ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

India and Climate Change: Some International Dimensions

Industrial countries have never been sympathetic to India's idea of controlling carbon emissions based on per capita targets. They prefer targets based on reductions in total emissions by developing countries, comparable or equivalent to those undertaken by them. This paper offers a new approach that tries to bolster the case for a per capita emissions approach by distinguishing the co2 emissions intensity of production and consumption from energy use per capita. It also outlines some projections that could lead to a reasonable emissions trajectory for India and one that is consistent with global efforts at addressing climate change. Looking at the role of trade in climate change, it concludes that the outcome will be messy if the trading system is burdened with the task of settling environmental problems.


India and Climate Change: Some International Dimensions

Arvind Subramanian, Nancy Birdsall, Aaditya Mattoo

Industrial countries have never been sympathetic to India’s idea of controlling carbon emissions based on per capita targets. They prefer targets based on reductions in total emissions by developing countries, comparable or equivalent to those undertaken by them. This paper offers a new approach that tries to bolster the case for a per capita emissions approach by distinguishing the CO


emissions intensity of production and consumption from energy use per capita. It also outlines some projections that could lead to a reasonable emissions trajectory for India and one that is consistent with global efforts at addressing climate change. Looking at the role of trade in climate change, it concludes that the outcome will be messy if the trading system is burdened with the task of settling environmental problems.

The sections in this paper on equitable burden sharing are based on work with Nancy Birdsall, while those on trade and cap-and-trade are based on work with Aaditya Mattoo. Kevin Ummel and Dan Hammer provided superb assistance with the data. We are grateful to Nitin Desai, Devesh Kapur, Pratap Bhanu Mehta, and conference participants for valuable discussions and comments.

Arvind Subramanian ( is at the Peterson Institute for International Economics, the Center for Global Development, and Johns Hopkins University, Nancy Birdsall is at the Center for Global Development and Aaditya Mattoo ( is at the World Bank, Washington.

Economic & Political Weekly

august 1, 2009 vol xliv no 31

his paper consists of three sections, covering three broad issues relating to climate change. First, arguably the biggest and most important challenge for India in the climate change negotiations will be ensuring equitable burden sharing on the future time path of global greenhouse gas (GHG) emissions. But the political and intellectual ground is shifting beneath India’s feet. There is likely to be less sympathy for the view long held by India that it should not be forced into commitments that will jeopardise its development trajectory. The Indian view that the focus should be on per capita emissions and that future emissions targets should take into account the historical “wrongs” of industrial countries finds increasingly fewer adherents, especially in the US. We outline very broadly and in a very preliminary fashion the contours of a different political and intellectual tack that India might take on the question of burden sharing.

The second point relates to trade. As we argued in a recent piece in Foreign Affairs, trade should be a secondary issue if the burden-sharing question can be resolved (Mattoo and Subramanian 2009). If not, then trade restrictive actions could proliferate and become a contentious issue. But the way World Trade O rganisation (WTO) law on the environment has been evolving – gradually but perceptibly – could, at the margin, give greater l egitimacy to these actions.

In the third section, we describe the two major policy approaches at the national level – cap-and-trade and carbon taxes

– for reducing GHG emissions and discuss the international implications of these approaches for a country like India.

1 Emerging Thinking: Is the Writing on the Wall?

The Boxer amendment to the Lieberman-Warner Climate Security Act of 2008 – the last piece of environmental legislation in the US Congress before the election of President Barack Obama – provides a reasonable indication of the way the US is likely to a pproach the international dimensions of climate change negotiations. Of course, the new Congress and new administration will craft entirely new legislation on climate change but one can reasonably expect that there will not be serious departures from the approach outlined in Lieberman-Warner-Boxer (LWB).1

The amendment’s stated purpose is “to promote a strong g lobal effort to significantly reduce greenhouse emissions”; and “to e nsure to the maximum extent practicable, that greenhouse gas emissions occurring outside the United States do not undermine the objectives of the United States in addressing global climate change”. Notably, for developing countries such as India, the amendment is very much animated by the spirit that it should prevent the “shifting of US jobs to foreign countries that would have lower manufacturing costs merely because they refuse to do their part to limit greenhouse gas emissions”. In turn, “doing their part”, also articulated in terms of “comparable action”, is likely to be interpreted in terms of “carbon tax comparability”. In other words, other countries will be judged by whether they impose carbon taxes similar or equivalent to the US or impose caps on emissions reductions similar to those that the US will embody in any new legislation.

The import provisions of this bill apply to “covered goods” from “covered countries”. Covered goods are energy-intensive goods such as steel and chemicals, and other goods that generate substantial GHG emissions in their production (cement, glass, pulp, paper, chemicals and industrial ceramics are explicitly cited). Covered imports are related closely to those US goods whose cost of production will be affected by any new legislation. Covered countries are those not on the excluded list, which comprises three categories: those who are taking comparable action to the US (a reduction of 70% in GHG emissions by 2050 and a r eduction of 20% by 2020) in limiting domestic GHG emissions relative to the base year of 2005; least-developed countries; and de minimis emitters of GHGs. Countries such as India will make it to the list only if the first criterion is met.

