, 350 GW for hydro
energy, and 15 GW for geothermal energy. The African Development Bank (2015) is
confident that when the continent unlocks its full potential for renewable energy, it could
tackle fundamental inefficiencies in its energy system, hugely expand power generation,
and contribute towards the development goal of universal access to energy. Hence,
policies facilitating the adoption, integration, and development of renewable energy
sources play a key role in mitigating climate change and can also have huge co-benefits
in terms of health and economic development.
The cost of renewable energy technologies can be considerably lowered by publicly
funded R&D programmes and by providing incentives (e.g. tax incentives) for private
programmes (Bossetti, 2015). These kinds of policies can directly promote development
and adoption of renewable energy sources by making them more cost-competitive. A
second way to promote adoption is to subsidize renewable energy technologies.
Germany’s Energiewende (see Box 17) is an example of probably the most aggressive
subsidy policy to date favouring renewable energy sources. Demand-side promotion
policies such as standards, energy certificates, or feed-in tariffs75 are another policy
option directly promoting adoption of renewable energy sources. Such policies have
contributed considerably to the adoption of solar and wind power in Europe (Bossetti,
2015). Because renewable energy – unlike energy from fossil fuels - is highly
unpredictable and volatile (depending, for instance, on weather conditions in the case of
wind or solar energy), infrastructure has to be adapted. Hence, public investments in new
energy storage and distribution infrastructure can facilitate the integration of renewables
into the existing energy infrastructure.
75
Feed-in tariffs are policy mechanisms that aim to provide long-term contracts to renewable energy
producers based on the cost of the technology, thus stimulating investments in renewable energy
production.
129
Box 17: Germany’s Energiewende policy
Germany has been pursuing an ambitious energy transformation policy for almost two decades
based on massively subsidizing solar and wind energy. The policy, known as Energiewende,
aims to achieve a transition from a coal- and nuclear-power-based energy system to a system
relying on renewable energy. Germany intends to reduce its GHG emissions by between 80 and
95 per cent (compared to 1995) within the next four decades. To achieve this target, it plans to
increase the share of renewable energy in totally consumed energy to 80 per cent and
simultaneously increase energy efficiency by 50 per cent (Sterner and Köhlin, 2015).
From 2000 to 2015, Germany increased its share of renewable energy in the energy production
sector from 6 to almost 30 per cent (Figure 38). However, the share of renewable energy in the
production of heat (9.9 per cent) and transport (5.4 per cent) is still relatively low. Prices of
renewable energy technologies have decreased substantially as a result of the massive subsidies
and continue to do so (Sterner and Köhlin, 2015). However, CO2 emissions per kilowatt of
produced energy have not yet dropped substantially, which Sterner and Köhlin attribute to the
simultaneous phasing out of nuclear energy, which has required continued use of fossil fuel
power stations.
Figure 38: Share of renewable energies in Germany’s energy market
Source: BEE (2015: 1).
While the Energiewende policy has clearly reduced costs and promoted widespread adoption of
renewables, the subsidies have also generated important costs for the German government and
for German households.
Source: Author's elaboration based on Sterner and Köhlin (2015) and BEE (2015).
2.1.2 Policies to promote technological solutions to capture and store carbon dioxide
emissions
Section 2.1.1 described different policy instruments to reduce the production of
anthropogenic CO2 emissions. In addition to these emission reduction policy instruments,
governments can also implement policies that aim to capture and store already-emitted
CO2 emissions before they reach the atmosphere.
130
Carbon capture and storage (CCS) technologies capture CO2 emissions from large
sources such as coal power plants,76 transport them to a carbon storage site, and store
them there so that they do not reach the atmosphere (Tavoni, 2015). Different
technological storage solutions are available, including geological and oceanic storage
sites.77 CCS has the capacity to make the ongoing large-scale use of fossil-fuel-based
energy compatible with climate change mitigation targets: if CCS technologies could be
sufficiently up-scaled, there would not be an immediate need to decarbonize the
economy. Tavoni (2015: 344) notes in his recent review on CCS technologies and
policies that “CCS could effectively allow for the procrastinated use of fossil fuels while
limiting – if not eliminating – their impact in terms of greenhouse gas emissions.” Given
the appealing nature of CCS, it is not surprising that it plays an important role in the
IPCC’s climate mitigation strategies (IPCC, 2014a).
While CCS technologies generate high hopes as an alternative mitigation option to reduce
the flow of emissions towards the atmosphere, the technology is still in the early phases
of development. In 2016, only 15 large-scale CCS pilot projects were operational on a
worldwide scale, with an additional four expected to enter into operation (see Figure 39).
These large-scale projects are complemented by small-scale pilot projects. So far, the
existing projects have the capacity to capture and store 28 million tonnes of CO2 per year
(Global CCS Institute, 2015).
Figure 39: Number of large-scale carbon capture and storage pilot projects per year
Source: Global CCS Institute (2015: 4).
76
Theoretically, it is also possible to apply CCS technologies to biomass burning, which could remove CO2
directly from the atmosphere, creating a so-called “negative emission technology” (Tavoni, 2015; see also
Section 2.1.3). Biomass is a renewable resource which extracts and binds CO2 out of the atmosphere during
its process of growing. By burning biomass, CO2 is emitted back into the atmosphere. If CCS technologies
are applied to biomass burning, the CO2 that biomass extracts from the atmosphere when it is growing
could be captured and stored. This is one way to remove CO2 from the atmosphere. See Section 2.1.3 for a
discussion of other negative emission technologies.
77
Most of these technologies store CO2 in its supercritical form in geological or oceanic storage sites.
Supercritical CO2 is a fluid state of CO2 in which it is held at or above its critical temperature and critical
pressure.
131
It should be noted, however, that several technological, political, and economic problems
would need to be solved in order to apply CCS technologies on a sufficiently large scale.
First, it is still not clear which CCS technology works best and if it will be able to operate
at the required scale. Policy measures are therefore required to launch and (partially) fund
additional pilot projects to test the different CCS designs (Tavoni, 2015). To do so, it
would also be necessary to overcome public fears opposing the installation of CCS
infrastructure. This has led to the cancellation of CCS pilot projects in Europe in the past.
It is also of crucial importance that storage solutions be safe and long-lasting, as leaking
CO2 escaping towards the atmosphere would undermine the entire purpose of CCS.
Second, the economic viability of CCS projects has to be improved, as the cost of
capturing a tonne of CO2 is currently estimated at about US$100 (Tavoni, 2015). Policies
to increase the price of carbon (thereby making higher-cost CCS technologies
economically viable) and policies to fund R&D could help achieve such a cost reduction
(Tavoni, 2015).
2.1.3 Policies to promote technological solutions aimed at removing carbon dioxide
directly from the atmosphere
While Sections 2.1.1 and 2.1.2 discussed policy options to reduce the flow of emissions
towards the atmosphere, this section discusses a policy instrument that aims to remove
CO2 directly from the atmosphere: carbon geoengineering.
Several technologies could remove CO2 from the atmosphere. One of them, already
briefly introduced in Section 2.1.2, is the application of CCS technologies to biomass.
