ARTICLE IN PRESS

0301-4215/$ - se

doi:10.1016/j.en

$Paper prese

China, May 26�CorrespondE-mail addr

Energy Policy 36 (2008) 1420–1429

www.elsevier.com/locate/enpol

Greenhouse gas emissions trading among Pacific Rim countries:An analysis of policies to bring developing countries

to the bargaining table$

Adam Rose�, Dan Wei

School of Policy, Planning and Development, University of Southern California, Los Angeles, CA 90089, USA

Received 31 August 2007; accepted 3 December 2007

Available online 20 February 2008

Abstract

This paper examines the aggregate net costs and individual country cost savings of greenhouse gas emissions trading among Pacific

Rim countries. We propose emission permit allocation rules designed to entice developing countries to participate. Absence of developing

country involvement has served as an excuse for the lack by participation by the United States in the first compliance period of the Kyoto

Protocol and may serve as a disincentive to even more countries in subsequent periods. Our analysis specifies permit allocation rules that

could result in no net costs, and even cost-savings, to developing countries for their involvement in the emissions trading market, while at

the same time providing extensive benefits to industrialized countries through access to lower-cost mitigation alternatives.

r 2007 Elsevier Ltd. All rights reserved.

Keywords: Cap and trade for greenhouse gas mitigation; Economics of climate policy; International cooperation

1. Introduction

The earth’s mean temperature has been rising steadily inrecent decades, and there is now a strong scientificconsensus that this phenomenon is primarily caused byhuman activity, most prominently the combustion of fossilfuels (IPCC, 2007). In response, some progress has beenmade on the design of policies to mitigate greenhouse gas(GHG) emissions. The Kyoto Protocol, for example, isbased on commitments from most industrialized countries(ICs) and a handful of developing countries (DCs) to capemissions. Individual cities and sub-national regionsaround the world have made commitments as well (Kouskyand Schneider, 2003; Peterson and Rose, 2006). Mitigationactions are currently taking place at all levels indepen-dently and more are about to take place cooperatively asthe first Kyoto Protocol compliance period of 2008–2012 isabout to begin.

e front matter r 2007 Elsevier Ltd. All rights reserved.

pol.2007.12.008

nted at the Fudan University Shanghai Forum, Shanghai,

, 2007.

ing author. Tel.: +12137408022.

ess: adam.rose@usc.edu (A. Rose).

Still, all of these efforts omit more than one-third ofGHG emitters by volume, and, moreover, the committedreductions may be modest in relation to what is needed toprevent serious climate change and its potentially devastat-ing consequences (IPCC, 2007). Although Northeastern,Mid-Atlantic, and Western States of the US havecommitted to GHG reduction, the Bush administrationhas rejected the Kyoto Protocol. Also, most DCs have notmade any mitigation commitments. China and India aremajor sources of GHGs, and, in fact, the former will soonoutpace the US in emitting carbon dioxide.DCs have explained their lack of commitment to

emissions caps with reference to the associated cost burdenand the fact that diversion of resources to address thisproblem would detract from their future economicgrowth.1 They point out that ICs were not required to doso at a similar point of their development and that it is

1DCs are contributing to the solution of the problem in a limited way

through the Kyoto ‘‘flexibility mechanisms,’’ such as the Clean Develop-

ment Mechanism (CDM), whereby they host low-cost mitigation options

sponsored by industrialized countries (e.g., low emitting steel mills and

power plants, or the planting of forests for sequestration).

ARTICLE IN PRESS

4

A. Rose, D. Wei / Energy Policy 36 (2008) 1420–1429 1421

hypocritical for ICs to press them on this issue. In turn, ICspoint to the increasing contribution of DCs to the problem.Many of them feel besieged by cheap imports from DCsand note that IC competitiveness in international marketswill further deteriorate if DCs do not shoulder some of thecost burden.2

The purpose of this paper is to illustrate how ICs andDCs can engage in a mutually beneficial, cooperative effortto address the climate change problem through the policyinstrument of GHG emissions trading. This approachholds the prospect of greatly reducing the overall cost ofachieving GHG emission reductions. This is accomplishedby issuing and then allowing the trading of emissionsrights, so that countries with relatively high mitigationcosts can purchase permits from countries with relativelylow mitigation costs, thus ensuring a least-cost solution.All countries gain from the exchange. Permit purchasingcountries avoid high-cost expenditures, and selling coun-tries reap permit revenues to more than offset the increasein mitigation they take on as they sell permits (see, e.g.,Barrett et al., 1992; Tietenberg, 2003). From most equitystandpoints, conditions underlying the situation work inthe right direction—DCs have relatively more low-costmitigation options, and hence the transfer of monetarypayments moves from relatively rich to relatively poorcountries. Moreover, equity can be fine-tuned by thenegotiation process according to political considerationsand/or various equity principles (Rose and Stevens, 1998).Equity objectives are implemented by the inter-countrydistribution of permits (and hence initial emission caps, orquotas) according to burden-sharing rules that correspondto the equity principles (Rose et al., 1998). In fact, thepolicy instrument can be designed in such a manner as toyield no harm or no risk to DCs, if they are allocated asufficient number of permits. Still, this represents mean-ingful involvement of DCs that can significantly reduce ICcompliance costs by offering them lower-cost alternatives.