On timing, the LWB initiative envisages trade actions against countries that do not live up to the standards set in it, beginning in 2019. That is, it envisages a transition period of 10 years. But the current mood and thinking is that the transition period is likely to become shorter and countries such as India will face the possibility of trade action earlier rather than later.

Last year, there was a flurry of excitement when Prime Minister Manmohan Singh made a new proposal at Heiligendamm on burden sharing. Essentially, he said that India would always r emain below industrial countries in terms of per capita CO2 emissions. But LWB and other legislative initiatives all see burden sharing in terms of cuts in absolute emissions or in terms of c arbon taxes with no reference to per capita emissions.

The decreasing salience of the per capita approach stems probably from three factors. The gravity of the situation has elevated the need to take effective action, which is all about total emissions in the atmosphere, above concerns of fairness in burden sharing. Second, the historical wrongs argument made by developing countries runs into the following riposte: if culpability for past actions is to be taken into account, should not such culpability be based on intent? And given that the problem became e vident only in the last 20 or so years, the responsibility of industrial countries for their cumulative emissions should pertain to this shorter period than a hundred years or so. A final, less assertively made, argument pertains to population. If industrial countries were to accept the per capita emissions perspective, they would implicitly be making allowance for the large populations, and the perceived ineffective population policies of developing countries. This they are increasingly unwilling to do.

An Alternative Approach?

Is there another way of making a more intellectually persuasive case for a per capita approach? In recent (and unfinished) r esearch with Nancy Birdsall, we are trying to mount such a case.

We begin by noting that overall emissions stem from consumption and production and that these two need to be looked at separately. For a number of major GHG-emitting countries, we break up emissions into these categories (the data broadly permit this distinction). Actually, we disaggregate aggregate emissions into three categories along the following lines.

Total CO2 emissions = CO2 emissions from consumption activity + CO2 emissions in production activity. –(1) In turn:

CO2 emissions from consumption per capita = (Energy consumed per capita) * (CO2 emissions from consumption per unit of energy consumed) –(2)

CO2 emissions in production = (CO2 emissions in production per unit of gross domestic product; GDP) * (GDP) –(3)

The three variables of interest are

  • (i) energy consumed per capita or what we would call the pure consumption or welfare measure that should be of key interest to developing countries;
  • (ii) CO2 emissions from consumption per unit of energy consumed, which is a measure of inefficiency of CO2 emissions generation in consumption. We call this consumption inefficiency;
  • (iii) CO2 emissions in production per unit of GDP, which is a measure of inefficiency of CO2 emissions generation in production. We call this production inefficiency.

    The aim is to disentangle what we call efficiency of CO2 emissions generation (in production and consumption) from pure e nergy consumption.

    Table 1 presents some basic data for 2005 for some of the key variables.2 For all countries, emissions from production still a ccounts for the bulk of total emissions. However, the share of emissions in consumption activity is substantially greater for i ndustrial countries (about 42% for the US, France and the UK) compared with India (22%) and China (14%).

    Table 1: Selected Indicators Related to CO Emissions (2005)


    Carbon Carbon Total Total Share of Total House Intensity Intensity Carbon Carbon Consumption hold Energy of of Con-Emissions Emissions Emissions Use (Electricity Production sumption Per Capita in Total and Road Use) Emissions Per Capita Country/Units Tonnes Tonnes CO/Tonnes Tonnes Million Tonnes of Oil


    CO/$1,000 of Oil Equi-CO/Person Tonnes COEquivalent/


    GDP valent (TOE) Person

    Bangladesh 0.38 5.79 0.26 36.3 0.36

    India 1.38 9.26 1.05 1,147.5 0.22 18.31

    Indonesia 1.15 4.64 1.55 341.0 0.30 73.18

    Philippines 0.55 4.00 0.92 76.4 0.33 67.74

    Pakistan 0.87 4.30 0.76 118.4 0.32 41.47

    Malaysia 0.90 3.53 5.45 138.0 0.27 391.49

    South Africa 1.63 5.44 7.05 330.3 0.21 218.19

    Italy 0.26 3.54 7.75 454.0 0.35 460.22

    Netherlands 0.31 3.68 11.21 183.0 0.31 657.41

    Germany 0.25 3.65 9.86 813.5 0.40 690.62

    Denmark 0.19 3.89 8.77 47.5 0.33 559.32

    United Kingdom 0.19 3.80 8.80 529.9 0.43 646.49

    France 0.16 2.35 6.38 388.4 0.42 557.24

    Japan 0.17 3.84 9.50 1,214.2 0.28 557.24

    United States 0.30 3.87 19.62 5,817.0 0.43 1,881.33

    august 1, 2009 vol xliv no 31


    Figure 1: CO2 Emission Intensity of Production from a historical perspective) and also the striking profligacy in consumption of industrial countries. Figures 2 and 3 plot respectively the energy use per capita and consumption inefficiency of a few countries. On pure energy consumption per capita, industrial countries fare less well. Their e nergy consumption per capita – which we argue is the welfare metric that all countries should be focusing on – has been consistently rising, not falling, except for a few countries. From the Indian perspective, the point of such an analytical