According to Barrett and Moreno-Cruz (2015), this approach could be useful but will be
limited in scale by design. A second technology is industrial air capture, a carbon
geoengineering technology that could be a viable alternative with the potential to be
scaled up to any level (Barrett and Moreno-Cruz, 2015). Industrial air capture is a
technology in which a capture solution (e.g. a chemical sorbent like alkaline liquid) is
exposed to air, resulting in a chemical reaction removing CO2 from the air. The CO2 is
subsequently filtered out of the CO2-enriched capture solution and is stored in suitable
storage sites while the chemical sorbent is recycled (Figure 40).
Figure 40: Direct industrial air capture system
Source: Carbon Engineering. Available at: http://carbonengineering.com/air-capture.
132
If industrial air capture is powered by renewable energy, the technology has the capacity
to remove any amount of CO2 from the atmosphere. If enough renewable energy is
available to operate this technology on a sufficiently large scale, no other policies to
stabilize CO2 concentrations would be needed. As such, carbon geoengineering is
considered to be the only true backstop technology to address climate change (Barrett and
Moreno-Cruz, 2015). However, this technology can only be implemented provided it is
further developed and available at an affordable cost.
Several studies reviewed in the survey by Barrett and Moreno-Cruz (2015) estimate the
costs of removing a tonne of CO2 through industrial air capture. These estimates range
from US$30 per tonne up to US$600 per tonne. Barrett and Moreno-Cruz argue that if
costs turn out to be as high as US$600 per tonne, the technology will not take off within a
reasonable time frame. If, however, costs are as low as US$30 per tonne, the technology
could revolutionize climate change policy and could even be implemented by a small
coalition of countries. This coalition could unilaterally mitigate climate change (Barrett
and Moreno-Cruz, 2015) by bypassing the collective action problem associated with
conventional emission reductions, and emission capture and storage policies.
Because this technology is not yet on the table of climate change policy negotiations (see
Section 4), and because there are still many unknowns about it, the first policy step
should be to fund and coordinate R&D efforts to evaluate the costs and risks of industrial
air capture technologies. Only if and when costs and risks are known can further policy
instruments to promote carbon geoengineering be designed and implemented.
2.2 Policies to promote technological solutions aimed at increasing the amount of
incoming solar radiation reflected back into space
Section 2.1 provided an overview of policy instruments to stabilize atmospheric GHG
concentrations. Alternatively, however, humans could also try to offset the positive
radiative forcing resulting from increasing GHG concentrations by reflecting more of the
incoming solar radiation back into the space. This approach, known as solar
geoengineering, is able to offset increases in mean temperature rather quickly. But it does
not address certain other impacts of increased GHG concentrations such as ocean
acidification (Barrett and Moreno-Cruz, 2015). For this reason, Barrett (2014) labels solar
geoengineering a quick fix but not a true solution to climate change.
Solar geoengineering can be implemented in many ways. One option seems to be
particularly promising in terms of costs: the injection of sulphate aerosols into the
stratosphere. Barrett and Moreno-Cruz (2015) cite a study by McClellan et al. (2012) that
provides an estimate of less than US$8 billion per year to offset the positive radiative
forcing expected to occur over coming decades due to rising GHG concentrations. Unlike
all other options, it seems that, in theory, even one willing country could implement such
a solution single-handedly (Table 8).78
78
This raises international governance questions, which today are far from settled. See Barrett and MorenoCruz (2015) for a short discussion on possible international governance frameworks.
133
Table 8: Comparison of policy options limiting climate change
Source: Author’s elaboration based on Barrett and Moreno-Cruz (2015).
However, as in the case of carbon geoengineering, many unknowns remain about solar
geoengineering. Most importantly, the technology may entail severe environmental risks
that are not yet (fully) understood (Barrett and Moreno-Cruz, 2015). Funding and
coordinating R&D to assess the risks, feasibility, and efficiency of solar geoengineering
is therefore needed before such technologies could be deployed.
Short summary
Section 2 reviewed the three fundamental policy options humans have to limit climate
change: (1) substantial reduction of the flow of greenhouse gas emissions reaching the
atmosphere, (2) direct removal of CO2 from the atmosphere by using carbon
geoengineering, and (3) offsetting the positive radiative forcing resulting from increasing
GHG concentrations by using solar geoengineering. Error! Reference source not
found.These three options, including their expected costs, risks, unknowns, and the
prospect of collective action, are summarized in Table 8. While policies to substantially
reduce the flow of emissions (carbon pricing, promotion of renewables, conventional
carbon capture and storage) are well known and entail only few risks, there are two main
caveats regarding their implementation: associated costs are high and there are collective
action problems. Despite this, actions to limit climate change have so far focused almost
exclusively on these options, which have also been implemented by several developed
countries and some developing countries. Carbon geoengineering could be an alternative
to these policies because it could be implemented by only a handful of nations, thereby
bypassing the collective action problem. However, the associated environmental risks and
especially the costs are higher than those associated with policies to reduce the flow of
emissions. Finally, solar geoengineering has the advantage of being easy to implement
single-handedly by only one country and of having very low costs. At the same time,
however, it does not address some impacts of climate change, and entails even higher
environmental risks than carbon geoengineering.
3 Climate change adaptation policy options
In addition to policies to limit climate change, humans can also try to reduce or even
avoid the anticipated impacts of climate change by adapting to climate change. IPCC
(2014c: 118) defines climate change adaptation as the “process of adjustment to actual or
expected climate and its effects. In human systems, adaptation seeks to moderate or avoid
134
harm or exploit beneficial opportunities. In some natural systems, human intervention
may facilitate adjustment to expected climate and its effects.”
Given that the anticipated impacts of climate change are expected to be largest in
developing and least developed countries (see Module 2), adaptation policies are
especially important for these countries. Drawing on Sauter et al. (2015, 2016), Figure 41
displays the projected population and damage shares of different regions.79 Regions
mostly composed of developing countries (especially South Asia, sub-Saharan Africa,
and the Middle East and North Africa) are expected to have a higher share in world
damages in proportion to their share in world population by 2050. In other words, these
regions will most likely be over-proportionally affected by the impact of climate change.
Regions mostly composed of developed countries (especially Europe and North America)
are anticipated to have lower damage shares in proportion to their population shares.
While these are only rough estimates, the numbers clearly indicate that adaptation
policies are of crucial importance for developing countries.
Figure 41: Potential climate change damage share in relation to population share by
region in 2050 (per cent)
Source: Mekonnen (2015: 88), adapted from Sauter et al. (2015, 2016).
Humans have a long history of adapting to a changing climate. Throughout history,
human beings have changed the locations or construction of their settlements, adapted
their agricultural technologies, and changed entire economic processes in response to
climate variations (Burton, 2006). Humans have been rather successful in doing so, but
history also provides examples of local adaption failures that resulted in the collapse of
entire societies (see Module 1). Adapting to human-induced climate change thus
represents a new chapter in a long history of human adaptation to climatic conditions, but
this time, the dimension of the challenge is global.