More specifically, this paper will examine the aggregatenet costs and individual country cost savings from both theperspectives of cost-effectiveness and fairness of GHGemission trading among Pacific Rim countries.3 This is ageographic area of great significance, consisting of 44percent of the world’s population, 61 percent of its energyuse, and 60 percent of its GHG emissions in 2004 (EIA,2006). Many DCs in the region are undergoing rapideconomic development, so that their total proportion ofthese totals will soon become dominant, e.g., China isprojected to soon outpace the US as the largest emitter ofCO2 (see energy use and GHG emission projections of

2For example, just prior to the signing of the Kyoto Protocol, the US

Senate voted 95 to 0 not to ratify it unless there was meaningful DC

involvement.3Note that our grouping of Pacific Rim countries differs slightly from

the formal APEC membership list. We have included Colombia, Costa

Rica, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, North

Korea, and Panama, but excluded Brunei, Papua New Guinea, and

Vietnam.

countries included in this study in Appendix Table A1). Itis a natural area of cooperation given its long history ofinternational trade, financial, and other institutionaltreaties. Agreements among trading partners help to reducetensions relating to the loss of individual country‘‘competitiveness.’’ They also facilitate brokering, monitor-ing, and enforcement of permit trades. Moreover, thepolicy has the potential of enticing some major players intothe arena. Five Western States of the US (California,Oregon, Washington, Arizona, and New Mexico) locatedon or near its Pacific Coast have agreed to a cap-and-tradeagreement, and are considering other partners both insideand outside of the US (Washington Post, 2007). Elsewherein the Pacific, the underlying conditions and projectedoutcomes of the policy can be designed to be highlyfavorable to DCs.

2. Background

Experience with pollution emission trading has generallybeen successful. The most prominent case is sulfurallowance trading in the US (see, e.g., Ellerman et al.,2000). More recently, emissions trading within the Eur-opean Union has generally been successful as well. Somerecent criticisms are valid (Ellerman and Buchner, 2007),but most of the problems can be remedied with designmodifications. On the global scene, several studies haveindicated that mitigation costs can be reduced by as muchas 50–75 percent below a system of fixed mitigation quotas(see, e.g., Rose and Stevens, 2001; Weyant, 1999).Emissions trading has recently been analyzed for the

Pacific Rim and individual countries, yielding both positiveand negative conclusions. Oxley and Macmillan (2004)examined implications of the Kyoto Protocol for Asia-Pacific Economic Cooperation (APEC) developing coun-tries, as well as the related question of future complianceperiods and DC involvement. The authors are quitenegative about the goals and prospects for this agreement.For example, they state that ‘‘Even if the system wereestablished, developing countries could not participateuntil they generate tradeable permits. This requirescommitments to regulate and reduce emissions which theywill not make.’’ (p. 6). However, this completely overlooksthe policy instrument design options proposed in thispaper. It also suggests that the CDM will not yield muchbecause of the cumbersome approval process.4 The authorssuggest that a superior policy approach would be to

CDM is effectively bi-lateral trading of emission commitments, while

the tradeable permit approach is a multi-lateral one. The former focuses

on individual projects and can become bogged down in details. The latter

operates through a financial intermediary in a much more streamlined and

efficient manner in terms of the various transactions costs (defined broadly

to also include monitoring and enforcement) (see, e.g., Collamer and

Rose, 1997). Note that CDM projects are undertaken through the United

Nations Global Energy Facility, which provides credits for ICs that they

may use to offset their commitments during the forthcoming Kyoto

compliance period.

ARTICLE IN PRESS

5Most studies indicate that banking and borrowing of permits does not

significantly lower the overall compliance cost (see, e.g., Stevens and Rose,

2002).6Zhang has pointed out some complications relating to the Russian hot

air issue. Loeschel and Zhang (2002) conclude that Russia was allocated

‘‘hot air’’ permits at Kyoto in order to entice it to sign the agreement.

After the US withdrew from the Kyoto Protocol, Russia used its increased

leverage to obtain further credits as a carbon sink. Thus, Russia is likely to

resist a modification such as the one we propose, but at the same time most

analysts anticipate that economic growth will make hot air a non-issue for

Russia by the year 2020.

A. Rose, D. Wei / Energy Policy 36 (2008) 1420–14291422

suspend mitigation commitments until lower-cost abate-ment technologies can be developed. They cite some studiesthat represent some of the most pessimistic analyses of theissue as indicating that economic impacts of the Kyoto willbe especially negative for nearly all APEC countries,though they note it will result in positive investment flowsto some East Asian members. They suggest the twocountries most adversely affected will be China andMexico, because they are significant energy exporters.

Oxley and Macmillan are also very critical of institu-tional design aspects of international permit trading.Concern about ‘‘hot air’’ trading, however, can easily beaddressed by forbidding such a practice. One of theirlegitimate concerns is the possibility that some permitallocations may be politically untenable if they call forenormous transfers between countries. However, theproposal presented here would not raise these concerns.Finally, they cite the lack of an institutional structure forthe trading system, but fail to acknowledge the effective-ness of financial intermediaries in general and theirsuccesses in permit trading in particular (see, e.g., Collamerand Rose, 1997).

Zhang (2003) has analyzed several concerns about Chinaentering an international permit trading arrangement.First, DCs like China consider it unfair to take anyabatement commitment before Annex I countries take thelead in doing so. However, our study performs simulationsfor year 2020, which is after the first (and possibly secondand third) Kyoto compliance period for ICs. Their secondconcern pertains to the lack of institutional support andcapability for operating emissions trading in DCs. How-ever, China is currently actively engaging in bilateralcooperation with ICs on CDM projects; it is reported thatby the end of January 2007, the Chinese government hadapproved nearly 300 CDM projects (China Daily, 2007).Though the institutional requirements of emission tradingmay have different features from CDM, experiencesaccumulated in CDM projects will enhance China’scapability to participate in multilateral emission tradingin the future. The jury is still out, however, on whether theinstitutional requirements and associated costs of interna-tional emissions trading will be lower than that of CDM.