    6 4 2Production Intensity (tonnes CO2 /$1,000 GDP) China



    2000 4000 6000 8000 GDP per capita, PPP

    .6 .5 .4 .3 .2Production Intensity (tonnes CO2 /$1,000 GDP) United States Germany United Kingdom Japan France

    15000 20000 25000 30000 35000 GDP per capita, PPP

    The data show that industrial countries have indeed made a lot of progress in production efficiency – which is measured as the CO2 emissions generated in production as a share of GDP (Figure 1).3 This figure plots this ratio for some major emitters. Currently, for example, India is about 41/2 times more inefficient than the US in production and even more so than some European countries (column 1 of table 1). China is almost twice as inefficient as India. But viewed historically, countries such as India, both in terms of the level and trajectory of CO2 emissions, do not appear to be doing badly.

    This inefficiency of India and China, however, is less pronounced on the consumption side (column 2). For every unit of consumption, India generates about two to three times more CO2 emissions than do the industrial countries.

    The greatest disparities, however, relate to basic energy consumption. Take the two categories of consumption for which there are good data – household electricity and transport. The US’ per capita consumption of energy is 49 times that of India and of transport 146 times that of India. Even the less profligate European countries consume substantially more than India. When both these categories are aggregated, we find that the US is 100 times more profligate than India (column 6).

    One advantage of disaggregating emissions as we have done is to isolate the policy or actionable elements. For example, in the popular debate, the contrast between the total emissions per capita of the US and Europe on the one hand and developing countries on the other (column 3) is pointed out. But this aggregate comparison is misleading. It obscures the high levels of inefficiency in developing countries today (although not necessarily

    Economic & Political Weekly

    august 1, 2009 vol xliv no 31

    exercise for the climate change negotiations would be to argue along the following lines. India should say that it is committed to contributing to climate change goals and that its chief contribution will be to strive to attain levels of production and consumption efficiency (of CO2 emissions) not based on historical levels but on those that can be found today in the industrial world. That is, in relation to production and consumption efficiency, India would not say that it will attain the level of efficiency that industrial countries attained at a comparable stage of their development experience (say 40 or 50 years ago). Instead, recognising the gravity of the situation, the need for a broad effort, and the fact that technology can and should be harnessed, India will strive to get close to the frontier in terms of the GHG-efficiency in production (this commitment may have to be conditioned to some

    Figure 2: Household Energy Use Per Capita

    .0015 .001 0005Weighted Average of Household Energy Use Per Capita (ktoe/person)China India Japan US Germany France UK

    10000 20000 30000 40000

    GDP per capita, PPP

    Figure 3: CO2 Emission Intensity of Consumption Weighted Average of Consumption Intensity (tonnes CO2 /ktoe)10 8 6 4 2 India China France UK Germany Japan US

    600 10600 20600 30600 40600 GDP per capita, PPP

    extent by domestic endowments. For example, India’s ability to meet frontier levels of production efficiency may be impeded by the abundant supply of domestic coal. Some allowance would have to be made for such country-specific factors).

    In return, India would seek two things. First, industrial countries would need to provide the financial and technological r esources to permit India to move to the efficiency frontier in p roduction. Second, and more importantly, industrial countries

    Table 2: Carbon Intensity in Production and Development

    Dependent Variable: Log of Production Carbon Intensity

    (1975-2000) Developed Countries Developing Countries

    Log of GDP per capita (PPP) -0.781 -1.373 -0.332 0.002 [t-stat] [-16.81] [-32.17] [-6.08] [-0.07]

    Country fixed-effects Y N Y N

    Sample size (n) 782 782 1,108 1,108

    R-squared 0.958 0.570 0.938 0.000

    Production carbon intensity is measured as tonnes of CO emissions in production per $1,000 GDP.


    Table 3: Household Energy Use and Development

    Dependent Variable: Developed Countries

    Developing Countries

    Household Energy Log of Total Log of Total

    Log of Total Log of Total Use Per Capita Household Household Household Household (1990-2005) Energy Use in Energy Use in Energy Use in Energy Use in

    Ktoe Ktoe Ktoe Ktoe

    Log of GDP per capita (PPP) 1.13 0.78 0.8 1.06 [t-stat] [10.90] [23.38] [18.30] [20.44]

    Country fixed-effects N Y N Y

    Sample size (n) 562 562 1,260 1,260

    R-squared 0.17 0.99 0.21 0.97

    Household energy use comprises electricity and transport use; ktoe is kilotonnes of oil equivalent.