Climate change adaptation policies have gained increased attention recently. Since about
10 years ago, adaptation has figured more prominently on the international policy agenda
as the failure of current policies to reduce emissions has become evident (Barrett and
79
Sauter et al. (2015, 2016) used a rough proxy of potential damages based on predicted increases in
extreme temperature events.
135
Moreno-Cruz, 2015). As will be shown in Section 4, these policies are currently one of
the two main pillars of international climate change policy (the second pillar being
policies to reduce the flow of emissions towards the atmosphere). Given that climate
change impacts, risks, and vulnerabilities differ locally, different adaptation policies are
needed in different places on the planet. The first step of any comprehensive adaptation
policy therefore consists of assessing local risks and vulnerabilities in order to identify
local needs (Noble et al., 2014). Once local needs are known, the government can choose
appropriate measures from a wide range of adaptation options. While a detailed
discussion of all these options is out of the scope of this teaching material, we provide a
schematic overview by using the IPCC’s classification of adaptation measures.
Table 9 lists the three broad categories of adaptation options as classified in the fifth
IPCC assessment report: structural and physical options, social options, and institutional
options (Noble et al., 2014).
136
Table 9: Categories and examples of adaptation options discussed in the fifth IPCC
assessment report
Source: Noble et al. (2014: 845).
Structural and physical options include the use of engineering (e.g. building sea walls or
coastal protection structures), the application of specific technologies (e.g. the use of new
crop varieties resistant to local anticipated climate impacts or the implementation of early
warning systems to prepare for the increased frequency of extreme weather events), the
use of ecosystems and ecosystem services (e.g. assisted migration of threatened species
or ecological restauration measures), and the delivery of specific services at various
levels (e.g. the creation of food banks to counter anticipated food shortages or the
137
implementation of vaccination programmes to limit expected health risks). Box 18
provides an example of a structural adaptation project in Bangladesh.
Box 18: Cyclone shelters and early warning systems in Bangladesh
Bangladesh has experienced a number of extreme weather and climate events over its history.
According to the Intergovernmental Panel on Climate Change (IPCC), approximately 500,000
people died during a category 3 cyclone in 1970. In 1991, another category 3 cyclone killed
140,000 people.
As climate change increases, the likelihood of such extreme weather and climate events,
including future cyclones, is probable. Bangladesh has therefore undertaken a collaborative
process between local communities, government, and private organizations to implement
adaptation measures. This collaboration has improved overall education about disasters (a
social adaptation measure), deployed early warning systems based on high-technology
information systems and measures such as training volunteers to distribute warning messages
by bicycle (a structural and physical adaptation measure), and built a network of cyclone
shelters (another structural and physical adaptation measure).
As a result of these adaptation efforts, Bangladesh was able to achieve a remarkable reduction
in mortality: during the category 4 cyclone in 2007, 3,400 people died, considerably down
from the 500,000 and 140,000 deaths recorded during the previously mentioned cyclones. The
achievement is even more remarkable considering that the country experienced population
growth of more than 30 million between the cyclone events.
Source: Author's elaboration based on Smith et al. (2014).
Social options aim to reduce local vulnerabilities of disadvantaged segments of the
population and actively address social inequities (Noble et al., 2014). They include
educational measures (e.g. awareness-raising and knowledge-sharing about adaptation
planning), informational measures (e.g. the development and dissemination of improved
climate forecasts to better identify risks), or behavioural measures (e.g. changing
agricultural practices, switching crop varieties, or developing household evacuation
plans). Table 9 outlined additional examples of social adaptation options. Box 19
provides an example of an adaptation project implemented in Bolivia that combines a
behavioural measure with a structural measure.
Box 19: The camellones project in Bolivia
In Bolivia, the frequency of extreme weather events has increased in recent years and is
expected to further increase in the future due to climate change. Increased frequency of floods
particularly threatens the poor population living in remote rural areas of the country.
Devastating floods in Beni in 2007 prompted people from locations around Trinidad to
participate in a project known as camellones that was jointly implemented by the Kenneth Lee
Foundation and Oxfam.
The project consists of a combination of behavioural adaptation – a change of an agricultural
practice – and a structural adaptation – the building of the camellones platforms. The aim of
the project was to avoid future agricultural losses due to increased occurrence of floods.
Instead of using a conventional farming technique, the project implemented the camellones
138
farming technique that is based on a traditional farming method. Several modern camellones
were built in 2007 and have since been used for farming (see photo). These modern camellones
are essentially earth platforms. Each platform measures roughly 500 m2 and varies in height
between 0.5 and 3 meters, depending on the predicted height of the local flooding and the
area’s capacity for water run-off (Oxfam, 2009). Because these platforms are above the flood
levels, they protect the crops from the flood. The ditches surrounding the camellones become
canals in the event of a flood and act as an irrigation and nutrient source during the dry season
(Oxfam, 2009).
Camellones in Bolivia
Source: http://valorandonaturaleza.org/noticias/manejo_precolombino_de_ecosistema_amaznico_
As a result of the new agricultural practices, farmers did not lose their crops and seeds during
floods in 2008. The system seems to be a sustainable solution to address increased flooding,
and it also preserves water for times of drought, acts as a natural seed bank, and improves soil
quality and ultimately food security (Oxfam, 2009).
Source: Author's elaboration based on Oxfam (2009).
Institutional options aim to increase the potential for adaptation (Noble et al., 2014)
through a variety of economic measures (e.g. disaster contingency funds or financial
incentives to implement adaptation options), legal and regulatory measures (e.g.
mandatory building standards, or laws to encourage contracting and insurance), and
governmental plans and programmes (e.g. national and regional adaptation plans to
139
coordinate adaptation efforts). Table 9 listed additional institutional options. An example
of an institutional adaptation policy is provided in Box 20.
Box 20: Nepal’s National Framework on Local Adaptation Plans for Action
In 2011, Nepal adopted the Nepal National Framework on Local Adaptation Plans for Action
(LAPA). The LAPA framework aims to integrate climate adaptation into local and national
development planning processes in a bottom-up, inclusive, responsive, and flexible manner
(Government of Nepal, 2011). Figure 42 outlines schematically how climate adaptation actions
are integrated into and harmonized with local and national planning. The LAPA framework
foresees seven steps: (1) information gathering; (2) vulnerability and adaption assessment; (3)
prioritization of adaptation options; (4) formulation of LAPAs; (5) integration of LAPAs into
local planning processes; (6) implementation of LAPAs; and (7) assessment of the progress of
LAPAs. These steps are followed by implementation of a village- or district-specific LAPA
(Government of Nepal, 2011).
Figure 42: The Local Adaptation Plans for Action framework
Source: Government of Nepal (2011: 6).
Note: NAPA: National Adaptation Programme of Action; VDC: Village Development Committees.
By 2014, a total of 70 LAPAs had been prepared (69 at the village level and one at the
municipality level) and implemented by the affected communities (Mimura et al., 2014).
Source: Author's elaboration based on Government of Nepal (2011) and Mimura et al. (2014).