Zhang also notes that an emissions trading program mayworsen the trade deficit the US has with China and therebyspur retaliation. However, our simulations indicate annualtransfers between the five western US states and Chinawould be only a small percentage of this imbalance, and anearly infinitesimal percentage of overall trading volumebetween these two countries. Still, we acknowledge thattrade is an increasingly sensitive topic, and that CDMinvestment as an alternative is much less so. Finally, unlikeAnnex I countries, for which emissions targets are setagainst their historical emission levels, Zhang indicates thatfor DCs any abatement targets must be linked to theirfuture emissions to be considered fair. This argument isconsistent with our strategy to set up emission caps forDCs based on their 2020 projected baseline emission levels,

though, we admit that there would be uncertainties withrespect to DCs’ economic growth rates and thus emissionprojections.

3. Policy design

This paper assumes the workings of a typical interna-tional emissions trading system for carbon dioxide. A fixedamount of permits is allocated to each country in a givenyear. Countries must either mitigate any CO2 emissions inexcess of their allocated amount or they will have to buypermits for each tonne emitted over the initial quota. If acountry sells permits, it must provide a unit of CO2

mitigation for each unit it sells. The permit price isdetermined by the interactions of buyers and sellers inthe market place. No banking or borrowing of permits overtime is allowed in the simple version of the model.5

The trading entities are the 31 countries, regions, or stateslisted in column 1 of Appendix Table A1. In our majorsimulation results below, trading entities are aggregated into11 groups. The group aggregation strategy is shown in thefirst column of Appendix Table A1. Permits are allocated onthe following bases for the reference simulation:

1.

Three ICs (Japan, Canada, and New Zealand) areallocated permits according to their Kyoto commit-ments.

2.

Australia is allocated permits according to its Kyotocommitment, even though it has not ratified theProtocol.

3.

Russia is allowed permits according to a strict ‘‘NoHarm Rule’’ or equal to 90 percent of its baselineemissions. This avoids the ‘‘hot air’’ selling problem thatwould arise if it were given permits on the basis of itsKyoto cap, which exceeds its current emissions becauseof the economic downturn associated with its economictransition.6

4.

The five Western States of the US are granted permitsequal to their recent commitments in the Five-Stateagreement.

5.

Developing countries are allocated permits on the basisof each of the following:a. Strict ‘‘No Harm Rule’’: This is an allocation of

permits such that, following trading, each developingcountry would incur zero cost, i.e., permit revenueswould exactly offset permit costs.

ARTICLE IN PRESS

7

gre

we

imp8

cur

refl

red

A. Rose, D. Wei / Energy Policy 36 (2008) 1420–1429 1423

b. Ninety Percent of Baseline Emissions: This representsa large amount of permits, which has the potential toresult in negative net costs (permit revenues exceedingtotal mitigation cost) for DCs.7

We

at d

con

lica

No

ren

ect

uct

The ‘‘No Harm Rule’’ was proposed by Edmonds et al.(1995) and stated as: ‘‘developing nations receive sufficientemissions rights to cover own emissions and to generatesufficient revenue to cover the economic cost of participa-tion in the protocol.’’ There is some ambiguity over itsoriginal formulation and subsequent use. Both of theversions of the rule that we simulate are consistent with thisdefinition and other examples found in the literature.

Each of the proposed allocation rules for DCs isconsistent with several altruistic principles, such as ‘‘Abilityto Pay,’’ and ‘‘Vertical Equity’’ (see Rose, 1992; Rose et al.,1998). The first is an ‘‘allocation-based’’ principle, whereeach entity’s permit amounts are determined at the initialstage of the process. The second is an ‘‘outcome-based’’rule that is based on the results of the post-trading stage.The first involves less uncertainty than the second becauseit is not based on an unknown second step. Theuncertainty, however, can be reduced by central authorityensuring a zero net cost to outcome through financialtransfers out of a common pool, whose funds might beprovided by ICs and/or any excess revenues (over zero netcost) from DCs. Note that both rules do involve an implicitmitigation commitment on the part of DCs, so that theyboth meet the requirement of serious DC involvement.Furthermore, potential criticism that permit tradingsystems result in costless or even financially beneficialoutcomes for DCs should be juxtaposed against thepotential savings of ICs, which we will show to be large.8

4. Policy simulations

We performed four simulations differentiated accordingto our two versions of the No Harm Rule and whether ornot the five Western States are included:

1.

Case 1. Strict No Harm Rule for DCs, and inclusion ofthe five Western States of the US.

9Note that the senior author has also employed a simultaneous equation

2. permit trading model developed by Zhang (see Zhang, 2000; Rose and

Case 2. Strict No Harm Rule for DCs, but excluding thefive Western States.

Zhang, 2004), but that model is not readily able to include outcome-based

3. permit distributions and has not yet been extended in the various ways

Case 3. Ninety Percent of Baseline Allocation for DCs,and inclusion of the five Western States.

noted above.10The interpolation was performed as follows: The marginal abatement

4.

cost (MAC) curves of the five US Western States have a semi-logarithmic

Case 4. Ninety Percent of Baseline Allocation for DCs,but excluding the five Western States.

acknowledge that in international negotiations there is often a

eal of disagreement over the establishment of baselines. However,

sider our analysis and results to be of a general nature and their

tions are not affected by this consideration.

te that we have used emission cap levels most closely approximating

t commitments. A more general analysis using game theory to

more strategic behavior would likely lead to a different set of GHG

ion levels.

