Alleviating Energy Poverty For The Poor

ARTICLE IN PRESS

Energy Policy 33 (2005) 1367–1372

Viewpoint

Alleviating energy poverty for the world’s poor Ambuj D. Sagar* Science, Technology, and Public Policy Program, Belfer Center for Science and International Affairs, John F. Kennedy School of Government, Harvard University, 79, John F. Kennedy St., Cambridge, MA 02138, USA

Abstract Improving energy services for poor households in developing countries remains one of the most pressing challenges facing the development community. The dependence of these households on traditional forms of energy leads to significant health impacts as well as other major disbenefits, yet there has been little progress in meeting this challenge. This viewpoint argues for an ‘energypoverty alleviation’ fund to help provide modern energy services to these households. It also proposes an approach through which to create such a fund, namely by introducing an incremental levy on petroleum. Notably, this scheme does not need a global agreement since a levy could be introduced by major oil-exporting countries. The implementation of this mechanism would result in a climatefriendly outcome (even before taking into account the elimination of products of incomplete combustion resulting from the traditional household use of biomass-based fuels) while providing immense socio-economic benefits to the world’s poor. Such an approach would allow significant progress on the sustainable development front while reducing global greenhouse gas emissions, and therefore is very much consistent with the United Nations Framework Convention on Climate Change. r 2004 Elsevier Ltd. All rights reserved. Keywords: Biomass; Climate change; Sustainable development; Equity

1. Introduction An estimated 2 billion people worldwide, mostly in rural areas, continue to suffer from energy poverty (World Bank, 1996, Goldemberg et al., 2000). This remains one of the major challenges facing the development community. Although much attention has been focused towards this problem, only limited progress has been made in tackling it. The per-capita energy consumption of these individuals, comprising the poorest third of humanity, is a minuscule fraction of that of citizens of industrialized countries and much smaller than even that of urban dwellers in developing countries. They do not have access to safe, clean fuels and subsist mainly on traditional energy sources such as animal dung, crop residues, and wood (Reddy et al., 1997; Goldemberg et al., 2000). The continuing reliance of poor households on such forms of energy comes with major disbenefits including: *

substantial, and often increasing, time and effort to procure firewood or other forms of biomass—for

*Corresponding author. Tel.: +1-617-496-6218; fax: +1-617-4960606. E-mail address: ambuj [email protected] (A.D. Sagar). 0301-4215/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.enpol.2004.01.001

*

*

example, in rural sub-Saharan Africa, many women have to carry 20 kg of fuel wood an average of 5 km every day (IEA, 2002); possibly high price per unit of energy services (since subsidies often increase as one goes up the energy ladder (Reddy et al., 1997)); and severe and widespread health impacts associated with indoor air pollution resulting from the inefficient combustion of energy sources in poor households, with women and children facing particular risk (Smith, 1993). Recent estimates by the World Health Organization suggest that about 1.6 million premature deaths can be annually attributed to indoor air pollution from biomass and coal use in poor households in developing countries (WHO, 2002, cited in Smith, 2003), which makes it the sixth largest health risk factor in developing countries. Indoor smoke from these solid fuels is, in fact, responsible for about 38 million disability-adjusted lost years (DALYs)1 in developing countries with their attendant social and economic costs (WHO, 2002).

