Instant ebooks textbook Carbon Abatement Costs and Climate Change Finance 1st Edition William R. Cline download all chapters
Instant ebooks textbook Carbon Abatement Costs and Climate Change Finance 1st Edition William R. Cline download all chapters
Instant ebooks textbook Carbon Abatement Costs and Climate Change Finance 1st Edition William R. Cline download all chapters
https://ebookultra.com
https://ebookultra.com/download/carbon-abatement-
costs-and-climate-change-finance-1st-edition-
william-r-cline/
https://ebookultra.com/download/climate-change-and-carbon-markets-a-
handbook-of-emissions-reduction-mechanisms-farhana-yamin/
ebookultra.com
https://ebookultra.com/download/climate-change-measures-and-trade-
considerations-1st-edition-george-r-fried/
ebookultra.com
https://ebookultra.com/download/climate-change-a-multidisciplinary-
approach-2nd-edition-william-james-burroughs/
ebookultra.com
https://ebookultra.com/download/climate-change-and-climate-modeling-
neelin-j-d/
ebookultra.com
Carbon Risk and Green Finance 1st Edition Aaron Ezroj
https://ebookultra.com/download/carbon-risk-and-green-finance-1st-
edition-aaron-ezroj/
ebookultra.com
https://ebookultra.com/download/understanding-climate-change-
feedbacks-1st-edition-panel-on-climate-change-feedbacks/
ebookultra.com
https://ebookultra.com/download/adjudicating-climate-change-state-
national-and-international-approaches-1st-edition-william-c-g-burns/
ebookultra.com
https://ebookultra.com/download/human-impacts-on-weather-and-
climate-2nd-edition-william-r-cotton/
ebookultra.com
https://ebookultra.com/download/climate-and-culture-change-in-north-
america-ad-900-1600-1st-edition-william-c-foster/
ebookultra.com
Carbon Abatement Costs and Climate Change Finance
1st Edition William R. Cline Digital Instant Download
Author(s): William R. Cline
ISBN(s): 9780881326307, 0881326305
Edition: 1
File Details: PDF, 1.11 MB
Year: 2011
Language: english
96 Policy Analyses in International Economics
July 2011
Carbon Abatement
Costs and Climate
Change Finance
William R. Cline
PE TERSON I N S T I T U T E F O R I N T E R N AT I O N A L E CO N O M I C S
Carbon Abatement
Costs and Climate
Change Finance
Carbon Abatement
Costs and Climate
Change Finance
William R. Cline
PE TERSON INS T I T U T E F O R I N T E R N AT I O N A L E CO N O M I C S
Washington, D C
Ju ly 2011
William R. Cline has been a senior fellow at Copyright © 2011 by the Peter G. Peterson
the Peterson Institute for International Eco- Institute for International Economics. All
nomics since its inception in 1981 and has held rights reserved. No part of this book may be
a joint appointment at the Center for Global reproduced or utilized in any form or by any
Development since 2002. During 1996–2001 means, electronic or mechanical, including
while on leave from the Institute, he was dep- photocopying, recording, or by information
uty managing director and chief economist of storage or retrieval system, without permission
the Institute of International Finance. He is from the Institute.
the author of 23 books, most recently Financial
Globalization, Economic Growth, and the Crisis of For reprints/permission to photocopy please
2007–09 (2010). He was senior fellow, Brook- contact the APS customer service department
ings Institution (1973–81); deputy director of at Copyright Clearance Center, Inc., 222
development and trade research, US Treasury Rosewood Drive, Danvers, MA 01923; or email
Department (1971–73); Ford Foundation visit- requests to: [email protected]
ing professor in Brazil (1970–71); and lecturer
and assistant professor of economics at Princ- Printed in the United States of America
eton University (1967–70). 12 11 10 5 4 3 2 1
The views expressed in this publication are those of the author. This publication is part of
the overall program of the Institute, as endorsed by its Board of Directors, but does not
necessarily reflect the views of individual members of the Board or the Advisory Committee.