The assumptions we adopted in our simulation modelare summarized as follows:

�

fun

em

par

par

ene

fos

val

a ¼par

all sectors are covered;

� only carbon dioxide emissions are considered; � emissions are based on consumption rather than

production processes that generate them;

� marginal cost curves embody direct mitigation costs

only;

� marginal cost curves do not include various transactions

costs;

� marginal cost curves do not distinguish between

producer vs. consumer allocation of permits;

� carbon sequestration is included as an option, though

the term ‘‘mitigation’’ is applied to it for the sake ofgenerality;

� the ‘‘no regret’’ range of mitigation is between 5 and 10

percent and is determined for each country/region asexplained below;

� emission caps are based on stated country commitments

for ICs and ‘‘No Harm’’ rules for DCs as explainedabove;

� the simulation target year is 2020.

The simulations are run using a non-linear programmingmodel developed by Rose and Stevens (see Rose et al.,1998). The model has two variants: one is ‘‘allocation-based’’ and the other is ‘‘outcome-based.’’ Note that themodel is sufficiently general to be able to simulate theworkings of several equity rules (see Rose et al., 1998), toinclude benefits of mitigation (in addition to the cost) in thecalculations (Rose and Stevens, 1998), to include institu-tional constraints such as supplementarity (Rose andStevens, 2001), and to include dynamic considerations(Stevens and Rose, 2002).9

The necessary data for the model include CO2 emissionprojections and marginal mitigation cost curves. The costcurves are presented in Fig. 1. They were obtained from avariety of sources, including Ellerman and Decaux (1998)and IPCC (2007) for most DCs, state energy and climatepolicy plans for Arizona and New Mexico (AZCCAG,2006; NMCCAG, 2006), and interpolation for other USstates.10

ctional form (MCA ¼ aþ b lnð1� RÞ, in which R is the percentage

ission reduction, a is the intercept parameter, and b is the slope

ameter). The MAC curves of CA, OR, and WA are specified by

ametric shifts from the Arizona state curve in direct proportion to their

rgy intensity weighted by the relative carbon content of the three major

sil fuels. The shift is accomplished by altering the a and b parameter

ues in the marginal cost function. The Arizona parameter values are

�72.8 and b ¼ �665.3. The energy intensity and the a and bameter values have a positive relationship. Thus, a state with an

ARTICLE IN PRESS

0

20

40

60

80

100

120

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%

Carbon Emissions Reduction

Marg

inal A

bate

ment C

ost

(2006 d

olla

rs p

er

tonne o

f C

O2)

Japan

Australia/NZ/Canada

Mexico

SoutheastAsia

China

AZCA

Fig. 1. Marginal abatement cost curves of major countries/regions/states.

A. Rose, D. Wei / Energy Policy 36 (2008) 1420–14291424

Note one important aspect of the marginal cost curvesrelating to the potential of conservation and other ‘‘NoRegrets’’ options (i.e., mitigation practices that pay ormore than pay for themselves). ‘‘Bottom-up’’ analyses aretypically based on engineering considerations, while ‘‘top-down’’ analyses are typically based on macro economicmodels. The former tend to be relatively optimistic, butoverlook institutional impediments and engineering reali-ties of real world applications. The latter tend to berelatively pessimistic because they take a superficial orcoarse-grained view of the technological options. Themitigation cost curves for the Western States used in thispaper are based on a more sophisticated analyses, however(see AZCCAG, 2006; NMCCAG, 2006). The estimates arebased on a consensus stakeholder process and represent acombination of engineering approaches, macro assump-tions, experience with demonstration projects, and adapta-tion of actual data from other jurisdictions. Thus, theycannot be readily characterized as either upper or lowerbound.11

A general consensus in the literature indicates a No-Regrets range of at least 10 percent for most DCs and lowerlevels for ICs (Halsnaes et al., 1994; Mosnaim, 2001; Wanget al., 2007). We have set the No Regrets level at 5 percentfor Japan and 7.5 percent for Asian Tigers, Australia, NewZealand, and Canada because they have already imple-

(footnote continued)

emission weighted fossil energy intensity half as large as Arizona would

have parameter values half as large. The relative weighted fossil fuel

intensity value of a state is proportional to the carbon intensity value

(ratio of CO2 emission and gross state product) of this state. For example,

in 2020 Arizona has a CO2/GSP ratio of 0.465, while California has a ratio

of 0.290. Thus, the a and b parameter values of Arizona are 1.603 times

those of the California’s.11Note also that the cost curves developed for the Western States are

based on mitigation actions using several methods of implementation

including cap and trade. Effects of additional transaction costs associated

with any differential from the implementation of a cap and trade system

should be added to the cost curves. However, in order to be consistent

with the cost curves of other Pacific Rim countries or regions derived from

other sources, for which marginal cost curves do not include various

transaction costs, we did not incorporate them into the cost curves of

Western States.

mented a great deal of the low-hanging fruit. We utilize a 10percent No-Regrets level for the five Western States(derived from the state energy and climate policy reportsof Arizona and New Mexico), however, because of theaggressive pursuit of these kinds of options by thestakeholders and policy makers in some areas in the US.Finally, note that many studies suggest that cost saving

may be as much as $100 per tonne of CO2 for someoptions, while others argue that cost-saving measures arelimited because, if they existed, the market would haveensured their implementation. Of course, this perspectiveomits several considerations, such as the potential costlessremoval of institutional barriers or the role of informeddecision making. Still, we strike a compromise of the twoviews by assuming that No Regrets options are cost neutralrather than cost saving, i.e., the mitigation cost curves inFig. 1 move along the X axis at zero rather than beginningin the negative cost range.Note also that we perform simulations only for the year

2020. We considered several alternative time periods. Theyear 2010 was considered too near for the implementationof policy proposed here. We also considered 2012–2016, asa possible Kyoto second compliance period, but consideredthis too short sighted. Instead, we take a long-term view forthis policy design, and thus have utilized the year 2020 as arepresentative of its potential.