1 One DALY represents one healthy year of life lost by an individual due to disease/adverse health condition.

ARTICLE IN PRESS 1368

A.D. Sagar / Energy Policy 33 (2005) 1367–1372

For these reasons, the continued dependence on traditional fuels for household energy use presents a particularly troubling aspect of the energy poverty problematique. Why, then, has little progress been made on this front? Government agencies, non-governmental organizations (NGOs), development organizations and other actors have been involved in a variety of programs to enhance—quantitatively and qualitatively—energy services for the poor. But the funds channeled toward rural energy constitute only a small fraction of the total funds targeted for the expansion and transformation of the energy sector in developing countries; it is the commercial energy segment that has received much of the attention as well as investments by developing country governments, private actors, and bilateral/multilateral development agencies. Furthermore, reducing the problems associated with the use of traditional fuels in poor households remains only one of a range of objectives of rural energy programs.2 As a result, even though efforts such as those promoting cleaner energy-conversion devices (e.g., improved cookstoves) as well as lesspolluting fuels such as biogas have shown some success (see, for example, Smith et al., 1993), the resources and attention devoted to improving household energy use are not at all commensurate with the magnitude of the problem. There also remain numerous barriers to developing and disseminating new rural energy technologies as well as ensuring their effective use over the long-term. While this is particularly germane for technologies such as solar PV, which are hindered by their high costs as well as technological complexity, it is also relevant for other devices such as improved cookstoves, biomass gasifiers, and biogas digesters. As a result, the dominant approaches for providing improved energy services to poor households—promoting the cleaner use of biomass or other renewables— have brought only limited progress. Furthermore, the time needed for the development and widespread deployment of advanced renewable technologies, as well as their economics, also hinders them in making a significant contribution on this front in the near future. All of this has contributed to a distinct lack of progress on provision of modern energy services for household use in rural areas.

provide savings in greenhouse gas (GHG) emissions compared to conventional options for enhancing rural energy supply such as electricity grid extension or diesel gensets. But climate-driven programs have mostly bypassed household energy use for cooking and heating since this generally utilizes biomass-based energy sources that have been traditionally regarded as carbon neutral. This reliance by the poor on the so-called ‘climate-friendly’ biomass energy sources combined with their low levels of energy use has resulted in a lack of motivation for policy-makers and NGOs to focus on this group in the climate context. Recent research has somewhat upended the conventional wisdom regarding the climate neutrality of biomass combustion in households. It has been shown that inefficient combustion of traditional biomass fuels in cookstoves yields significant gaseous products of incomplete combustion (PICs) that are GHGs (Smith et al., 2000). For example, it is estimated that, for every megajoule (MJ) of heat delivered, even the use of renewably harvested wood in the currently prevalent mix of household stoves in India leads to two-thirds the carbon-equivalent emissions of the ‘basic’ GHGs (i.e., CO2, CH4, and N2O) in comparison to kerosene stoves (Smith et al., 2000). For non-renewable wood, dung, or crop residues, emissions of these GHGs (for the fuel/ stove combinations prevalent in India) are well in excess of those from kerosene stoves on a per-unit-heatdelivered basis.3 It should also be noted that when other PICs such as non-methane hydrocarbons and carbon monoxide are also included in the calculations, these numbers rise even further (Smith et al., 2000). Additionally, the emission of dark particles from biomass combustion also contribute to what UNEP has termed the ‘‘Brown Cloud’’ (UNEP and C4, 2002)— this has received attention recently because of its potential implications for regional climate patterns, reduced agricultural productivity, and respiratory ailments over long distances. One would expect these links between rural energy use and the climate issue to provide some impetus towards widespread deployment of devices such as improved cookstoves or fuels such as biogas whose use would reduce emissions of GHGs as well as dark particles which in turn would result in climate benefits while also yielding immense ancillary social and health benefits. Yet this has not happened, at least not on a scale that is

2. The missing link to the Climate Convention

3 For the case of India, it is estimated that the global warming commitment (GWC) of the basic set of GHGs emitted by the use of various fuels for delivering 1 MJ of energy service is as follows (in equivalent grams carbon as CO2): kerosene: 39; renewable wood: 26; non-renewable wood: 165; crop residues: 66; and dung: 158 (Smith et al., 2000). Note that the emissions for any fuel source vary with the stove in which it is burnt, and therefore these estimates of average emission from these fuel source are based on estimates of the distribution of their use in various stoves in the Indian context.