Contents
Preface ix
Acknowledgments xiii
1 Overview 1
Policy Context 1
Method and Plan of the Study 2
Principal Findings 3
v
5 Abatement Costs through 2050 33
Copenhagen Convergence 34
Abatement Costs under the Copenhagen Convergence Scenario 39
Other Model–Based Estimates 48
Alternative Abatement Paths 50
8 Synthesis 81
The $100 Billion Copenhagen Commitment 81
Estimates of This Study 83
Appendices
Appendix A Data Sources and Further Statistical Tables 91
Appendix B Estimating Atmospheric CO2 Concentrations 99
under Alternative Emissions Paths
Appendix C Other Greenhouse Gases and Aerosols 103
Appendix D Alternative Policy Paths for CO2 Emissions, 2010–50 107
Appendix E Abatement Costs of the Alternative Policy Paths 113
Appendix F Cost-Minimizing Reallocation of Abatement over Time 117
Appendix G Abatement Cost Function Estimates Based on EMF 22 121
References 131
Index 135
Tables
Table 2.1 Carbon dioxide emissions: Growth decomposition, 10
1990–2006
Table 2.2 Business as usual baseline emissions of carbon dioxide 13
Table 2.3 Decomposition of emissions growth, 2010–20 14
Table 2.4 Decomposition of emissions growth, 2020–30 15
Table 2.5 Business as usual baseline, 2030–50 17
Table 3.1 Country targets and initiatives under the Copenhagen 21
Accord
Table 4.1 Multiplicative abatement cost parameter (a) in the 27
Nordhaus RICE model
vi
Table 4.2 Abatement cost function parameters, EMF 22 28
synthesis model
Table 4.3 Ackerman et al. abatement cost parameters from 31
McKinsey for 2030
Table 5.1 Business as usual baseline and Copenhagen Convergence 37
target emissions, 2020 and 2050
Table 5.2 Business as usual baseline and Copenhagen Convergence 38
abatement path: Annual rates of growth in carbon plus
energy efficiency
Table 5.3 Percent reduction in emissions from business as usual 40
baseline, Copenhagen Convergence scenario
Table 5.4 Abatement costs for the Copenhagen Convergence 41
policy path: RICE model basis
Table 5.5 Abatement costs for the Copenhagen Convergence 44
policy path: EMF 22 synthesis model basis
Table 5.6 Abatement costs in 2030 for the Copenhagen Convergence 47
policy path: McKinsey/Ackerman et al. basis
Table 5.7 2050 emissions under Copenhagen Convergence and 54
UNDP scenarios
Table 5.8 Cumulative CO2 emissions after 2010 and atmospheric 55
concentrations
Table 6.1 Shadow price of CO2 under Copenhagen Convergence 60
abatement in the absence of international offsets trading
Table 6.2 Purchases or sales of CO2 emissions rights, global shadow 64
price, and savings from emissions trading: RICE cost
function basis
Table 6.3 Global CO2 emissions under alternative variants of the 67
Copenhagen Convergence policy scenario
Table 6.4 Present value of abatement costs under alternative 69
Copenhagen Convergence scenarios: RICE cost basis
Table 7.1 Annual investment required for Copenhagen Convergence 77
abatement, 2020–40
Table 8.1 Abatement costs for the Copenhagen Convergence 84
policy path
Table 8.2 Developing-country financing needs for abatement and 85
adaptation and financial flows from offset purchases
Table A.1 Population 92
Table A.2 Baseline per capita GDP 94
Table A.3 Business as usual baseline CO2 emissions per capita 96
Table A.4 Business as usual baseline energy consumption 97
Table C.1 Radiative forcing of other greenhouse gases and aerosols 104
by 2050 under business as usual low scenario B1p
Table D.1 CO2 emissions path under Copenhagen Convergence 108
scenario
Table D.2 CO2 emissions path under UNDP (2007) 109
vii
Table D.3 CO2 emissions path under Chakravarty et al. (2009) 110
Table D.4 CO2 emissions path under Frankel (2008) 111
Table E.1 Abatement costs: UNDP (2007) 113
Table E.2 Abatement costs: Chakravarty et al. (2009) 115
Table E.3 Abatement costs: Frankel (2008) 116
Table F.1 Emissions profile under cost-minimization discounting 118
at 1.5 percent per year
Table F.2 Emissions profile under cost-minimization discounting 119
at 3 percent per year
Table F.3 Emissions profile under cost-minimization discounting 120
at 5 percent per year
Table G.1 Regression results for abatement cost function 124
Table G.2 Abatement costs for the Copenhagen Convergence 127
policy path: EMF 22 synthesis model basis constrained
Table G.3 Investment required for Copenhagen Convergence 129
abatement: 2020 and 2030
Figures
Figure 4.1 Comparison of RICE and EMF 22 based abatement 29
cost estimates for uniform cuts from baseline
Figure 5.1 Business as usual and Copenhagen Convergence 36
abatement emissions paths, 1990–2050
Figure 5.2 Alternative policy paths for C02 emissions, 2010–50 53
Figure B.1 Business as usual emissions and increased atmospheric 100
concentrations: SRES scenario estimates and regression
equation
Figure G.1 Abatement cost with purchases of emissions rights 126
viii
Preface
In this study, William Cline continues his work on climate change that began
with his path-breaking 1992 book, The Economics of Global Warming, and includes
his most recent book Global Warming and Agriculture: Impact Estimates by Country
(2007). This new analysis focuses on the abatement costs likely to be required
to keep global warming within the internationally endorsed ceiling of 2°C
warming above preindustrial temperatures. Cline employs three leading abate-
ment cost models to examine a “Copenhagen Convergence” policy scenario
in which the major nations achieve their 2020 emissions curbs pledged at
Copenhagen in December 2009 and thereafter converge to uniform low per
capita emissions by 2050.
The momentum for action on greenhouse gas abatement has faltered
recently, at least in the United States with the failure of Senate action on legis-
lation on emissions restraint after the Waxman-Markey bill passed the House
of Representatives in 2009. The estimates in this study should provide some
reassurance that forceful action is possible without imposing prohibitive
economic costs. The central model estimates indicate that abatement costs
would amount to about one-third to two-thirds of one percent of GDP for
industrial countries by 2030 and moderately less for developing countries.