5. Results

The results of our simulation are presented in Tables 1–4and Figs. 2–4. We begin with some highlights of the overallresults before going into detail:

Pure No Harm:

�

The equilibrium permit price is $5.55/tCO2. � 3.6 billion tCO2 are mitigated. � All DCs are permit sellers (China receives $3 billion in

permit revenues).

� All ICs are permit buyers. � All countries are either better off or no worse off as a

result of trading.

� Total cost savings to the Pacific Rim countries is $38.5

billion, or 90 percent of pre-trading costs.

Ninety Percent of Baseline Permit Allocation:

�

The equilibrium permit price is $1.88/tCO2. � 2.5 billion tCO2 are mitigated. � All DCs are permit sellers (China reaps $1.1 billion in

permit revenues).

� All ICs are permit buyers. � All countries are better off as a result of trading,

although the Asian Tigers incur a positive net cost.

� Total cost savings to the Pacific Rim countries is $41.1

billion, or 98 percent of pre-trading costs.

ARTICLE IN PRESS

Table 1

Emission trading simulation among five Western States and Pacific Rim countries or regions in year 2020 (‘‘No Harm Rule’’ for non-Annex I countries

and Russia) (million $2006 or otherwise specified)

Region/country Before trading After trading

Emission

reduction

(% from baseline)

Mitigation

cost

Trading

cost

Net cost Cost saving Emission

reduction

(million tCO2)

Emission

reduction

(% from baseline)

Asian Tigers 11.8 293 �293 0 78 221 15.5

Australia/NZ 39.4 83 662 745 3317 84 16.2

Canada 43.7 129 1219 1348 7883 130 16.2

C. America 16.1 14 �14 0 4 14 19.6

China 19.0 3016 �3016 0 848 2094 25.7

Japan 21.7 50 1026 1076 7230 79 6.5

Mexico 16.1 115 �115 0 34 117 19.6

Russia 16.3 374 �374 0 111 412 19.4

S. America 16.1 51 �51 0 15 52 19.6

Southeast Asia 16.7 295 �295 0 85 253 21.2

Western States 33.8 30 1250 1280 18,904 112 11.2

Total 20.6 4450 0 4450 38,510 3567 20.6

Permit price ¼ $5.55/tonne CO2.

Total permits bought/sold ¼ 748.83 million tonnes CO2.

Table 2

Emission trading simulation among Pacific Rim countries or regions except for the US in year 2020 (‘‘No Harm Rule’’ for non-annex I countries and

Russia) (million $2006 or otherwise specified)

Region/Country Before trading After trading

Emission

reduction

(% from baseline)

Mitigation

cost

Trading

cost

Net cost Cost saving Emission

reduction

(million tCO2)

Emission

reduction

(% from baseline)

Asian Tigers 10.4 125 �125 0 33 184 12.9

Australia/NZ 39.4 41 444 484 3578 74 14.4

Canada 43.7 63 807 870 8361 115 14.4

C. America 14.7 7 �7 0 2 13 17.4

China 16.3 1376 �1376 0 381 1727 21.2

Japan 21.7 20 660 680 7627 73 6.0

Mexico 14.7 56 �56 0 16 103 17.4

Russia 14.9 183 �183 0 54 369 17.4

S. America 14.7 25 �25 0 7 46 17.4

Southeast Asia 14.9 139 �139 0 40 218 18.3

Total 17.9 2034 0 2034 20,099 2921 17.9

Permit price ¼ $3.45/tonne CO2.

Total permits bought/sold ¼ 554.71 million tonnes CO2.

A. Rose, D. Wei / Energy Policy 36 (2008) 1420–1429 1425

Omitting Western States:

�

This results in relatively lower permit prices andemission reductions � Cost savings are increased for ICs because the absence

of the US dampens the bidding up of permit prices

� Gains are decreased for DCs because permit prices are

relatively lower.

Overall results:

�

ICs reap enormous cost reductions (under the secondversion of No Harm Rule, US Western States cost

reductions equal $20 billion; Canada’s equal $9 billion;Japan’s equal $8 billion).

� Negative net costs for DCs are small in comparison to

IC gains.

� DCs are better off having the Western States included,

because the latter bid up the permit price, whichincreases DC permit sales revenues.

More details of the results are now provided startingwith Table 1, which pertains to the strict ‘‘No Harm’’ rule(zero net cost for all DC Pacific Rim entities). The firstcolumn indicates the emission reduction before trading interms of a percentage reduction from baseline (this is the

ARTICLE IN PRESS

Table 3

Emission trading simulation among five Western States and Pacific Rim countries or regions in year 2020 (10% reduction from baseline for non-Annex I

countries and Russia) (million $2006 or otherwise specified)

Region/country Before trading After trading

Emission

reduction

(% from baseline)

Mitigation

cost

Trading

cost

Net cost Cost saving Emission

reduction

(million tCO2)

Emission

reduction

(% from baseline)

Asian Tigers 10.0 41 �18 23 2 152 10.7

Australia/NZ 39.4 16 260 276 3786 65 12.6

Canada 43.7 25 468 494 8738 100 12.6

C. America 10.0 3 �7 �4 4 11 15.2

China 10.0 494 �1085 �591 591 1392 17.1

Japan 21.7 6 370 376 7930 67 5.5

Mexico 10.0 22 �59 �36 36 91 15.2

Russia 10.0 74 �217 �143 143 327 15.4

S. America 10.0 10 �26 �16 16 41 15.2

Southeast Asia 10.0 53 �123 �71 71 185 15.5

Western States 33.8 3 437 441 19,744 105 10.5

Total 14.6 747 0 747 41,062 2536 14.6

Permit price ¼ $1.88/tonne CO2.