The climate issue has also provided some impetus to many rural renewable energy programs since these 2

Other objectives include providing electric power for rural industries, for improved healthcare facilities, as well as for public goods such as lighting.

ARTICLE IN PRESS A.D. Sagar / Energy Policy 33 (2005) 1367–1372

required to make a significant dent in this problem. Furthermore, it is unlikely that GHG-mitigation-project developers will devote much attention to projects of this nature, given that transaction costs as well as monitoring and verification requirements for such small-scale individual emitters may impose a high burden. Hence, this large group of rural and urban poor— comprising one-third of humanity—has been, and is likely to remain, virtually excluded from the climate discussions, in large part because its members are responsible individually only for very low levels of GHG emissions (although the aggregate emissions from this large group are not insignificant). This is an unfortunate situation, and somewhat ironical. The benefits of international cooperation are likely to flow to major current or future GHG emitters to help them reduce their emissions, while low-emitting individuals or households, who by virtue of their (admittedly involuntary) energy-use patterns already contribute to the achievement of the primary objective of the United Nations Framework Convention on Climate Change (UNFCCC), continue to be ignored (Sagar and Banuri, 1999). This amounts to something akin to a ‘‘polluters get paid’’ principle (Sagar, 1999). While those emitting large quantities of GHGs in the South do need financial and technical assistance to curtail the growth of their emissions, this should not result in the exclusion of lowemitters from the UNFCCC discussions, especially since the climate issue is a major driver, and increasingly so, in the transformation of the global energy sector. In fact, the principle of equity stated in the UNFCCC as well as the sustainable development imperative highlighted in both the Climate Convention and the Kyoto Protocol clearly suggest that greater attention must to be paid to this group of individuals who already contribute only low GHG emissions.

3. An ‘energy-poverty alleviation’ fund Given all this, the question is how to help in the sustainable development of these individuals in a manner that is consistent with the aims and principles of the UNFCCC. The dominant approach of promoting the cleaner use of biomass or other advanced renewable technologies has brought only limited progress for the reasons mentioned earlier. It has been pointed out that the household energy needs of this group correspond to a small fraction of the global energy consumption (Goldemberg et al., 2000).4 Recently, Smith (2003) has also suggested the substitu4 Goldemberg et al. (2000) suggest that the annual cooking needs of the 2 billion people not served by modern fuels correspond to about 1% of global commercial energy consumption or 3% of global oil consumption.

1369

tion of traditional fuels by fossil fuels for these individuals—while eminently logical, given the large ancillary benefits and the marginal increase in GHG emissions, such an approach would require substantial funds and also may find some opposition from climate/ environmental groups who are ideologically averse to substituting renewables by fossil fuels. I highlight here the conceptual outline of a specific mechanism through which to promote an energy transition in poor households that would enhance their human development while remaining consistent with the aims and principles of the Climate Convention.

3.1. An ‘energy-poverty alleviation’ levy on petroleum The basic concept underlining this mechanism is rather straightforward. It involves levying an incremental price increase on petroleum that could be termed an ‘energy-poverty alleviation’ levy, and then using the funds collected to provide cleaner-burning fuels such as kerosene or LPG for meeting the household energy needs of this poorest part of humanity. Helping move these individuals up the energy ladder would be a major step towards enhancing their welfare. Additionally, some part of the funds collected could also be used to develop improved technologies for kerosene or LPG use among the rural/urban poor.5 How large would such a levy have to be to make a useful contribution? We assume that the annual basic energy needs of these 2 billion people would be covered by about 2.1 exajoules (EJ) of delivered energy.6 Since energy delivery devices for kerosene and LPG have an energy efficiency of about 50% (Baldwin, 1987, cited in Goldemberg et al., 2000), this translates to about 100 million tons oil equivalent (mtoe) of supplied energy. Since the current annual global consumption of crude oil is about 3500 million tons (MT) (IEA, 2003), we would need to levy a price increase of about 3.7% to raise sufficient funds to purchase about 100 mtoe of kerosene or LPG (assuming that the prices of these refined products are about 30% higher than the crude oil precursor (see, for example, Maples, 2000)). While a global approach to such a levy would be most desirable, it would require agreement by a large number of nations (including those that are both major producers as well as consumers such as the United 5 Just as the funding for improved cookstove programs is not very large (Barnes et al., 1993), there is also little directed effort to develop improved kerosene or LPG stoves that could be used in poor households. 6 This is based on the estimate by the World Energy Council (1999) that the annual basic energy needs of the poor can be met by about 1.04 GJ of useful energy per capita. This estimate relies on an assessment by the Indian Advisory Board on Energy, and is slightly lower than that indicated in Goldemberg et al. (2000).