140
Identifying adaptation needs and implementing adaptation options is costly and requires
funding. As risks and vulnerabilities and thus adaptation needs are unevenly distributed
and are often highest in poor countries, local adaptation requires globally coordinated
financing. As Section 4 will show, negotiations on adaptation financing are a central part
of global climate change negotiations.
Short summary
Section 3 reviewed adaptation policies that aim to reduce or even avoid the anticipated
impacts of climate change by adapting economies to climate change. Given that climate
change impacts, risks, and vulnerabilities differ locally, different adaptation policies are
needed in different places on the planet. The section explained that the first step of any
comprehensive adaptation policy therefore consists of assessing local risks and
vulnerabilities in order to identify local needs. Once local needs are known, the
government can choose appropriate measures from a wide range of adaptation options.
The section also highlighted that adaptation policies are especially important for
developing and least developed countries, as the anticipated impacts of climate change
are expected to be largest in these countries.
4 The international climate change policy architecture
Having discussed the different options for climate change limitation and adaptation
policies, we now turn our attention towards the international climate change policy
framework. Multilateral climate change negotiations to date have mainly focused on
emission reduction policies and, more recently, have also started to integrate adaptation
policies. We structure our discussion around three milestones in the development of the
multilateral climate change policy framework: (a) the signature of the 1997 Kyoto
Protocol, which was the first substantial international agreement to reduce GHG
emissions by establishing country-specific, legally binding emission targets for
developing countries; (b) the failure of the COP15 in Copenhagen that marked the end of
the Kyoto Protocol approach; and (c) the successful conclusion of the COP21 in Paris,
which marked the start of a new era of international climate policy architecture. Unlike
what has happened over the past two decades, the Paris Agreement does not impose
legally binding emission targets for developed countries, but rather is built on voluntary
national contributions by all countries. As such, it represents a major change from the
Kyoto Protocol approach, which had a more limited scope albeit high ambitions, to a
regime based on broad participation, even if its ambitions are initially relatively modest.
4.1 From Rio to Paris – 25 years of climate change negotiations80
The 1992 Conference in Rio adopted the United Nations Framework Convention on
Climate Change (UNFCCC), which required signatory countries to undertake national
inventories of GHG emissions and develop action plans to reduce national emissions.
Signatory countries of the UNFCCC then held yearly conferences (called Conferences of
the Parties - COPs) starting in 1995. From the very beginning of the UNFCCC process, a
80
This section draws on Flannery (2016).
141
key principle guiding the negotiations was that of common but differentiated
responsibilities. This principle is enshrined in Principle 7 of the Rio Declaration on
Environment and Development:
“In view of the different contributions to global environmental
degradation, States have common but differentiated responsibilities.
The developed countries acknowledge the responsibility that they bear
in the international pursuit of sustainable development in view of the
pressures their societies place on the global environment and of the
technologies and financial resources they command.”
In other words, the principle of common but differentiated responsibilities takes into
account that the bulk of cumulative human-induced GHG emissions have been emitted by
the current developed countries. For this reason, these countries should bear a greater
responsibility in solving the climate change issue.
The Kyoto Protocol was the first substantial international agreement to reduce GHG
emissions. Signed at the COP3 in Kyoto in 1997, it fully embraced the principle of
common but differentiated responsibilities (Flannery, 2016). The protocol used the
UNFCCC classification of countries into Annex I, Annex II, and Non-Annex I Parties.
Annex I Parties encompassed industrialized member countries of the Organisation for
Economic Co-operation and Development (OECD) as of 1992 and countries with
economies in transition (EIT Parties), which included Russia, Central and Eastern
European countries, and Baltic countries. Annex II Parties consisted of a subset of Annex
I countries, namely those countries that were members of the OECD in 1992. Non-Annex
I Parties were mostly developing countries.
Based on this country classification, the Kyoto Protocol imposed legally binding
mitigation targets for Annex I Parties and established a market mechanism to facilitate
reductions in GHG emissions. In accordance with the principle of common but
differentiated responsibilities, it excluded all developing countries (Non-Annex I Parties)
from any legally binding obligation to reduce emissions. Moreover, it imposed an
obligation on Annex II parties to provide financial support to Non-Annex I Parties.
The Kyoto Protocol used the so-called top-down approach (Flannery, 2016): negotiations
conducted at the international level focused essentially on determining country-specific,
legally binding emission targets for Annex I countries. Each country then had to
implement the required emission reductions using suitable policy instruments (many of
which were described in Section 2.1). The overall goal of the protocol was to reduce
emissions of all major greenhouse gases by 5 per cent by 2012 relative to 1990.
To enter into force, the Kyoto Protocol needed to be ratified by at least 55 countries,
responsible for at least 55 per cent of 1990 CO2 emissions. While it was signed in 1997,
the required conditions for entering into force were only met in 2005. The first phase of
the protocol thus started in 2005 and was set to end in 2012. The United States, at the
time the country with the highest GHGs emissions worldwide, never ratified the protocol.
142
Its absence from the Kyoto Protocol illustrates the collective action problem (see Module
3) and has been one of the factors that considerably weakened the Kyoto process.
After 2005, efforts quickly concentrated on negotiating the roadmap for future
developments to shape international climate change policy after the first commitment
phase of Kyoto expired in 2012. During COP13 in Bali in 2007, a roadmap was agreed
upon with two key points. The first was to prepare a second Kyoto commitment period
with legally binding emission reductions for Annex I Parties for the COP15 to be held in
Copenhagen. The second was to open negotiations for a new agreement involving all
UNFCCC parties. According to Flannery (2016), it was this latter decision that signalled,
for the first time, that the principle of common but differentiated responsibilities could
evolve.
The COP15 in Copenhagen in 2009 resulted in an outcome that many observers qualified
as a failure: the conference did not agree on a legally binding extension of the Kyoto
Protocol. Flannery (2016: 72) states that by failing to reach such a legally binding
extension, “Copenhagen dealt a deathblow to the top-down approach in which nations
negotiated terms for one another’s actions as the basis for agreement. Going forward,
national pledges will be based on voluntary submissions that reflect national
circumstances and priorities.” Such an approach based on voluntary national
contributions is generally called a bottom-up approach. In retrospective, Copenhagen
indeed marked the end of top-down approaches in international climate change
negotiations. Such an outcome could have been anticipated, however, as many heads of
state (including those of the United States and the People's Republic of China) had
announced earlier that they would not sign a legally binding agreement. After the
Copenhagen failure, negotiations remained stalled for several years. It was only in 2012
at the COP18 in Doha that there was agreement on a second Kyoto commitment period
(until 2020).
With regard to global GHG emissions, the top-down approach of the Kyoto Protocol can
be considered a failure (Barrett et al., 2015): the IPCC notes that while the group of
Annex I countries indeed managed to collectively meet their Kyoto target by reducing
their aggregate emissions more than 5.2 per cent below 1990 levels by 2012, these
reductions were largely offset by emission growth in Non-Annex I countries (Stavins et
al., 2014).81 Consequently, global GHG emissions, instead of being reduced, have been
growing at an unprecedented rate over the past two decades.