Total permits bought/sold ¼ 815.72 million tonnes CO2.

Table 4

Emission trading simulation among Pacific Rim countries or regions except for the us in year 2020 (10% reduction from baseline for non-annex I countries

and Russia) (million $2006 or otherwise specified)

Region/country Before Trading After Trading

Emission

reduction

(% from baseline)

Mitigation

cost

Trading

cost

Net cost Cost saving Emission

reduction

(million tCO2)

Emission

reduction

(% from baseline)

Asian Tigers 10.0 19 5 25 0 138 9.7

Australia/NZ 39.4 9 181 190 3872 60 11.7

Canada 43.7 14 325 339 8893 93 11.7

C. America 10.0 1 �4 �2 2 10 14.2

China 10.0 247 �531 �285 285 1235 15.1

Japan 21.7 3 252 255 8051 65 5.4

Mexico 10.0 12 �31 �19 19 84 14.2

Russia 10.0 41 �119 �79 79 306 14.4

S. America 10.0 5 �14 �9 9 38 14.2

Southeast Asia 10.0 27 �63 �35 35 169 14.1

Total 13.4 380 0 380 21,246 2198 13.4

Permit Price ¼ $1.27/tonne CO2.

Total permits bought/sold ¼ 601.49 million tonnes CO2.

12Note that the column sum for trading cost is zero because the value of

sales is exactly offset by the value of purchases.

A. Rose, D. Wei / Energy Policy 36 (2008) 1420–14291426

complement of the amount of emissions for which permitsare granted). For example, 11.8 percent for the AsianTigers means that this group of countries receives permitsequal to 88.2 percent (100 minus 11.8 percent) of itsbaseline emissions. Higher percentages for ICs indicatetheir relatively greater mitigation commitments. Still therange of 11.8 percent to 19.0 percent for DCs is asignificant commitment and potentially reduces CO2 by2.4 billion tonnes beyond business as usual (i.e., no PacificRim DC involvement above and beyond the KyotoProtocol and Western States agreement).

Mitigation costs after trading are listed in column 2 ofTable 1, and trading costs (purchases are denoted by

positive entries and sales by negative entries) are listed incolumn 3. Net costs are simply a summation of columns 2and 3. Note that, true to the specification, all cost entriesare zero for DCs, and, of course, positive for ICs.Cost savings indicate the difference between pre-trading

and post-trading costs plus permit transactions. Note thatdespite the large net costs, ICs receive by far the larger costsavings in percentage terms.12 Note that 748.83 milliontonnes of CO2 are transacted and that the total emissionreduction is 3567 million tonnes, including additional

ARTICLE IN PRESS

-4,000

-2,000

0

2,000

4,000

6,000

8,000

Asian

Tiger

s

Austra

lia/N

Z

Can

ada

C. A

mer

ica

China

Japa

n

Mex

ico

Rus

sia

S. Am

erica

South

east A

sia

Wes

tern

Sta

tes

in m

illio

n 2

006 d

olla

rs

Trading CostNet Cost

Cost Saving

20,000

Fig. 2. Trading cost, net cost, and cost saving of countries/regions for

simulation Case 1.

-2,000

0

2,000

4,000

6,000

8,000

Asian

Tiger

s

Austra

lia/N

Z

Can

ada

C. A

mer

ica

China

Japa

n

Mex

ico

Rus

sia

S. Am

erica

South

east A

sia

Wes

tern

Sta

tes

in m

illio

n 2

00

6 d

olla

rs

Trading Cost

Net Cost

Cost Saving

20,000

Fig. 3. Trading cost, net cost, and cost saving of countries/regions for

simulation Case 3.

-600

-300

0

300

600

900

1,200

1,500

Ne

t C

ost

(in

mill

ion

$2

00

6) Japan

China

Canada

Australia/NZ

Mexico

Southeast Asia

No HarmAll Pacific Rim

No Harmw/o WS

10% MitigationAll Pacific Rim

10% Mitigationw/o WS

Fig. 4. Net cost impacts of alternative permit trading arrangements.

A. Rose, D. Wei / Energy Policy 36 (2008) 1420–1429 1427

reductions of 2414 million tonnes due to DC involvement,or a 2.1-fold increase in mitigation. The post-tradingmitigation levels are presented in the last column and showa narrowing of the range from column 1. China undertakesthe largest mitigation after trading—25.7 percent (up from

its original 19 percent)—and Japan lowers its mitigationafter trading from 21.7 percent to 6.5 percent. Of course allcountries are better off from trading, and the actions ofboth China and Japan are in their self-interest in additionto providing for the common good.The results presented in Table 2 pertain to the case of the

pure ‘‘No Harm Rule’’ but with US states omitted. Firstnote that the permit price drops more than $2 per tonne to$3.45, because the states are no longer bidding it up. Otherfigures in the table change significantly as well. The pre-trading emission reduction for DCs goes down becausetheir obligations for permit trading are lower in this caseand they are less able to offset relatively higher mitigationcosts because of lower permit prices than in Table 1. Totalimpacts for DCs are still the same, however, because of thezero net cost requirement. Non-US ICs are relatively betteroff than in Case 1 because the permits they purchase arerelatively cheaper. Net cost for this group (not includingthe US) drops from $3170 million in Table 1 to $2034million in Table 2, or a decrease of 35.8 percent. The totalemission reduction is lower because not only is the USexcluded but also the institutional commitments by DCsare lower (compare the last two columns of the two tables).Tables 3 and 4 present the results of the second version of

the No Harm Rule, in which DCs are allowed to gainoutright from inclusion in permit trading, since they aregranted a relatively larger level of permits (90 percent ofbaseline) than in the pure NoHarm case. All countries (exceptAsian Tigers) are better off in Case 3 (Table 3) than itscounterpart in Case 1, because more permits are sold and at alower price. At the same time, the amount of CO2 emissionreduction is not as great as in Case 1. Note, however, that theAsian Tigers’ net cost is a positive $23 million. This outcomecan be adjusted through a small transfer of dollars or byincreasing the initial permit allocation.Table 4 presents the results of the second version of the