    Table 4: Carbon Intensity of Consumption and Development

    Dependent Variable: CODeveloped Countries

    Developing Countries


    Emissions Per Unit of Log Carbon Log Carbon

    Log Carbon Log Carbon Household Energy Consumption Intensity of Intensity of Intensity of Intensity of (1990-2005) Consumption Consumption Consumption Consumption

    Log of GDP per capita (PPP) 0.05 -0.11 0.07 0.06 [t-stat] [1.65] [-3.91] [6.93] [2.25]

    Log of production carbon intensity 0.26 0.07 0.21 0.07 [t-stat] [14.58] [3.38] [21.32] [5.04]

    Country fixed-effects N Y N Y

    Sample size (n) 549 549 1,068 1,068

    R-squared 0.52 0.97 0.32 0.92

    Dependent variable is in tonnes of CO emissions in consumption per tonne of oil equivalent.


    would have to recognise the need for equity in energy consumption which would have to be measured in terms of consumption per capita. Here, India should not be expected to depart significantly from the pattern of per capita consumption exhibited by industrial countries over the course of their development (after all, basic needs – energy needs – do not change). This point could acquire greater salience if the contrast in the levels of energy consumption per se between India and industrial countries – which is staggering – is highlighted.

    We try and elaborate on these ideas. We run regressions for each of these three measures, relating them to development (proxied by per capita GDP; purchasing power parity-based). We run these regressions for the industrial countries and developing countries (separately) for the period for which data are available (1990-2005; Table 2 for production inefficiency; Table 3 for e nergy use per capita; and Table 4 for consumption inefficiency). We can then look at the elasticity of each of these three variables with respect to GDP per capita. We use these elasticities to project India’s future emissions in production and consumption under three scenarios (Table 5).4

    In the first – baseline – scenario, we assume that India’s future emissions behaviour will resemble that of developing countries during the period 1990-2005 (column 2, Table 5). By emissions behaviour, we mean that the elasticity of the different emissionrelated variables with respect to development (that we estimate in Tables 2-4) in the projection period will be similar to that for the recent past. In other words, this might be called the business-asusual scenario where India does not undertake any serious commitments to reduce CO2 emissions. This scenario also provides a reference for evaluating the effort in other scenarios. Table 5 shows that total emissions increase almost 10-fold and per capita emissions rise to about six million tonnes CO2 per c apita. Clearly, this scenario would not be conducive to reaching an i nternational agreement that would keep global emissions within safe limits.

    In the second scenario (column 3, Table 5), we assume that I ndia’s future emissions intensity behaviour will resemble that of industrial countries in the period 1990-2005 (see the assumptions at the bottom of Table 5).5 That is, the production and consumption intensity elasticities are assumed to be for India for the future what they were for industrial countries during the period 1990-2005. In effect, India’s commitment then is to approach t oday’s technological frontier in terms of emissions intensity. In this scenario, India’s energy use per capita, however, continues to grow according to that of developed countries (that is, the elasticity of energy use per capita with respect to development is about 1.06). In this scenario, India’s total emissions increase by about 134% and emissions per capita increase to about 1.6. Note that in this scenario, India’s “effort” as it were is the difference between current developing country elasticities (in production and consumption efficiency) and current industrial country elasticities. In terms of production efficiency, this elasticity is currently -.33 for developing countries and -1.7 for industrial countries, which would represent a fivefold improvement. Of course, in practice,

    Table 5: India's CO Emissions in 2050, Projections


    2005 Scenarios for 2050
    (1) (I) Current Policies (2) (2) Industrial Countries’ Performance in 1990-2005 (3) (3) Industrial Countries’ Likely Performance in Future(4)
    Emissions in production (MTCO2) 891 5,835 289 45
    Emissions in consumption (MTCO2) 256 4,295 2,396 1,008
    Total emissions (MTCO2) 1,147 10,130 2,686 1,053
    PopulationEmissions per capita 1,096.4 1.05 1,658 6.11 1,658 1.62 1,658 0.63
    Annual average per capita GDP
    growth rate (2005-2050) 5%
    Annual average population
    growth rate (2005-2050) 0.92%
    Elasticity of CO2 emissions intensitof GDP wrt development y -0.332 -1.7 -2.55
    Elasticity of energy use per capita wrt development 1.06 1.06 0.78 Elasticity of CO2 emissions intensity of consumption wrt development 0.06 -0.11 -0.165 august 1, 2009 vol xliv no 31 Economic & Political Weekly

    achieving this improvement would require better technology and resources, which could become the subject of the Copenhagen bargain in December this year.6 Note that even with this very “responsible” behaviour by India, total emissions would increase substantially. The call by the United Nations Development Programme (UNDP) for developing countries to cut emissions by 20% would not be met by India in this scenario.

    In a third scenario (column 4, Table 5), we assume that India’s CO2 emissions intensity in production and consumption will converge not to today’s technology frontier in the industrial countries but some future technology frontier. In addition, we assume that India’s energy use will decline in line with the pattern exhibited by industrial countries in the recent past. Only with these assumptions – which imply a huge effort by India – could there be any possibility of something like the UNDP targets being met.

    In sum, the most reasonable scenario for India (scenario two) would involve a bargain in which there is a concession by developing countries on the frontier of CO2 emissions on production and consumption, but agreement that on the basic welfare metric of energy consumption per capita, there should be comparability or some equivalence between the development experiences of the industrial countries and the future development trajectory of developing countries. Energy needs do not change but the efficiency of meeting them does and India could offer to make a s erious effort on the latter.