In parallel with negotiations on a second Kyoto commitment period, negotiations were
taking place to develop a post-2020 climate change policy framework. To this end, the
COP17 in Durban established the Ad Hoc Working Group on the Durban Platform for
Enhanced Action (ADP) with a mandate to (a) increase the ambition of the climate
change limitation policy for the pre-2020 period, and (b) negotiate a global agreement for
the post-2020 period by 2015 (Flannery, 2016). Compared to the political landscape of
the 1990s that led to adoption of the Kyoto Protocol, the situation has changed
considerably. Besides the obvious failure of the top-down approach of the Kyoto
81
This growth can be partially attributed to carbon leakages from developed countries.
143
Protocol, GHG emissions from developing countries have been growing rapidly. While
the United States had been the largest emitter of CO2 until the mid-2000s, the People's
Republic of China has been the largest emitter since 2005. Other developing countries
such as India are currently also among the top emitters worldwide. Following intense
negotiations, the ADP’s work ultimately led to the outcome of the COP21: the Paris
Agreement.
4.2 The Paris Agreement
Negotiations for a post-2020 climate change framework initiated by the ADP concluded
successfully in 2015 with the adoption at COP21 of the Paris Agreement. The agreement
brings the 197 parties to the United Nations Framework Convention on Climate Change
(all United Nations Member States, the State of Palestine, the Cook Islands, Niue and the
European Union) under a common legal framework. This represents a change from the
Kyoto Protocol, which had a more limited scope, albeit high ambitions, to a regime based
on broad participation, even if ambitions initially are relatively more modest. This
departure from the past two decades of a top-down climate policy, which suffered from
limited coverage of only a small subset of countries required to cut emissions, towards a
bottom-up framework, which involves all UNFCCC parties, is important given the
paramount significance of long-term action by all countries to address climate change.
Together with the new architecture combining bottom-up Nationally Determined
Contributions (NDCs) – i.e. national commitments to reduce GHG emissions, with topdown procedures for reporting and synthesis of NDCs by the UNFCCC Secretariat – the
Paris Agreement represents significant progress towards a climate regime that could limit
climate change. Prior to the Paris Conference, 186 parties had submitted Intended
Nationally Determined Contributions (INDCs),82 and two more did so during the
conference, bringing the total to 188 INDCs. The Paris Agreement will take effect in
2020 when the Kyoto Protocol ends. In the interim, parties agreed to promote climate
action, ramp up financing, and begin implementation of their climate plans. They will
have an opportunity, as part of a collective review in 2018, to update these plans. Equally
important is the fact that the private sector has been involved in a major way in
contributing to climate action. Over 5,000 global companies in 90 countries covering all
industrial sectors pledged actions to combat climate change.
It is important to note that the new policy architecture incorporates several elements that
have been identified by game theory models (see Module 3) as being able to increase
cooperation. NDCs can be viewed as national commitments to reduce emissions (to some
extent) regardless of what other countries are doing. If they turn out to be credible
commitments, they will be a key element fostering cooperation. The financial transfer
mechanisms and the repeated follow-up procedures (see Section 4.2.2), as well as the
prospect of linking development and climate change policy issues (see Section 4.2.3), are
additional elements that strengthen cooperation under the Paris Agreement. So far, these
elements seem to have positively affected the agreement’s signature and ratification
process. The Paris Agreement was opened for signature on 22 April 2016. By 29 June
82
Note that INDCs became NDCs when the Paris Agreement was concluded.
144
2016, 179 countries had signed it. To date, 19 of the 179 signatories have also deposited
their instruments of ratification, acceptance, or approval, accounting in total for 0.18 per
cent of total global GHG emissions (UNFCCC, 2016). According to Article 21, §1, of the
Paris Agreement, the agreement will enter into force 30 days after the date on which at
least 55 parties to the convention, accounting in total for at least 55 per cent of total
global GHG emissions, have deposited their instruments of ratification, acceptance,
approval, or accession with the Depositary.83
Similar to the Kyoto Protocol, the Paris Agreement reflects the principle of common but
differentiated responsibilities and thus takes on board the equity and fairness concerns of
developing countries. The following sections review the content and the ambition of the
agreement, explain the follow-up procedures, and discuss climate financing.
4.2.1 Objectives of the Paris Agreement
Parties to the agreement aim to limit “the increase in the global average temperature to
well below 2°C above pre-industrial levels and to pursue efforts to limit the temperature
increase to 1.5°C above pre-industrial levels” (Article 2.1.a of the agreement). This
formulation represents a win for small island developing states and other developing
nations that argue that a temperature increase above 1.5°C would be devastating for them.
Parties also aim to “reach global peaking of GHG emissions as soon as possible,” without
specifying a date. They will “undertake rapid reductions thereafter in accordance with
best available science, so as to achieve a balance between anthropogenic emissions by
sources and removals by sinks of greenhouse gases in the second half of this century”
(Article 4.1). This does not mean that emissions would go to zero, but that they would go
low enough that they could be offset by natural processes or advanced technologies that
are able to remove greenhouse gases from the air (see Sections 2.1.2 and 2.1.3).
Put simply, the Paris Agreement requires any country that ratifies it to act to stem its
GHG emissions in the coming century, with the goal of peaking GHG emissions “as soon
as possible” and continuing the reductions as the century progresses. Parties will aim to
prevent global temperatures from rising more than 2°C by 2100 with an ideal target of
keeping the temperature rise below 1.5°C.
4.2.2 Follow-up procedures and financing
The Paris Agreement calls on all countries to submit a new NDC every five years
(Articles 3, 4, 7, 9, 10, 11, and 13) that should represent a “progression” over the prior
one, and should reflect the country’s “highest possible ambition” (Article 4.3). This
process is crucial, because the commitments in current NDCs are not sufficient to limit
warming to below 2°C, much less 1.5°C. The UNFCCC Secretariat’s assessment of the
collective impact of over 146 INDCs submitted by 1 October 2015 (i.e. prior to the Paris
Conference) concluded that they will result in a fall in global emissions and keep the rise
83
Note that the agreement specified the Secretary-General of the United Nations as Depositary. He or she
will be responsible for ensuring the proper execution of all treaty actions related to the agreement.
145
in global warming to around 2.7°C by 2100. While not enough to avert a dangerous
warming of the earth, these commitments are important steps forward.
Implementation of the agreement will be assessed at “global stocktakes” every five years,
with the first global stocktake scheduled for 2023 (Articles 14.1 and 14.2).
In addition to climate change mitigation, the agreement also stipulates that countries will
“engage in adaptation planning processes” (Article 9) to ensure that they are ready to
cope with the effects of climate change. For impacts to which countries cannot adapt, the
agreement contains a “loss and damage” section, suggesting that these cases will be
addressed through a variety of means including “risk insurance facilities, climate risk
pooling and other insurance solutions” (Article 8.4.f). This provision is another key win
for small island states and other vulnerable developing nations.