But participation of the major emerging-market economies will be necessary
for success and by 2050 the cutbacks from the business-as-usual baseline for
China and some others among them are surprisingly close to the deep cuts in
industrial countries—because of the rapid rise of emissions otherwise associ-
ated with rapid growth.
The Institute has sought to contribute to the debate on the new inter-
national economic architecture needed to curb global warming. In 2008, the
Institute published Leveling the Carbon Playing Field by Visiting Fellow Trevor
ix
Houser and coauthors from the World Resources Institute. Senior Fellows
Gary Hufbauer and Jeffrey Schott have written working papers on World
Trade Organization (WTO) rules relating to climate change and the prospects
for US-NAFTA cooperation on abatement, and Hufbauer, Steve Charnovitz,
and Jisun Kim published a major book in 2009 entitled Global Warming and the
World Trading System.
The Peter G. Peterson Institute for International Economics is a private,
nonprofit institution for the study and discussion of international economic
policy. Its purpose is to analyze important issues in that area and to develop
and communicate practical new approaches for dealing with them. The
Institute is completely nonpartisan.
The Institute is funded by a highly diversified group of philanthropic foun-
dations, private corporations, and interested individuals. About 35 percent
of the Institute’s resources in our latest fiscal year was provided by contribu-
tors outside the United States. The Doris Duke Charitable Foundation has
provided generous support for much of the Institute’s recent work on climate
change, including the present study. In 2009–10 the Foundation provided
support to focus on developing countries’ financing needs for climate action.
The present study translates the abatement cost estimates into prospective
financing needs (chapter 7), including for adaptation. It turns out that the
benchmark Copenhagen Accord figure of $100 billion annually by 2020 is
broadly supported by the calculations developed here.
The Institute’s Board of Directors bears overall responsibilities for the
Institute and gives general guidance and approval to its research program,
including the identification of topics that are likely to become important
over the medium run (one to three years) and that should be addressed by
the Institute. The director, working closely with the staff and outside Advisory
Committee, is responsible for the development of particular projects and
makes the final decision to publish an individual study.
The Institute hopes that its studies and other activities will contribute to
building a stronger foundation for international economic policy around the
world. We invite readers of these publications to let us know how they think we
can best accomplish this objective.
C. Fred Bergsten
Director
June 2011
x
PETER G. PETERSON INSTITUTE FOR INTERNATIONAL ECONOMICS
1750 Massachusetts Avenue, NW, Washington, DC 20036-1903
(202) 328-9000 Fax: (202) 659-3225
C. Fred Bergsten, Director
Ex officio
* C. Fred Bergsten
Nancy Birdsall
Richard N. Cooper * Member of the Executive Committee
Barry Eichengreen
Honorary Directors
Alan Greenspan
Frank E. Loy
David Rockefeller
George P. Shultz
Acknowledgments
I thank Thomas Emmons and Yimei Zou for painstaking and inspired research
assistance. For comments on an earlier draft, I thank without implicating
Trevor Houser, Caio Koch-Weser, and an anonymous reviewer.
xiii
1
Overview
The major nations have come tantalizingly close to beginning more meaningful
action to curb global warming. The Copenhagen Accord of December 2009,
ratified and more formally incorporated into the United Nations climate nego-
tiations system at Cancún, Mexico, in December 2010, set a target of limiting
warming to 2°C above preindustrial levels. For the first time, the Copenhagen
Accord took the crucial step of incorporating the major developing countries
into an international action program. The sole previous (and still extant) inter-
nationally agreed mechanism, the Kyoto Protocol, was incapable of meaningfully
curbing global warming because its limits on emissions applied only to industrial
countries, yet the share of these countries in global emissions is already slightly
less than half and under business as usual will shrink to less than 40 percent by
2030 and about 30 percent by 2050 (tables 2.2 and 2.5 in chapter 2).
Policy Context
The eventual damages from unrestrained global warming could be severe. With
3°C warming by late in this century, agricultural production potential would
fall by 3 to 16 percent globally, 17 to 28 percent in Africa, 29 to 38 percent in
India, 20 to 30 percent in Pakistan, 9 to 21 percent in the Middle East and
North Africa, and 13 to 24 percent in Latin America, depending on whether
“carbon fertilization” benefits partially offset heat stress (Cline 2007, pp. 69,
96). With even 2°C warming, on a time scale of centuries, the Greenland ice cap
would likely melt, raising sea level 7 meters. With considerably higher warming
and on a time scale of millennia, sea levels could rise 60 to 80 meters, the
oceans become anoxic from shutdown of the ocean conveyor belt, and anaer-
obic bacteria emit hydrogen sulfide in toxic amounts, a mechanism that may
have caused the Permian-Triassic mass extinction 251 million years ago (Cline
1
2011, Kump et al. 2005). Stakes and time scales such as these bedevil fine-point
calculations of optimal abatement and argue for an insurance approach that
focuses on even small probabilities of disastrous impacts (Weitzman 2007).