    2 The WTO, Trade and the Environment

    Aaditya Mattoo and I argued in a recent paper that the salience and use of trade policy and trade restrictive actions in relation to climate change will be determined largely by the success of the Copenhagen or some successor process that addresses the question of equitable burden sharing. If there is agreement on the terms of a fair and equitable bargain on the future responsibilities relating to GHG emissions, trade restrictive action and trade frictions more broadly can be minimised.

    On the other hand, if this is not possible, the possibility of trade frictions will increase considerably because trade will unfairly bear the burden of having to mediate on what are fundamentally development and environmental issues rather than trade issues.

    Trade measures for the environment are broadly taken for two reasons (see Subramanian 1992 for a more elaborate taxonomy of such measures). They can be taken for competitive reasons or to change the environmental behaviour of partner countries.

    In the context of climate change, competitiveness arises b ecause differential commitments to CO2 reduction will effectively translate into widely varying taxes on energy use and hence into varying costs (across countries) of production, especially in energyintensive sectors. This creates the risk of carbon leakage because trade and investment will flow towards countries with lower e nergy taxes and costs. In turn, this will create the demand to use trade policy to offset the different taxes on energy.

    One important question is whether and to what extent current WTO rules permit action to offset the competitiveness impact of differential energy or CO2-related taxes. WTO law and jurisprudence are inherently very complicated, so ex ante assessments

    Economic & Political Weekly

    august 1, 2009 vol xliv no 31

    must be made with considerable caution and caveats. At the end of the day, it is the dispute settlement procedure that settles legal matters. What is clear, however, is the WTO jurisprudence has gradually been evolving in the direction of allowing countries more policy space on the environment; that is, allowing greater permissiveness in the use of trade restrictive action on environmental grounds. This greater permissiveness relates to both types of trade measures – those taken on competitive grounds and those taken to influence partner country behaviour.

    What is the best guess about current WTO rules on competitiveness in the most important practical scenario relating to climate

    Table 6: Exports of Resource-Intensive Goods, India and China, 2007

    Steel Cement Paper Aluminium Chemicals
    Export to World
    China Value 47,070 1,151 6,788 7,282 792
    Share of total 3.87 0.09 0.56 0.60 0.07
    India Value 7,067 220 438 766 1,525
    Share of total 4.84 0.15 0.30 0.52 1.04
    Exports to OECD
    China Value 23,141 612 3,541 3,791 453
    Share of total 3.32 0.09 0.51 0.54 0.06
    India Value 3,409 0.39 145 135 278
    Share of total 5.48 0.00 0.23 0.22 0.45

    All values are in millions $, all shares are in %. Source: UN Comtrade.

    change? Suppose, in the future, India were to apply a 10% tax on carbon while the US levies a 50% tax. Suppose that India exports steel to the US and say that carbon-related costs comprise 30% of the value of steel output. Would the US be permitted to say that the cost of domestic production is greater than competing goods from India because of India’s low energy tax and that it is justified in levying an additional charge of say of 12% (50-10 times 30%) on steel from India to level the playing field?

    The WTO has rules on what are called border tax adjustments (BTAs) which govern situations like that outlined above. Hufbauer et al (2009) argue that in all probability the answer to the above question is in the affirmative. They base this judgment on a reading of the language in Article III: 2 (a) of the General Agreement on Tariffs and Trade (GATT) on the Agreement on Subsidies and Countervailing Measures, which has more e xplicit rules on BTAs relating to exports, as well as on the finding of a GATT panel back in the late 1980s on certain measures that the US took as part of the Superfund programme.7 I ndeed, the LWB bill seems very much to be predicated on the assumption that its provisions will not run afoul of WTO rules relating to BTAs.

    In some ways, therefore, it appears that on competitive grounds, developing countries might have to accept that WTO rules permit trade actions. Three points, however, need to be borne in mind. First, how important could such actions be for I ndia? That depends very much on India’s status as an exporter of resource-intensive goods. Table 6 suggests that at least currently India is not a major exporter of resource-intensive goods to O rganisation for Economic Cooperation and Development (OECD) markets where trade actions might be contemplated. Exports of such goods to all OECD countries account for less that 7% of I ndia’s total exports.

    Second, another threat may be action to stem the flow of investment that migrates for reasons related to different environmental regulations. Just as the North American Free Trade Agreement (NAFTA) debate was about the giant sucking sound of fleeing capital to cheaper wage locations, similar concerns will arise in the climate context. Here the WTO has no rules affecting i nvestment behaviour, so the current “law and jurisprudence” or rather the lack thereof will not prevent countries from taking a ction to prevent a so-called green race to the bottom or a race to the green bottom. Given President Obama’s recent speech to Congress where he railed against tax breaks for companies that ship jobs abroad, it is quite conceivable that action could be taken to prevent an environmentally motivated outflow of investment. Again, these are most likely to affect the energy-intensive sectors. To the extent that India relies more on domestic capital to finance and operate the domestic technology sector, the less these actions by the US and European Union (EU) will matter.