The agreement states that developed countries “shall provide financial resources to assist
developing country parties with respect to both mitigation and adaptation” (Article 9.1) –
i.e. to help them brace for impacts and shift to cleaner energy systems. The text suggests,
however, that wealthier developing countries can also contribute such funds if they so
choose. Several such countries have considered doing so (e.g. Brazil), and the People's
Republic of China has already pledged to provide US$3.1 billion over a three-year
period. Developed countries should report on their climate donations every two years
(Article 9.5). Such financial assistance will to a large extent determine the level of
implementation of NDCs by developing countries. The commitment of developing
countries to address climate change is thus largely contingent on financial assistance from
wealthier countries.
In this context, developed countries should continue their existing collective financing
mobilization goal agreed upon in Cancún at COP16 through to 2025. That goal was to
provide funds equal to US$100 billion per year by 2020. Prior to 2025, the parties to the
Paris Agreement shall set a new collective quantified financing goal of at least US$100
billion per year, taking into account the needs and priorities of developing countries
(Article 9, §3, Decision 54). To put this goal of US$100 billion per year into perspective,
note that available climate financing in 2014 was estimated at US$62 billion (Mai et al.,
2016). As Figure shows, roughly one-third of this amount came from multilateral sources,
slightly more from bilateral sources, and the remaining part from private sources.
Considerable efforts will thus have to be made to reach the US$100 billion per year target
by 2020.
146
Figure 43: Estimated climate financing in 2014
Source: Mai et al. (2016: 28)
a
Includes only revenues from developed countries.
Financial resources to support climate mitigation and adaptation actions in developing
countries are managed by four funds: the Green Climate Fund, the Global Environment
Facility, the Least Developed Countries Fund, and the Special Climate Change Fund
(Decision 59). While raising the required US$100 billion per year is clearly the most
pressing challenge, several other challenges remain on the spending side and still need to
be addressed. Mai et al. (2016: 27) note that there “are concerns on the spending side
about the balance between mitigation and adaptation (currently most is on the former),
allocating funding across countries and projects accounting for efficiency and equity, and
avoiding paying for projects that would have gone ahead without funding.”
4.2.3 Paris Agreement and development
Compared to the previous climate change policy architecture, the Paris Agreement marks
a change for developing countries: they are now also called upon to implement climate
change limitation policies by submitting their own NDCs, which was not the case under
the Kyoto Protocol. These countries have a strong interest in limiting climate change, as
they are over-proportionally exposed to the associated risks (see Module 2 and also
Section 3 of this Module). However, they also have a strong interest in rapid economic
growth that would allow them to eradicate poverty and increase standards of living
(Collier, 2015).
To pursue these two objectives simultaneously, the transformation towards low-carbon
and resource-efficient economies must proceed in a way that allows for decoupling
economic growth and social development from climate change. UNCTAD (2015: 1)
notes that for developing countries, “such a decoupling should facilitate, and not
undermine, opening new trade and investment opportunities, generating jobs, growing
national economies, widening access to basic necessities and essential services and
reducing poverty.” To be able to achieve these objectives while limiting climate change,
developing countries thus firmly demanded the inclusion of the principle of common but
differentiated responsibilities in the Paris Agreement.
147
Given that they succeeded and that the Paris Agreement is built upon this principle,
developing countries can count on financial and technological support from developed
countries to implement their NDCs. The financial transfer mechanisms aiming at
transferring at least US$100 billion per year from developed to developing countries by
2020 is one of the two key factors expected to enable developing countries to implement
their NDCs. Technology transfer from developed to developing countries in areas such as
renewable energy and energy conservation technologies is the second key factor for
successful implementation of developing countries’ NDCs. Financial and technological
transfers should jointly ensure that the implementation of NDCs does not hurt developing
countries’ growth prospects.
Several observers are confident that due to the finance and technology-transfer
mechanisms developing countries might not only be able to contribute to limit global
GHG emissions by implementing their NDCs, but could also benefit from important cobenefits of these policies. The African Development Bank (2015: 3) notes that the need to
respond to climate change represents “an opportunity to drive the economic
transformation that Africa needs: climate-resilient, low-carbon development that boosts
growth, bridges the energy deficit and reduces poverty. Climate change gives greater
urgency to sound, growth-stimulating policies irrespective of the climate threat.” The
African Development Bank (2015) identifies several important co-benefits in different
sectors. The African energy sector could greatly benefit from a transformation towards
renewable energies, which could tackle fundamental inefficiency in Africa’s energy
systems and generate important investment opportunities. A second set of major cobenefits relates to the agricultural sector. Implementation of climate-smart agriculture
could increase Africa’s annual agricultural output from the current US$280 billion to an
estimated US$880 billion by 2030, which would offer the potential to increase food
security and generate jobs (African Development Bank, 2015). The third major set of cobenefits lies in low-carbon and climate-resilient investments in African cities. Such
investments could make cities less vulnerable to anticipated climate change impacts, and
also improve economic productivity, security, air quality, and public health, as well as
reduce poverty (African Development Bank, 2015). The combination of finance and
technology-transfer mechanisms with policies that help mitigate climate change and also
produce significant co-benefits stimulating growth is thus key for the successful
transformation of developing economies into low-carbon economies.
Many developing countries were among the first parties to sign the Paris Agreement: 18
of the 19 countries that had ratified the agreement as of this writing are developing
countries. Many developing countries have also already submitted detailed NDCs. Box
21 briefly illustrates the objectives and approaches of the NDC of one such developing
country, Kenya.
Box 21: Kenya’s Nationally Determined Contribution objectives and approaches
Kenya’s experience is typical for a lower-income country that aspires to high growth but also
wants to pursue climate change policies. According to Kaudia (2015), Kenya’s GHG emissions
were estimated to be 73 MtCO2-eq in 2010, with roughly 75 per cent of them attributable to
activities such as agriculture, forestry, and free-range rearing of livestock. Historically, Kenya’s
148
contribution to cumulative global GHG emissions has been very low, ranging around 0.1 per cent,
with cumulative per capita emissions of less than 1.26 MtCO2-eq (the global average is 7.58
MtCO2-eq). Kaudia (2015) notes that the country plans to attain 10 per cent GDP growth by
2030, which would double its emissions if no additional climate change policies were
implemented. The Intended Nationally Determined Contribution Kenya submitted before the
COP21 in Paris foresees reducing the country’s emissions by 30 per cent compared to their 2014
levels (see the low-carbon pathway in Figure 44). Kenya thus intends to pursue ambitious growth
while simultaneously committing to an ambitious emission reduction goal.
To achieve this ambitious goal, Kenya has already implemented several policy measures to limit
climate change. These include policies to indirectly price carbon dioxide by taxing older and
hence fuel-inefficient vehicles and by limiting the age of vehicles that can be imported to a
maximum of eight years. Kenya has also integrated climate change into its national planning
process and established, in cooperation with the World Bank, what are called Climate Innovation
Centers, which have had a positive impact through various climate-change-related investment
projects (Kaudia, 2015).
Figure 44: Kenya’s carbon dioxide abatement potential by sector
Source: Kaudia (2015) based on data from Government of Kenya (2012).