The central policy questions then become: How expensive is insurance
against global warming, and who will pay the insurance policy premiums? This
study seeks to examine the prospective costs of abatement through 2050 at the
global level and in detail for the 25 largest economies (with the European Union
treated as a single economy). It also attempts to shed light on the magnitude of
abatement investment requirements and adaptation costs in developing coun-
tries by 2020 and to relate them to the target of $100 billion in international
support pledged by industrial countries in the Copenhagen Accord.
1. The RICE model developed by William Nordhaus (2010b); those of McKinsey (2009) as compiled
by Ackerman et al. (2010); and specially estimated equations based on model results compiled in
Energy Modeling Forum 22 (see appendix G).
Principal Findings
Among the most important findings of this study are the following:
n The Copenhagen pledges for 2020 would cut global carbon dioxide emis-
sions by 9 percent from the bau baseline, from 35.9 to 32.7 billion tons of
carbon dioxide (GtCO2).
2. This path would require somewhat deeper cuts for the European Union and United States by
2050 than the two offered in their Copenhagen Accord submissions. Note also that the Copenhagen
Convergence scenario limits concentrations by 2050 to somewhat less than 450 ppm for carbon
dioxide alone, but somewhat more for CO2-equivalent of all greenhouse gases, as discussed later.
3. These include the influential proposal by the United Nations Development Program (UNDP
2007) as well as two leading proposals by academic researchers (Chakravarty et al. 2009, Frankel
2008). Using the RICE model cost parameters, a range of abatement costs corresponding to the
range of policy scenarios is explored.
OVERVIEW 3
n To achieve a path limiting warming to 2°C and atmospheric concentra-
tions of CO2 to 450 ppm, it will be necessary to cut 2050 emissions from
a baseline of 53.2 to 13.3 GtCO2 (a reduction of 75 percent from baseline)
(table 5.1 in chapter 5).
n In 1990–2006, efficiency of output relative to energy use rose at 1.4 percent
annually for the 25 largest economies, while energy efficiency per unit
of carbon required did not grow at all. From 2010 to 2030, baseline EIA
projections indicate 2 percent annual output/energy efficiency growth but
again zero energy/carbon efficiency growth. To achieve the needed emis-
sions cuts from 2020 to 2050, it will be necessary for the sum of these two
efficiency growth rates to reach 6 percent annually (table 5.2). Most of
this acceleration likely will need to come from a shift to carbon-spare or
carbon-free energy.
n To mobilize global political support for abatement, particularly in devel-
oping countries, it would be appropriate after 2020 to pursue international
convergence to uniform per capita emissions by 2050. The required 2050
level would be 1.43 tons of carbon dioxide (tCO2) per person, compared
with today’s levels of 19.3 tons in the United States, 5.0 tons in China, and
1.4 tons in India.
n This Copenhagen Convergence path would be only modestly more ambi-
tious than the 2050 cutbacks of 80 percent from 1990 levels pledged at
Copenhagen by the European Union and 83 percent from 2005 levels
pledged by the United States (reaching, respectively, 84 and 89 percent
from 1990 and 2005 levels).
n The 75 percent global cutback from bau baseline by 2050 would represent
89 percent cuts for industrial countries on average and 69 percent cuts for
developing countries. Because of their rapid baseline growth in output
and carbon, such emerging-market economies as China, Malaysia, South
Africa, and South Korea would need to cut emissions by 85 to 90 percent
from the 2050 baseline, surprisingly similar in depth to cuts needed by
industrial countries.
n Despite these seemingly daunting cutbacks, on the basis of the leading
abatement cost models it turns out that the costs would be modest. By
2030, in the prominent RICE model abatement costs would amount to
about one-fourth of one percent of GDP (0.28 percent for industrial coun-
tries, 0.19 percent for developing countries); by 2050 the RICE-based esti-
mate reaches 1.2 percent of world product (1.6 percent industrial countries,
1.0 percent developing countries; table 5.4). Costs in the leading bottom-
up model (McKinsey) are considerably lower. Costs are somewhat higher in
a family of regional cost functions estimated from results of the Stanford
Energy Modeling Forum (EMF) 22 model survey, but not radically so. In
the preferred variant, the EMF 22–based costs by 2030 amount to 0.67 per-
cent of GDP for industrial countries and 0.62 percent for developing coun-
tries (table G.2 in appendix G).
The broad implication of the estimates of this study is that that although
the costs and financing requirements for aggressive international action
limiting climate change will be substantial, they should be of a sufficiently
manageable magnitude that the outcome is likely to depend more on political
will rather than economic feasibility.
4. Investment would be needed already by 2020 in China despite no change from baseline in that
year in order to set on track the abatement needed during 2020–30 to follow the Copenhagen
Convergence path. The abatement cost estimate used as the basis for the investment calculation
is the average of the estimates from the RICE model and the preferred variant of the EMF 22 esti-
mates.