    Finally, it should again be remembered that all these issues acquire salience only if there is no basic deal on burden sharing. If

    Table 7: Comparing Cap-and-Trade (CAT) and Carbon Tax (CT) Regimes

    partner country behaviour where there are environmental externalities. But it is still unlikely that the WTO would sanction more broad-based trade restrictions – that is, going beyond actions to offset competitiveness – to affect climate change actions by developing countries. Of course, as under the Montreal Protocol, an international treaty that phases out substances that cause ozone depletion, whatever agreement emerges from Copenhagen could include a provision to allow for trade sanctions. But these trade sanctions should serve as enforcement mechanisms to be put in place only after cooperation is secured, not as sticks to induce cooperation in the first place.

    3 Cap-and-Trade, Carbon Taxes and International Implications

    Table 7 summarises the major ways in which future action to r educe the emission of GHGs can be implemented. In order to highlight their implications clearly, the table sets out the alternatives sharply. Essentially there are four types of policy regimes that can prevail. Each country can implement actions through a carbon tax or through a cap-and-trade (CAT) system. Under the former, countries would aim to levy

    Measure International Economic Economic Equity Political Economy Political Ease oftaxes or charges in proportion to the car Implementation Efficiency Efficiency of Resource Economy Monitoring

    bon content of fuels and products. Under

    without with Transfer of Competi-Uncertainty Uncertainty tiveness CAT, governments would establish a ceil-Concerns

    ing (“cap”) on total emissions and assign

    Rankings : 1 = best and 4 = worst

    or auction permits to industrial users to

    National carbon tax Uniform globally 1 4 4 4 (because large and 1 1

    emit up to a certain amount of GHG.

    too transparent)

    These rights held by industrial users

    National carbon tax Differentiated 3 4 3 3 (less necessary) 4 1 globally could be bought and sold (hence the

    National cap-and-trade International 1 1 1 (if caps 1 (implicit in 2 4 “trade” part of cap and trade). The idea trading reflect equity) national caps)

    is that firms that are able to cut their

    National cap-and-trade No international 3 2 1 (if caps 1 (implicit in 3 3

    emissions easily will sell some of their

    trading reflect equity) national caps)

    all countries agree in Copenhagen to reduce carbon emissions across the board, then there should be little basis for trade r estrictions in particular sectors. With economy-wide emissions targets, governments would retain the flexibility to mandate r eductions across sectors of their economy as they see fit. Accordingly, they would be immune to punitive action from their trading partners in specific sectors.

    The international community will have an opportunity to d esign a new regime to manage both climate change and trade at the Copenhagen summit, the most significant such meeting since the conference in Kyoto in 1997. The key objective of that regime should be to ensure the participation of all major carbon-emitting countries, including developing nations. The negotiations should therefore not use the threat of trade sanctions as a tool to force cooperation, as these tend to alienate developing nations. Instead, as Nicholas Stern, former chief economist of the World Bank, has proposed, participation and compliance should be secured through transfers of finance and technology – particularly since most developing countries see climate change as a problem caused by emissions from the industrial world (Stern 2007, 2009).

    But under current rules are these trade sanctions to force cooperation permitted? Again, the law and jurisprudence are fuzzy. Successive disputes, especially the Shrimp Turtle dispute, have expanded the scope of trade measures that can be used to affect

    rights to those less able to do so. The theory is that this trading will lead to a price of emissions that in principle should not be different from that established under a carbon tax. The carbon tax is a price-based policy regime while CAT is a quantity-based regime.

    These two regimes – tax and CAT – can in turn be implemented in two ways. Take two countries, the US and India, and say both adopt a carbon tax regime. Now, both countries need not levy the same tax on carbon. If they do, we will have an international carbon tax regime with a uniform global tax (row 1 in Table 1). If, however, they levy different taxes, there will be no common i nternational price for carbon emissions (row 2). In a similar f ashion, one can think of two CAT regimes: the first in which there is international trading (trading across countries of the rights to emit that each government gives its firms) as depicted in row 3; and a second in which trading across countries is not a llowed (row 4).

    These four regimes can be compared along a number of economic and political dimensions. These comparisons can then form the b asis for countries’ judgments about their different p olicy options.

    Take first the criterion of economic efficiency. For convenience, we label the four regimes as follows: uniform global carbon tax regimes as CTU; carbon tax regimes with different taxes across countries as CTD; CAT with international trading as CATU; and CAT without international trading as CATD. (In the table, these

    august 1, 2009 vol xliv no 31


    four regimes are ranked along the different dimensions, with a rank of 1 denoting the best ability to meet the relevant criterion and 4 the worst ability.)

    On the criterion of economic efficiency (column 1) – one that leads to the same international price for carbon emissions – the CTU and CATU regimes are by design clearly superior to the other two. But price and quantity-based regimes can have different e fficiency consequences under conditions of uncertainty (col umn 2). In a famous paper, Weitzman (1974) showed that quantity-based regimes can be superior under uncertainty. In the case of climate change, quantity-based regimes have the appeal that they more directly target the outcome of interest – namely, the quantity of emissions. Given the considerable uncertainty about the reactions of consumers and producers to price signals, there will be an inevitable process of trial and error to establish the right amount of taxes to achieve the desired quantity outcome. Given the tricky and persistent nature of GHG effects, this trial and error – especially errors that overestimate the ability of taxes to influence quantities – may not be affordable from a scientific perspective.