No Harm Rule, but where the US Western States areomitted. Similar to Table 2, the permit price drops from$1.88 to $1.27 without the US states to bid up the permitprice. Compared to Case 2 in Table 2, all countries arebetter off (except the Asian Tigers), since more permits aresold and at a lower price. Compared to Case 3, all DCs areworse off, since fewer permits are transacted by ICswithout the US, and DCs have to sell the permits at a lowerprice. The net gain of this group drops from $840 million inCase 3 to $406 million in Case 4, or a decrease of 51.7percent. Contrarily, non-US ICs are better off than in Case3 because they can buy permits at a lower price. The netcost for this group (not including the US) drops from$1146 million in Table 3 to $783 million in Table 4, or adecrease of 31.7 percent. The total emission reduction islower than in Case 3 due to the exclusion of the US andlower commitments of DCs. Actually, the total emissionreduction here is the lowest among the 4 cases.Figs. 2 and 3 illustrate the results for Cases 1 and 3,

respectively. They dramatize the huge cost savings for ICs.Fig. 4 shows a comparison of the net cost results for each

ARTICLE IN PRESS

Table A1

Fossil energy use and carbon dioxide emissions in 2020

Fossil energy use

(quadrillion Btu)

CO2 emissions

(million metric tonnes)

Asian Tigers 23.6 1422

Hong Kong 1.3 90

North Korea 1.5 109

Singapore 2.3 148

South Korea 13.3 723

Taiwan 5.2 352

Australia/NZ 7.4 515

Australia 6.3 469

New Zealand 1.1 46

Canada 16.6 799

Central American

Countries

1.5 73

Costa Rica 0.3 10

El Salvador 0.2 10

Guatemala 0.3 17

Honduras 0.2 10

Nicaragua 0.1 7

Panama 0.4 20

China 91.8 8159

Japan 23.6 1218

Mexico 10.3 595

Russia 39.6 2117

South American

Countries

5.6 266

Chile 2.0 98

Colombia 2.0 88

Ecuador 0.6 36

Peru 1.0 43

Southeast Asia 19.9 1192

Indonesia 7.8 486

Malaysia 4.2 243

Philippines 2.2 118

Thailand 5.7 345

Western US States 16.6 999

Arizona 2.0 154

California 9.9 600

New Mexico 1.0 70

Oregon 1.3 89

Washington 2.4 86

Source: EIA (2007) for large countries. Regional growth rates estimated by

EIA were applied to individual countries in Central America, parts of

South America, and parts of Southeast Asia.

A. Rose, D. Wei / Energy Policy 36 (2008) 1420–14291428

country/region in relation to our four policy designs. Forthe non-US ICs, net costs decrease without the entry of theUS states or where the policy design follows the secondversion of No Harm Rule to DCs. Therefore, in Case 4, thenon-US ICs achieve their mitigation commitments at thelowest net costs. DCs can enjoy net gains from tradingunder the second version of No Harm Rule; however, theirgains decrease if the US states do not participate in thetrading.

Again, we note that we have omitted many considera-tions relating to policy design and international negotia-tions. As such, our results might best be interpreted asupper bounds on the emission reductions that might beexpected from DCs in the Pacific Rim. Moreover, thismaximum is more likely to be closer to the ‘‘90 percent ofBaseline Emission’’ variant than the pure ‘‘No Harm’’Rule, because the former provides a more attractiveoutcome.

Note also that the mitigation cost data on which theanalysis is based has some limitations. Some cost curves aredated, others are interpolated from regional averages, anda few are extrapolated. The intent of this paper is toprovide general insights rather than specific dollar values.The authors believe the results are robust to likely updatingand refinements of the mitigation cost data, however.

6. Conclusion

This paper contributes to the recent literature onregional cooperative approaches to climate change policyby suggesting parameters for a system of tradable GHGemission permits for the Pacific Rim in a way that makesthis policy approach attractive to developing countries.The ‘‘No Harm’’ rules for the allocation of permits implyno costs or no risks to DCs, while at the same time helpingto lower GHG emissions and save ICs substantialmitigation costs. It should be noted that while the emphasisof the paper has been on attracting developing countries,the ‘‘No Harm’’ rules result in outcomes that are likely toprovide relatively greater gains (both in terms of levels andpercentages) to some ICs than for many DCs. The equityof the relative distributional impacts aside, it should bekept in mind that all countries gain from the policiesproposed in this paper.

The paper, however, has omitted several importantissues of designing and implementing this policy instru-ment, such as financial intermediation, monitoring, andenforcement. However, it clearly demonstrates the greatpotential for cooperation across Pacific Rim countries inaddressing the important problem of climate change.

Acknowledgments

We wish to thank ZhongXiang Zhang, Tom Peterson,and Lisa Schweitzer for their helpful comments on anearlier draft. The authors are, of course, responsible forany remaining errors or omissions.

Appendix A

See details in Table A1.

References

Arizona Climate Change Advisory Group (AZCCAG), 2006. Arizona

Climate Change Action Plan /http://www.azclimatechange.gov/

download/O40F9347.pdfS.