OVERVIEW 5
2
Baseline Emissions under
Business as Usual
Framework
For any of the major emitting countries, let Et = emissions of carbon dioxide
(excluding from deforestation, not considered in the present study) in year t.
7
Let Nt = population in year t. Let qt = real GDP per capita, at 2005 constant
international purchasing power parity (ppp) dollars. Let t = output per unit
of energy, or “energy efficiency.” Let t = energy per unit of carbon dioxide, or
“carbon efficiency.” Referring to constant rates over a period of a decade, for
the decade and country in question the path of population will follow:
Nt = N0ent (2.1)
where n is the rate of growth of population, subscript 0 denotes the year at the
beginning of the decade, and e is the base of the natural logarithm. Similarly,
output per capita will follow the path:
qt = q0egt (2.2)
where g is the growth rate of per capita income. Total production will then
equal:
Qt = Ntqt (2.3)
Total carbon dioxide emissions will then equal total output divided by
the energy efficiency parameter (output per unit of energy) times the carbon
efficiency parameter (energy per unit of carbon), or:
Qt
Et = (2.4)
OtJt
The time path for the energy and carbon efficiency parameters, in turn,
will be:
Ot = O0ewt (2.5)
and
Jt = J0ect (2.6)
where w is the annual rate of increase in energy efficiency of output and c is the
annual rate of increase in carbon efficiency of energy.
Substituting, the time path for carbon dioxide emissions will be:
N0entq0egt
Et =
O0ewtJ0ect
(2.7)
N0q0
= e(n+g–w–c)t = E0e(n+g–w–c)t
O0J0
The annual rate of change of emissions will then be the ratio of the deriva-
tive of this expression to the expression itself:
1. For the 2006–10 five-year plan, China pledged to reduce energy use per unit of GDP by 20
percent. This amounted to a pace of 3.7 percent per year, compared with 4.8 percent achieved in
1990–2006 (table 2.1).
Efficiency
Level, 2006
Growth rates, 1990–2006 (annual percent) parameters, 2006b
Country/ (million
economy metric tons) Total n g w c O
J
Argentina 173 2.70 1.13 2.76 1.36 –0.17 6.66 0.40
Australia 372 1.48 1.18 1.98 1.07 0.60 5.38 0.33
Brazil 352 3.27 1.48 1.23 –0.51 –0.05 7.47 0.64
Canada 544 1.19 1.01 1.75 1.18 0.39 4.32 0.50
China 6,099 5.79 0.84 8.84 4.82 –0.93 3.16 0.31
Egypt 167 4.91 1.87 2.20 –0.13 –0.71 5.37 0.37
European Union 4,050 –0.23 0.22 1.96 1.57 0.83 7.17 0.45
India 1,509 4.89 1.74 4.33 2.51 –1.32 4.73 0.37
Indonesia 333 4.97 1.52 2.96 1.03 –1.53 4.32 0.54
Iran 467 4.50 0.86 2.76 –2.07 1.19 3.70 0.37
Japan 1,293 0.61 0.20 1.12 0.24 0.47 7.50 0.41
Kazakhstan 193 –1.88 –0.46 1.85 2.52 0.75 2.36 0.32
Malaysia 188 7.49 2.07 3.80 –0.84 –0.78 4.36 0.36
Mexico 436 0.78 1.47 1.59 0.77 1.51 7.89 0.41
Pakistan 143 4.57 2.28 1.89 0.41 –0.81 4.76 0.56
Russia 1,564 –1.77 –0.25 0.08 1.46 0.13 2.69 0.43
Saudi Arabia 381 3.58 3.25 0.73 –1.45 1.85 3.95 0.38
South Africa 414 1.36 1.38 0.78 –0.05 0.85 3.27 0.31
South Korea 475 4.22 0.72 4.63 0.10 1.03 5.31 0.46
Taiwan 273 4.82 0.73 4.52 0.43 –0.01 5.62 0.38
Thailand 272 6.53 0.99 3.60 –0.77 –1.17 4.19 0.38
Turkey 269 3.80 1.66 2.47 0.53 –0.21 9.09 0.35
Ukraine 319 –4.07 –0.64 –1.81 1.33 0.29 2.05 0.43
United States 5,748 1.04 1.10 1.81 1.75 0.12 5.49 0.40
Venezuela 171 2.12 1.77 0.73 0.32 0.06 4.44 0.36
BASELINE EMISSIONS UNDER BUSINESS AS USUAL 11
2. Specifically estimated by EIA for the largest economies and based on regional trends for others;
see appendix A. Note that “European Union” in the table is for the 27 current members of the
European Union. The EU estimate is calculated as the base level for 2007 for these countries specif-
ically, and then for 2020 and 2030 it is based on the proportionate emissions growth for “OECD
Europe” in the EIA projections.
3. Data for major industrial countries are from the EIA estimates. For countries without detailed
estimates in the EIA source, the energy, economic growth, and emissions projections are based on
averages for the regions in which the countries are located.