Kaudia (2015) reports that reducing the rate of deforestation by way of large-scale tree planting
would be the least-cost solution to tackle climate change in Kenya and other low-income
countries. As shown in Figure 44, Kenya’s forestry sector clearly has the biggest CO2 abatement
potential. This cost-effective solution would preserve or even increase natural carbon sinks and
thus contribute to reducing the country’s emissions. Kaudia (2015) therefore recommends that
low-income countries that rely on natural capital to develop their green growth strategy should
mainly focus on environmental and natural resource management policies.
Source: Author’s elaboration based on Kaudia (2015).
149
Short summary
Section 4 reviewed the international climate change policy architecture. It discussed the
past 25 years of international climate change policy that started with the adoption of the
United Nations Framework Convention on Climate Change (UNFCCC) in 1992. The
section showed that climate change policy has been shaped by the principle of common
but differentiated responsibilities. It also showed that over past years, climate change
policy shifted from a top-down approach – the Kyoto Protocol – towards a mixed
approach upon which the Paris Agreement relies. This latter approach combines bottomup nationally determined contributions – i.e. national commitments to reduce greenhouse
gas emissions – with top-down procedures for reporting and synthesis of nationally
determined contributions by the UNFCCC Secretariat.
5 Exercises and discussion questions
1. Describe the two fundamental options available to limit climate change.
2. List policy instruments capable of reducing atmospheric greenhouse gas
concentrations.
3. Compare carbon taxes to carbon cap and trade systems and discuss their relative
advantages and disadvantages.
4. What are carbon capture and storage technologies?
5. Compare solar geoengineering to carbon geoengineering. What are the
fundamental differences?
6. Discuss the objectives, costs, risks, unknowns, and collective action prospects of:
 Substantial reductions of the flow of carbon dioxide emissions reaching the
atmosphere
 Carbon geoengineering
 Solar geoengineering
7. Discuss the importance of climate change adaptation projects for your country.
Find examples of planned or implemented climate change adaptation projects.
8. Identify the fundamental difference between the Paris Agreement and prior
international climate change agreements.
9. Define the concept of common but differentiated responsibilities. Why is this
concept of central importance to many developing countries?
10. Go to http://unfccc.int/focus/indc_portal/items/8766.php and download the INDC
of your country. Discuss the objectives and approaches of your country’s INDC.
150
11. What are the challenges and opportunities for climate change policies in
developing countries?
151
Annex 1: Some useful databases
Annex 2: Selected additional reading material
152
References
African Development Bank (2015). Africa’s climate opportunity: adapting and thriving.
African Development Bank. Abidjan.
Aldy JE (2016). Frameworks for evaluating policy approaches to address the
competitiveness concerns of mitigating greenhouse gas emissions. Resources for the
Future Discussion Paper 16-06.
Anderson B, and Di Maria C (2010). Abatement and allocation in the pilot phase of the
EU ETS. Environmental and Resource Economics 48: 83–103.
Archer D, Eby M, Brovkin V, Ridgwell A, Cao L, Mikolajewicz U, Caldeira K,
Matsumoto K, Munhoven G, Montenegro A, and Tokos K (2009). Atmospheric lifetime
of fossil fuel carbon dioxide. Annual Review of Earth and Planetary Sciences 37: 117–34.
Barrett S (2014). Solar geoengineering’s brave new world: thoughts on the governance of
an unprecedented technology. Review of Environmental Economics and Policy 8(2): 249–
69.
Barrett S, and Moreno-Cruz J (2015). The alternatives to unconstrained climate change:
emission reductions versus carbon and solar geoengineering. In: Towards a Workable
and Effective Climate Regime, Barrett S, Carraro C, and de Melo J, eds. CEPR Press and
Ferdi. London.
Barrett S, Carraro C, and de Melo J (2015). Introduction. In: Towards a Workable and
Effective Climate Regime, Barrett S, Carraro C, de Melo J, eds. CEPR Press and Ferdi.
London.
BEE (2015). Factsheet – Renewables from Germany.
http://www.energiewende2015.com/wp-content/uploads/2015/03/Factsheet-Renewablesfrom-Germany.pdf.
Bel G, and Joseph S (2015). Emission abatement: untangling the impacts of the EU ETS
and the economic crisis. Energy Economics 49: 531–33.
Bossetti V (2015). The role of renewables in the pathway towards decarbonisation. In:
Towards a Workable and Effective Climate Regime, Barrett S, Carraro C, and de Melo J,
eds. CEPR Press and Ferdi. London.
Burton I (2006). Adaptation to climate change – international policy options. Pew Center
on Global Climate Change. Arlington, VA.
Collier P (2015). Curbing carbon without curbing development. In: Towards a Workable
and Effective Climate Regime, Barrett S, Carraro C, and de Melo J, eds. CEPR Press and
Ferdi. London.
153
Edenhofer O, Kadner S, Stechow C, and Mi J (2015). Beyond the 2°C limit: facing the
economic and institutional challenge. In: Towards a Workable and Effective Climate
Regime, Barrett S, Carraro C, and de Melo J, eds. CEPR Press and Ferdi. London.
Ellerman AD, Convery FJ, and De Perthuis C (2010). Pricing Carbon: The European
Union Emissions Trading Scheme. Cambridge University Press. Cambridge, UK.
European Commission (2015). The EU emissions trading system (EU ETS). Available at:
http://ec.europa.eu/clima/policies/ets/index_en.htm.
Field CB, Barros VR, Mach KJ, Mastrandrea MD, van Aalst M, Adger WN, Arent DJ,
Barnett J, Betts R, Bilir TE, Birkmann J, Carmin J, Chadee DD, Challinor AJ, Chatterjee
M, Cramer W, Davidson DJ, Estrada YO, Gattuso JP, Hijioka Y, Hoegh-Guldberg O, and
Flannery BP (2015). The state of climate negotiations. In: Towards a Workable and
Effective Climate Regime, Barrett S, Carraro C, and de Melo J, eds. CEPR Press and
Ferdi. London.
Flannery BP (2016). Carbon taxes, trade, and border tax adjustments. Resources for the
Future Policy Brief 16-02.
Georgiev A, Alessi M, Egenhofer C, and Fujiwara N (2011). The EU emission trading
system and climate policy towards 2050. Center for European Policy Studies. Brussels.
Gillingham K, Kotchen, MJ, Rapson DS, and Wagner G (2013). Energy policy: the
rebound effect is overplayed. Nature 493: 475–76.
Global CCS Institute (2015). The global status of CCS: 2015. Summary Report.
Melbourne, Australia.
Government of Kenya (2012). National climate change mitigation analysis. Ministry of
Environment and Natural Resources. Nairobi.
Government of Nepal (2011). National Framework on Local Adaptation Plans for Action.
Government of Nepal, Ministry of Environment. Singhdurbar.
IEA (2011). 25 Energy efficiency policy recommendations – 2011 update. International
Energy Agency. Paris.
IPCC (2014a). Climate Change 2014: Mitigation of Climate Change. Contribution of
Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, Edenhofer, O, Pichs-Madruga R, Sokona Y, Farahani E, Kadner S,
Seyboth K, Adler A, Baum I, Brunner S, Eickemeier P, Kriemann B, Savolainen J,
Schlömer S, von Stechow C, Zwickel T, and Minx JC, eds. Cambridge University Press.