    If there were no international dimensions to policy actions, a carbon tax regime would be superior to a CAT regime because of two important virtues. CAT systems can typically only target the larger (industrial) sources of emissions, while a carbon tax can affect emissions throughout the economy, including consumers. The second virtue is that a carbon tax system is much easier to administer. There is no need to award or assign permits to t housands of producers.

    For a country like India, three other criteria for assessing these regimes are important: equity, the political economy of resource transfers, and the political economy of competitiveness and the trade frictions that they could provoke.

    Take equity first. Clearly, the question of equity or equitable burden sharing on climate change is a prior issue that will have to be settled through the Copenhagen process or some successor initiative. CAT and carbon tax regimes are or will very much be derivatives of that process. However, if the burden-sharing issue is not resolved early or clearly enough, the choice of regimes can have subtle effects pertaining to equity. The important point here is that the carbon tax regime makes a country’s climate change effort much more transparent. If India imposes say a relatively small carbon tax, that could easily be construed as an inadequate effort. In contrast, a CAT regime is much less t ransparent. The main reason that CAT regimes have gained p olitical support in the US is because they are not visibly seen as a tax. That same invisibility might also have virtues in an international context.

    Of greater importance but related to the above is the political economy of international resource transfers. From the perspective of developing countries such as India, any incremental burden sharing must be accompanied by transfers – of resources and technology. Ideally, efficiency and equity would be combined by having a uniform global tax on carbon (for efficiency reasons) and a system of transfers from these tax collections to address the equity question. But one can safely assume that in the US the notion of financial transfers from the carbon taxes collected to India and China would be completely anathema. A carbon tax regime will therefore not be conducive to equitable burden sharing and resource transfers to help address climate change. A CAT regime could be superior

    Economic & Political Weekly

    august 1, 2009 vol xliv no 31

    from an Indian perspective on this score, again because the caps work through quantities rather than prices (of course, trading under CAT regimes will establish prices for carbon emissions but less visibly than simple carbon taxes). It is not only in India’s interest to adopt a CAT regime but also in its interest that countries such as the US adopt the same regime so as to facilitate resource transfers.

    These two regimes will also have an impact on the political economy of competitiveness. For reasons similar to those discussed above but with greater salience, CAT regimes will better suit countries like India. Suppose India were to adopt a carbon tax regime and suppose that it chose to impose a relatively small tax, consistent with its conception of equitable burden sharing, certainly smaller than what the US and EU should and are likely to adopt. Energy-intensive sectors in industrial countries will quickly point to the lower taxes in India and call for offsetting a ction, especially on trade. This political economy will come into play even if the carbon tax that India adopts were the result of an agreement pursuant to the Copenhagen process. With a CAT r egime on the other hand, it will be more difficult for energy- intensive industries to point fingers because the Indian “effort” will be less transparent.

    So, on balance, a CAT regime may have some advantages for India. In some ways, the best scenario for India might be a CAT regime with international trading. The CAT will, on the one hand, by being less visible, facilitate differential burden sharing across countries and also provoke fewer calls for trade action. On the other hand, it could lead to efficiency because there should be a tendency for the price of carbon to converge.

    Of course, any advantages of a CAT regime should also take into account the political economy within India that might create very high rent-seeking and related costs of implementing such a regime.


    This paper has tried to address and raise questions on three i ssues relating to climate change. The most important objective for India has to be to secure equitable burden sharing, one that does not compromise its future development trajectory. A per capita emissions perspective – future targets based on per c apita emissions – would have best secured this objective. But the political and intellectual ground is shifting beneath India’s feet on this score. The recent signs – at least from US legislative i nitiatives – are that industrial countries are unlikely to be sympathetic to per capita emissions-based targets. I nstead, they are seeking targets based on reductions in total emissions by developing countries comparable or equivalent to those u ndertaken by them.

    We offer a new intellectual approach that tries to bolster the case for a per capita emissions approach. We distinguish the CO2 emissions intensity of production and consumption from energy use per capita. India could argue that energy needs should not be expected to be very different around the world and that future targets should respect that. However, India could offer to make a contribution to improving the emissions intensity of production and consumption in the country, based on transfers of techno logy and resources. We outline some projections based on these assumptions that could lead to a reasonable emissions trajectory for India and one that is consistent with global efforts at addressing climate change.

    The second issue relates to the role of trade in climate change. trading system cannot and should not discharge that responsibil-If equitable burden sharing on future emissions reductions ity. But, if that burden is thrust upon it, the evolution in WTO law can be achieved, trade and trade restrictions will become less and jurisprudence has been such as to slowly tilt the balance in i mportant. If not, the scope for trade restrictions will increase. favour of trade restrictive actions to achieve environmental There could be a messy outcome in which the trading system is o bjectives. That is potentially a worrisome development for

    burdened with the task of settling environmental problems. The d eveloping countries.


    1 A subsequent legislative initiative sponsored by John Dingell and Richard Boucher in the House of Representatives was very similar in spirit to the Lieberman-Warner-Boxer effort. Since the elections in 2009, Congressmen Henry Waxman and Ed Markey have been leading the charge on climate change legislation, and trade provisions are likely to feature in it as the bill winds its way through Congress.