4. Actual CO2 emissions growth from 2000 to 2006 averaged 3.1 percent (van Vuuren and Riahi
2008, 241). In comparison, the highest of six scenarios considered by the IPCC (A1B) called for
3.4 percent annual growth from 2000 to 2010; the median scenario rate was 2.0 percent. Scenario
A2 was at the median through 2010 but second highest by 2050, with average annual growth for
2000–50 amounting to 1.7 percent (versus 2.4 percent in the highest path, A1T). Calculated from
IPCC (2001, 801).
5. However, in scenario A2 deforestation and land use still account for about 4 GtCO2 by 2030 and
about 3 GtCO2 by 2050.
In December 2009, heads of state met in Copenhagen for the 15th Conference
of Parties of the United Nations Framework Convention on Climate Change
(UNFCCC). In the resulting Copenhagen Accord they pledged targets for
emissions reductions by 2020. For the first time, major emerging-market and
developing countries undertook abatement pledges. Their previous omission
had been a serious shortcoming of the Kyoto Protocol. Subsequently incor-
porated more formally into the UNFCCC at the 16th Conference of Parties
in November–December 2010 in Cancún, Mexico, the Copenhagen Accord
pledges provide the best basis for identifying prospective abatement paths
through 2020. Together with policy scenarios developed in this study for
subsequent international action, they serve as the basis for the abatement cost
estimates compiled in chapters 5 and 6.
As set forth in chapter 5, the 2020 targets are only a modest beginning
in the task of limiting global warming. They would reduce global emis-
sions from about 35.9 billion tons of carbon dioxide (GtCO2) under busi-
ness as usual (bau) to 32.7 GtCO2 (table 5.1 in chapter 5), a cutback of only
9 percent. Reductions would be 17 percent from 2005 levels for the United
States and 20 percent below 1990 levels for the European Union. Because bau
emissions paths are virtually flat through 2020 for these two leading world
economies, the corresponding cutbacks from baseline are both approximately
17 percent as well (tables 2.2 and 5.3). Among emerging-market economies,
Brazil, Indonesia, Mexico, South Korea, and South Africa pledged cutbacks
of approximately one-third from the 2020 baselines. In contrast, China and
India pledged cutbacks in carbon intensity of GDP (by 40 to 45 percent and
20 to 25 percent from 2005 levels, respectively). However, because of antici-
pated increases in energy efficiency even in the bau baseline, the resulting emis-
19
sions levels for these two largest emerging-market nations would not require
cutbacks from baseline. Even so, the limits would constrain any tendency
toward “carbon leakage” through increased carbon-based export activity that
could result from complete free-riding.
third from bau baselines otherwise attained by 2020. In the case of Brazil, about
two-thirds of the overall cut of 36 to 39 percent was to come from forestry and
agricultural measures. Considering that deforestation accounts for about half
ABATEMENT INITIATIVES 21
of Brazil’s emissions, the implication is that industrial emissions (the emis-
sions considered in the analysis of this study) would be cut from the bau base-
line by about 24 percent.1
China and India chose the other main alternative of the major emerging-
market economies: statement of a target expressed as a reduction in the carbon
intensity of GDP. China pledged a reduction of 40 to 45 percent in carbon
dioxide emissions per unit of economic output by 2020 compared with the
2005 base. It also pledged to increase the share of non-fossil fuels in primary
energy consumption to 15 percent by 2020 and set specific targets for increased
forest coverage. Its submission specifically referred to UNFCCC Article 4, para-
graph 7, which premises developing-country action on industrial-country
implementation of commitments on financial resources and transfer of tech-
nology. For its part, India set as its goal a reduction in the nonagricultural
emissions intensity of GDP by 20 to 25 percent by 2020 from the 2005 level.
India also cited Article 4, paragraph 7.
The bau projections set forth above provide a basis for examining the
meaning of the pledges by China and India. China’s CO2 emissions stood at
5.63 billion tons in 2005 (CDIAC 2009). Its bau baseline is at 9.54 GtCO2 by
2020 (table 2.2). The baseline emissions multiple from 2005 to 2020 is thus
1.69. In comparison, for 2005–20 per capita GDP growth is expected to average
6.9 percent per year (appendix table A.2) and population growth 0.6 percent,
placing total growth at 7.5 percent per year. This growth would increase GDP
by a multiple of 2.96. Shrinking the ratio of emissions to GDP by 42.5 percent
would translate to an expansion of emissions from 2005 to 2020 by a multiple
of 1.70.2 So the pledged reduction in carbon intensity is almost identical to
the expected bau baseline. For purposes of the calculations in this study, it is
assumed that China has zero departure from baseline by 2020 in the central
policy scenario (Copenhagen Convergence) based on the Copenhagen Accord.3
The same calculation for India shows the following. Baseline emissions rise
by a multiple of 1.46, from 1.42 GtCO2 in 2005 (CDIAC 2009) to 2.08 GtCO2
in 2020 (table 2.2). Real GDP grows at 6.3 percent through 2020 (5 percent per
capita plus 1.3 percent population growth), yielding an expansion multiple of
2.50 from 2005 to 2020. Shrinking the GDP expansion by the goal for reduced
1. That is, of a total cut of 36 percent from bau, two-thirds or 24 percent would be in forestry and
agriculture, implying a cut from baseline of 48 percent in these sectors. The remaining 12 percent
of the total would amount to 24 percent of the half of total emissions coming from the industrial
sector.