Cambridge, UK and New York.
IPCC (2014b). Climate Change 2014: Synthesis Report. Contribution of Working Groups
I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate
154
Change, Pachauri RK, and Meyer LA, eds. Cambridge University Press. Cambridge, UK
and New York.
IPCC (2014c). Annex II: glossary. In: Climate Change 2014: Synthesis Report.
Contribution of Working Groups I, II and III to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change, Pachauri RK, and Meyer LA, eds.
Cambridge University Press. Cambridge, UK and New York.
Kaudia AA (2015). Policy options in low-income countries: achieving socially
appropriate climate change response objectives. In: Towards a Workable and Effective
Climate Regime, Barrett S, Carraro C, and de Melo J, eds. CEPR Press and Ferdi.
London.
Kortum S, and Weisbach D (2016). Border adjustments for carbon emissions basic
concepts and design. Resources for the Future Discussion Paper 16-09.
Mai F, Keen M, Papaioannou MG, Parry I, Pattillo C, and Ter-Martirosyan A (2016).
After Paris: fiscal, macroeconomic and financial implications of climate change. IMF
Staff Discussion Note 16/01. International Monetary Fund. Washington, DC.
Mansur ET (2012). Upstream versus downstream implementation of climate policy. In:
The Design and Implementation of U.S. Climate Policy, Fullerton D, and Wolfram C, eds.
University of Chicago Press. Chicago.
McClellan J, Keith DW, and Apt J (2012). Cost analysis of stratospheric albedo
modification delivery systems. Environmental Research Letters 7(3).
Mekonnen A (2015). A view from Africa. In: Towards a Workable and Effective Climate
Regime, Barrett S, Carraro C, and de Melo J, eds. CEPR Press and Ferdi. London.
Mimura N, Pulwarty RS, Duc DM, Elshinnawy I, Redsteer MH, Huang HQ, Nkem JN,
Sanchez, and Rodriguez RA (2014). Adaptation planning and implementation. In:
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, Field CB, Barros VR, Dokken DJ, Mach
KJ, Mastrandrea MD, Bilir TE, Chatterjee M, Ebi KL, Estrada YO, Genova RC, Girma
B, Kissel ES, Levy AN, MacCracken S, Mastrandrea PR, and White LL, eds. Cambridge
University Press. Cambridge, UK and New York.
Noble IR, Huq S, Anokhin YA, Carmin J, Goudou D, Lansigan FP, Osman-Elasha B, and
Villamizar A (2014). Adaptation needs and options. In: 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, Field CB, Barros VR, Dokken DJ, Mach KJ, Mastrandrea MD, Bilir
TE, Chatterjee M, Ebi KL, Estrada YO, Genova RC, Girma B, Kissel ES, Levy AN,
155
MacCracken S, Mastrandrea PR, and White LL eds. Cambridge University Press.
Cambridge, UK and New York.
OECD (2012). Energy and Climate Policy Bending the Technological Trajectory.
Organisation for Economic Co-operation and Development Publishing. Paris.
Oxfam (2009). Climate change, poverty and adaptation in Bolivia – case study. Available
at: https://www.pembina.org/reports/c4d-case-study-oxfam.pdf.
Parry I, Veung C, and Heine D (2014). How much carbon pricing is in countries’ own
interests? The critical role of co-benefits. IMF Working Paper 14/174. International
Monetary Fund. Washington, DC.
Sauter C, Grether JM, and Mathys N (2015). Geographical spread of global emissions:
within-country inequalities are increasing. IRENE Working Paper No. 15-01. Neuchâtel.
Sauter C, Grether JM, and Mathys N (2016). Geographical spread of global emissions:
Within-country inequalities are large and increasing. Energy Policy 89: 138–49.
Smith KR, Woodward A, Campbell-Lendrum D, Chadee DD, Honda Y, Liu Q, Olwoch
JM, Revich B, and Sauerborn R (2014). Human health: impacts, adaptation, and cobenefits. In: 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, Field CB, Barros VR,
Dokken DJ, Mach KJ, Mastrandrea MD, Bilir TE, Chatterjee M, Ebi KL, Estrada YO,
Genova RC, Girma B, Kissel ES, Levy AN, MacCracken S, Mastrandrea PR, and White
LL, eds. Cambridge University Press. Cambridge, UK and New York.
Stavins R, Zou J, Brewer T, Conte Grand M, den Elzen M, Finus M, Gupta J, Höhne N,
Lee MK, Michaelowa A, Paterson M, Ramakrishna K, Wen G, Wiener J, and Winkler H
(2014). International cooperation: agreements and instruments. In: Climate Change 2014:
Mitigation of Climate Change. Contribution of Working Group III to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change, Edenhofer O,
Pichs-Madruga R, Sokona Y, Farahani E, Kadner S, Seyboth K, Adler A, Baum I,
Brunner S, Eickemeier P, Kriemann B, Savolainen J, Schlömer S, von Stechow C,
Zwickel C, and Minx JC, eds. Cambridge University Press. Cambridge, UK and New
York.
Stern S (2015). Why Are We Waiting? The Logic, Urgency, and Promise of Tackling
Climate Change. MIT Press. Cambridge, MA.
Sterner T (2007). Fuel taxes: an important instrument for climate policy. Energy Policy
35: 3194–202.
156
Sterner T, and Köhlin G (2015). Pricing carbon: the challenges. In: Towards a Workable
and Effective Climate Regime, Barrett S, Carraro C, and de Melo J, eds. CEPR Press and
Ferdi. London.
Tavoni M (2015). Carbon capture and storage: promise and delusion? In: Towards a
Workable and Effective Climate Regime, Barrett S, Carraro C, and de Melo J, eds. CEPR
Press and Ferdi, London.
Toman M (2015). International cooperation in advancing energy technologies for deep
decarbonisation. In: Towards a Workable and Effective Climate Regime, Barrett S,
Carraro C, and de Melo J, eds. CEPR Press and Ferdi. London.
UNCTAD (2015). Trade and climate change policy beyond 2015. Policy Brief No. 36.
United Nations Conference on Trade and Development. Geneva.
UNFCCC (2007). Impacts, Vulnerabilities and Adaptation in Developing Countries.
United Nations Framework Convention on Climate Change Secretariat. Bonn.
UNFCCC (2016). Paris Agreement - status of ratification. United Nations Framework
Convention on Climate Change. Available at:
http://unfccc.int/paris_agreement/items/9444.php.
Wang X, and Murisic M (2015). Taxing carbon: Current state of play and prospects for
future developments. In: Towards a Workable and Effective Climate Regime, Barrett S,
Carraro C, and de Melo J, eds. CEPR Press and Ferdi. London.
World Bank (2015). State and Trends of Carbon Pricing 2015. World Bank. Washington,
DC.
World Energy Council (2013). World energy perspective - energy efficiency policies:
what works and what does not. World Energy Council. London.
157