    2 It must be emphasised that the data presented in this paper is based on ongoing research and could hence be subject to revisions, especially since the manner in which the data are being examined is new.

    3 A measurement issue relates to whether GDP should be measured at market-based exchange rates or in purchasing power parity terms. Here, we use the former measure because the bulk of emissions in production tend to be in the manufacturing sector which, because it is largely tradable, should be valued at market exchange rates.

    4 In all the scenarios, we assume a long-run per capita growth rate for India of 5% and, consistent with United Nations projections, a population growth rate of 0.092%.

    5 Implicit in using this production elasticity is the idea that India’s future production intensity of emissions will be similar to that of the industrial countries. One reason to doubt this idea is that India’s development pattern has been atypical, relying more on services than manufacturing. If this pattern continues, and given that services are less CO2 intensive than manufacturing, India could end up being less of an emitter than other countries at their comparable stage of development. The services-intensity of Indian development raises a broader question about future emissions paths for other countries too that might skip the manufacturing stage of development.

    6 Calls for the transfer of technology have been made repeatedly in the past by developing countries in a host of areas but without concreteness on the mechanisms and policy instruments that should be used. Such mechanisms could take several forms: international funding for technological upgrading by Indian firms and for paying for the associated intellectual property and licence fees; international public funding of the underlying research and development combined with free or low-cost dissemination of the fruits of any such research. For a brief description of some of the problems encountered by Indian companies in accessing climate changerelated technologies, see Ghosh (2009).

    7 A complication arises if in the example the Indian measure on carbon is not a tax but a regulation. Then the question is whether any restriction on Indian goods coming into the US could be justified under one of the exception provisions (Article XX of the GATT) in the WTO.


    Ghosh, Arunabha (2009): “Competing Needs: Clean Coal is Key” ( 04/27201218/Business-At-Oxford--Competing.html?h=B)

    Hufbauer, Gary Clyde, Steve Charnovitz and Jisun Kim (2009): Global Warming and the World Trading System (Washington: Peterson Institute for International Economics).

    Mattoo, Aaditya and Arvind Subramanian (2009): “From Doha to the Next Bretton Woods: A New Multilateral Trade Agenda”, Foreign Affairs, 88 (1) (January/February): 15-26.

    Planning Commission of the Government of India (2006): Integrated Energy Policy.

    Stern, Nicholas Herbert (2007): The Economics of Global Climate Change: The Stern Review (Cambridge UK: Cambridge University Press).

    – (2009): “Transatlantic Perspective on Climate Change and Trade Policy”, Keynote address, P eterson Institute for International Economics, 4 March.

    Subramanian, Arvind (1992): “Trade and the Environment: A Nearly Empty Box?” The World E conomy. Weitzman, Martin L (1974): “Prices vs Quantities”, The Review of Economic Studies, 41 (4): 477-91.


    Centre for Study of Social Exclusion and Inclusive Policy Faculty of Social Sciences, Varanasi – 221005

    National Seminar

    “Repositioning ‘Subalternity’ in Globalized India: The Dynamics of Social Inclusion”

    November 19-21, 2009

    Call for Papers

    We invite researchers and academicians to submit abstract and full papers on the following sub themes of the seminar.

  • 1. Mapping subaltern studies in India: Hegemonic expressions of power in 21st century.
  • 2. From ‘servitude to assertion’: Struggle for agency and search for identity.
  • 3. Historiography from the edge – marginal voices
  • 4. Gender, exclusion and empowerment – documenting feminist narratives.
  • 5. Production of knowledge, didactics and subalternity: contemporary debates and future concerns.
  • Deadlines Submission of Abstract (150-200 words, with full address): 31st August 2009. Full paper submission (5,000-8,000 words, typed in Harvard style): 15th October 2009.

    Travel Allowance and Accommodation

    Authors of the accepted papers will be provided II/III tier AC fare, boarding and lodging facilities on the campus. Outstation participants are requested to contact the co-ordinator for lodging facilities.


    A nominal fee of Rs. 1,000/- for teachers & Rs. 500/- for Students/Research scholars is to be paid in the form of demand draft along with the registration form. Demand Draft is to be drawn in favour of the Coordinator, CSSEIP, BHU. Full papers, along with Registration Fee should reach the Coordinator on or before 15th October 2009.

    For further information please contact the Convener of the Seminar on the following address:

    Professor Ajit K. Pandey, Coordinator, CSSEIP Convener of the Seminar Dept. of Sociology, Faculty of Social Sciences, BHU, Varanasi – 221005 Seminar Office: GF-1, Jodhpur Colony (in front of Faculty Guest House), Banaras Hindu University, Varanasi, UP 221005, INDIA Contact No.: (+91) 9795 996951; 9335 669697; 9450 962335, 9889 279897 Email:

    august 1, 2009 vol xliv no 31

    To read the full text Login

    Get instant access

    New 3 Month Subscription
    to Digital Archives at

    ₹826for India

    $50for overseas users


    (-) Hide

    EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

    Back to Top