Cancún Agreements
The 16th Conference of Parties of the UNFCCC in Cancún in effect legiti-
mated within the United Nations framework the politically crucial but more
ad hoc pledges offered by major countries a year before at Copenhagen, where
obstruction by six countries had forced the UNFCCC merely to “take note”
ABATEMENT INITIATIVES 23
of rather than adopt the Copenhagen Accord.7 This time a forceful chairman,
Mexican Foreign Minister Patricia Espinosa, declared that “consensus does
not mean unanimity” and 193 Parties to the UNFCCC adopted the Cancún
Agreements despite the opposition of a single country, Bolivia (Stavins 2010).
Beyond endorsing the Copenhagen pledges, the Cancún agreements
provided for a mechanism to monitor the progress of countries in meeting
the commitments; established the Cancún Adaptation Framework for assis-
tance to vulnerable countries in adjusting to climate change already under way
(including a new climate fund managed by the World Bank); and spelled out
steps for national action plans for curbing deforestation. For the first time
in an official United Nations accord the agreements also adopted the objec-
tive of limiting global warming to no more than 2C above preindustrial levels
(UNFCCC 2010b, Houser 2010b, Stavins 2010).
The European Union and some other parties (including Japan, Australia,
Russia, and Canada) seek to transform the Cancún Agreements and the miti-
gation pledges of the Copenhagen Accord into a legally binding interna-
tional treaty as the successor to the Kyoto Protocol, which expires in 2012. It
seems far more likely, however, that progress will have to occur on the more
informal basis implied by the very name of the Ad Hoc Working Group set
up at Copenhagen. In the pivotal case of the United States, climate legisla-
tion stalled after passage of the Waxman-Markey bill in 2009, as the Great
Recession shifted priorities and landmark health care legislation as well as
financial regulatory reform preempted attention. For some time it may fall to
the Environmental Protection Agency as well as three state-level regional initia-
tives to implement what amounts to a Plan B for mitigation (Cline 2010b).
For the purposes of the present study, the principal implication of the
Cancún Agreements is that they strengthen the international political commit-
ment (albeit not in a legally binding fashion) to the abatement targets that
emerged in the Copenhagen Accord and that are the basis for the estimates for
abatement costs and emissions levels by 2020.
7. UNFCCC (2010a, Decision 2/CP.15). The formal name for the Copenhagen undertakings,
subsequently used frequently in the Cancún Agreements (UNFCCC 2010b), is Ad Hoc Working
Group on Long-term Cooperative Action Under the Convention. At Copenhagen the dissenting
countries were Bolivia, Cuba, Nicaragua, Sudan, Tuvalu, and Venezuela. For an account of the
negotiations at Copenhagen, see Houser (2010a).
This chapter sets forth the next building block in the analysis of this study:
equations and parameters for estimating abatement cost as a function of the
depth of emissions cut from baseline. Three sets of estimates are used in this
study. Two are from the “top down” school, in which overall economic output
is cut back in a nonlinear fashion as emissions are reduced. The widely used
“RICE” model of William Nordhaus (2010b) has a nearly cubic cost function.
In a second set of top-down estimates the survey of other leading models in
the Stanford Energy Modeling Forum (EMF) provides the basis for new esti-
mates of region-specific abatement cost curves developed in the present study
(appendix G). The third set of estimates is from the “bottom up” analysis of
McKinsey (2009), in which the focus is on a sequence of successively more costly
operations to cut emissions (with initial activities such as a shift to fluorescent
lighting having negative cost if institutional obstacles can be overcome).
The broad effect is that the cheapest abatement is found in the McKinsey
curves; next is the RICE abatement cost, which can be very low in initial
cutbacks but mounts rapidly; and highest are the EMF 22 costs, which begin
much higher but rise less rapidly with less-than-quadratic parameters. With
the quantitative estimates of the cost functions obtained in this chapter, it is
possible to apply the abatement policy scenarios examined in the next chapter
to arrive at estimates of prospective costs of international action to limit
climate change.
RICE Model
The form used by Nordhaus (2008) for calculating abatement cost, and also
applied in the new EMF-based estimates developed in the present study, is as
follows:
kt = DtPEt (4.1)
1. Values are from the online version of the RICE model available at http://nordhaus.econ.yale.
edu/RICEmodels.htm. Note that Nordhaus uses 1 and 2 for the terms α and here. The model
provides specific estimates for the following economies: United States, European Union, Japan,
Russia, China, and India. All other estimates shown here are from the relevant regional groupings
in the model: Africa, Latin America, Middle East, Eurasia, Other Asia, and Other High Income.
k = 0.029 × 0.402.8 = 0.0022, or 0.22 percent of GDP. As indicated in the table, for
a given percent cut from baseline, by 2050 the cost is only about half as large
(as a percent of GDP) as it would be in 2020. The declining cost, combined
with a relatively high long-term discount rate, contributes to the “ramp-up”
profi le (relatively moderate initial cuts followed by rising proportional cuts)
of the optimal abatement paths found in studies by Nordhaus (2008, 2010a).