Barrett, S., Grubb, M., Roland, K., Rose, A., Sandor, R., Tietenberg, T.,

1992. Combating Global Warming: A Global System of Tradable

Carbon Emission Entitlements. UNCTAD, Geneva, Switzerland.

China Daily, 2007. Climate Fund Seeks to Create a Green China /http://

www.chinadaily.com.cn/bizchina/2007-02/12/content_807247.htmS.

ARTICLE IN PRESSA. Rose, D. Wei / Energy Policy 36 (2008) 1420–1429 1429

Collamer, N., Rose, A., 1997. The changing role of transactions costs in

the evolution of joint implementation. International Environmental

Affairs 9 (4), 274–288.

Edmonds, J., Wise, M., Barns, D., 1995. Carbon coalitions: the cost and

effectiveness of energy agreements to alter trajectories of atmospheric

carbon dioxide emissions. Energy Policy 23, 309–335.

EIA, 2006. International Emissions, Energy Consumption, and Population

Tables /http://www.eia.doe.gov/emeu/international/contents.htmlS.EIA, 2007. International Energy Outlook 2007 /http://www.eia.doe.gov/

oiaf/ieo/index.htmlS.

Ellerman, A.D., Decaux, A., 1998. Analysis of Post-Kyoto CO2 Emissions

Trading Using Marginal Abatement Curves. MIT Joint Program on the

Science and Policy of Global Change, Report No. 40, Cambridge, MA.

Ellerman, A.D., Buchner, B., 2007. The European Union emissions

trading scheme: origins, allocation, and early results. Review of

Environmental Economics and Policy 1 (1), 66–87.

Ellerman, D.A., Joskow, P.L., Schmalensee, R., Montero, J., Bailey,

E.M., 2000. Markets for Clean Air: The US Acid Rain Program.

Cambridge University Press, Cambridge, UK.

Halsnaes, K., Mackenzie, G.A., Swisher, J.N., Villavicencio, A., 1994.

Comparable assessment of national GHG abatement costs. Energy

Policy 22 (11), 925–934.

IPCC, 2007. Climate Change 2007. The IPCC Fourth Assessment Report

/http://www.ipcc.ch/S.

Kousky, C., Schneider, S.H., 2003. Global climate policy: will Cities lead

the way? Climate Policy 3, 359–372.

Loeschel, A., Zhang, Z.X., 2002. The economic and environmental

implications of the US repudiation of the Kyoto Protocol and the

subsequent deals in Bonn and Marrakech. Weltwirtschaftliches Archiv

138 (4), 711–746.

Mosnaim, A., 2001. Estimating CO2 abatement and sequestration

potentials for Chile. Energy Policy 29 (8), 631–640.

New Mexico Climate Change Advisory Group (NMCCAG), 2006. New

Mexico Climate Change Action Plan /http://www.nmclimatechange.us/

ewebeditpro/items/O117F10150.pdfS.Oxley, A., Macmillan, S., 2004. The Kyoto Protocol and the APEC

Economies. Australia APEC Study Center, Melbourne, Australia.

Peterson, T.D., Rose, A., 2006. Reducing conflicts between climate policy

and energy policy in the US: the important role of the states. Energy

Policy 34 (5), 619–631.

Rose, A., 1992. Equity Considerations of Tradable Carbon Emission

Entitlements. In: Barrett, S., et al. (Eds.), Combating Global Warming:

Study on a Global System of Tradable Carbon Emission Entitlements.

U.N. Conference on Trade and Development, New York.

Rose, A., Stevens, B.K., 1998. Will a global warming agreement be fair to

developing countries. International Journal of Environment and

Pollution 9 (2/3), 157–178.

Rose, A., Stevens, B., 2001. An economic analysis of flexible permit

trading in the Kyoto Protocol. International Environmental Agree-

ments 1 (2), 219–242.

Rose, A., Zhang, Z.X., 2004. Interregional burden-sharing of greenhouse

gas mitigation in the United States. Mitigation and Adaptation

Strategies for Global Change 9 (3), 477–500.

Rose, A., Stevens, B., Edmonds, J., Wise, M., 1998. International equity

and differentiation in global warming policy: an application to

tradable emission permits. Environmental and Resource Economics

12, 25–51.

Stevens, B., Rose, A., 2002. A dynamic analysis of the marketable permits

approach to global warming policy: a comparison of spatial and

temporal flexibility. Journal of Environmental Economics and

Management 44 (1), 45–69.

Tietenberg, T., 2003. The tradable-permits approach to protecting the

commons: lessons for climate change. Oxford Review of Economic

Policy 19 (3), 400–419.

Wang, K., Wang, C., Lu, X.D., Chen, J.N., 2007. Scenario analysis on

CO2 emissions reduction potential in China’s iron and steel industry.

Energy Policy 35 (4), 2320–2335.

Washington Post, 2007. Western States Agree to Cut Greenhouse Gases

/http://www.washingtonpost.com/wp-dyn/content/article/2007/02/26/

AR2007022601321.htmlS.Weyant, J. (Ed.), 1999. The Costs of the Kyoto Protocol: a Multi-Model

Evaluation. Energy Journal 20 (Special Issue on the Cost of the Kyoto

Protocol).

Zhang, Z.X., 2000. Estimating the size of the potential market for the

Kyoto flexibility mechanisms. Weltwirtschaftliches Archiv—Review of

World Economics 136 (3), 491–521.

Zhang, Z.X., 2003. Why Has China not Embraced a Global Cap-and-

Trade Regime? Working paper available at SSRN (Social

Science Research Network): /http://papers.ssrn.com/sol3/papers.

cfm?abstract_id=1008592S.