The 2010 version of the RICE model (Nordhaus 2010b), from which table
4.1 is taken, derives the abatement cost parameters from models with national
and regional detail in the Fourth Assessment Report (IPCC 2007b) and the
Energy Modeling Forum (EMF) 22 report (Clarke, Böhringer, and Rutherford
2009), as indicated in Nordhaus (2010a, 6).
EMF 22 Models
Appendix G applies the same functional form of equation 4.1 to estimate cost
functions from the large set of model results compiled in EMF 22. It turns
out that the estimated equations show a strong pattern of higher abatement
costs than those of the RICE model for moderate emissions cutbacks but more
comparable estimates for large cutbacks. Correspondingly, the equations esti-
mated in appendix G show much less nonlinearity than indicated by the RICE
parameters, with the exponent on the proportionate cutback always less than
quadratic rather than nearly cubic.
Table 4.2 reports the EMF-based estimates of the multiplicative cost coef-
ficient (equation 4.1) as well as the exponential coefficient (), which varies
by region and time period rather than being universally a single value (2.8 in
the RICE model). Figure 4.1 illustrates results comparing the RICE cost func-
tion with the EMF-based estimate for selected regions (the United States, the
European Union, China, and Latin America) under standardized cuts from base-
line (20 percent in 2020, 40 percent in 2030, 60 percent in 2040, and 80 percent
in 2050). In all cases the EMF-based equations show higher cost than the RICE
equations. For example, for the United States, a cutback of 40 percent from base-
line in 2030 would cost 1.15 percent of GDP in the EMF-based estimates but
only 0.22 percent of GDP in the RICE-based estimates. The divergence is the
greatest for developing countries. Thus, again for a 40 percent cut in 2030, the
EMF-based cost would be about 2½ percent of GDP for China and for Latin
America, whereas the RICE-based cost would be only 0.36 percent of GDP for
China and 0.25 percent for Latin America.2
2. The EMF-based equations are specific to the region for China but for Latin America are from
the EMF general category “Group 3” of developing countries other than the BRICs (Brazil, Russia,
India, and China).
percent of GDP
6
US-EMF 22
EU-EMF 22
China-EMF 22
5 Latin America-EMF 22
US-RICE
EU-RICE
4 China-RICE
Latin America-RICE
3
ABATEMENT COST FUNCTIONS
0
2020 2030 2040 2050
a. Cuts from baseline: 20 percent in 2020, 40 percent in 2030, 60 percent in 2040, 80 percent in 2050.
Source: Author’s calculations, based on tables 4.1 and 4.2.
29
McKinsey Model
For the bottom-up school, the most prominent abatement cost estimates
are those of McKinsey (2009). Ackerman, Stanton, and Bueno (2010) have
obtained detailed regional estimates from the McKinsey study and developed
summary equations approximating cost curves for 2030 by region. These
curves are of the form:
R (4.2)
z’ = A
B–R
where z' is marginal cost per unit of carbon dioxide reduction, R is the amount
of the reduction, B is the quantity of reduction at which the marginal cost
curve turns vertical (infinite cost for an additional unit of abatement), and A is
a cost parameter. The CRED model of Ackerman, Stanton, and Bueno (2010)
estimates this equation for nine regions, based on the McKinsey estimates.3
Importantly, the equation form chosen by Ackerman et al. does not allow for
the well-known initial negative marginal cost in the McKinsey estimates but
instead treats costs in the initial range as close to zero but not negative.
To obtain an estimate of total cost of abatement, it is necessary to inte-
grate the marginal cost function over the span from zero to the cutback of R.
This integral is:
R
z = z’ = A[B ln B – R] (4.3)
0 B–R
Table 4.3 reports these bottom-up abatement parameters for 2030, in
billions of dollars at 2005 prices (parameter A) and billion tons of carbon (B).4
For example, for the United States in 2030 the reduction at which marginal
cost turns infinite is 1.39 GtC, or 5.1 GtCO2. Considering that the baseline
emissions for 2030 stand at 6.24 GtCO2 (table 2.2), marginal cost turns infi-
nite at 82 percent cutback from baseline. For most countries the estimates in
table 4.3 apply 2030 baseline shares in total emissions for the relevant region
to obtain the B scaled to the country in question.5
3. The nine regions are United States, Europe, Other High Income, Latin America, Middle East,
Russia and non-EU Eastern Europe, Africa, China, and South and Southeast Asia.
4. For industrial emissions only.
5. There is no attempt to estimate the parameters for Rest of World Developing, because in the
calculations later this grouping never carries out abatement because its per capita emissions are
always lower than the 2050 global convergence level. See appendix table A.3.
Avec ce refrain :