ARTICLE IN PRESS 1370

A.D. Sagar / Energy Policy 33 (2005) 1367–1372

States). This would significantly constrain the possibility of such a mechanism being implemented. An alternate, and possibly more workable, approach could be for OPEC members (or some larger group of oil-exporting nations) to agree among themselves to impose such a levy on the sales of their oil. OPEC members account for 40% of the world’s oil production and are responsible for almost 55% of global oil exports (EIA, 2003a); furthermore, OPEC members have much of the world’s swing oil production and therefore determine, to a large extent, global oil prices. Other major oil exporters include Russia, Mexico, and Norway. Exports of oil (crude as well as refined products) by these countries, along with OPEC members, add up to almost 50% of the global oil consumption (EIA, 2003a). Thus a levy imposed on the exports of oil by OPEC or this larger group of 14 countries (OPEC members plus Russia, Mexico, and Norway) should be able to raise significant funds that could underwrite, at least in part, an energy transition for the world’s poor. For example, a levy of 8% on crude exports (and an equivalent amount on refined product exports) by this ‘‘Group of 14’’ would raise sufficient funds to provide LPG or kerosene to cover fully the basic energy needs of these households. 3.2. Positive global climate implications and large socioeconomic benefits for the poor What are the carbon emission implications of such a scheme? While it would seem that a shift from biomass-based energy sources to fossil fuels such as kerosene and LPG would lead to an increase in GHG emissions, two issues need to be considered. First, there will be some reduction of oil consumption among current consumers due to the imposition of the levy (as illustrated by the elasticity-of-demand calculations below), which will lead to a reduction in carbon emissions. In addition, the elimination of the PICs resulting from inefficient combustion of biomass will help further offset the increase in carbon emissions that result from the switch from biomass to kerosene or LPG. As an illustrative calculation, let us assume a levy of 8% imposed by this ‘‘Group of 14’’ on their crude exports (and an equivalent amount on refined product exports). If we assume a long-term price elasticity of demand to be about –0.5 (i.e., a 10% increase in prices reduces demand by 5%),7 then this levy will result in a reduction in global oil consumption by 4%, i.e., down to 7 There is a wide variation in the estimates of the price elasticity of demand for oil. We use an aggregate price elasticity of –0.5, generally consistent with Gately and Huntington (2002) who estimate a price elasticity of –0.64 for OECD countries and –0.18 for non-OECD countries. The former group account for about 62% of the global oil consumption and the latter group for about 38% (EIA, 2003a).

about 3360 MT.8 Taking into account the 100 mtoe of LPG or kerosene that would be made available for rural energy households, the total global consumption would still be equivalent to about 3460 MT of oil.9 Oil accounts for about 42% of the carbon emissions globally from fossil-fuel use (EIA, 2003b) and carbon emissions from energy-use account for about two-thirds of the global GHG emissions in carbon-equivalent terms (IPCC, 2000). Hence, the implementation of such a levy would result in a decrease in global GHG emissions by about 0.3% due to the overall reduction in oil consumption.10 Accounting for the elimination of PICs from the use of traditional fuels would result in additional reduction in GHG emissions.11 It should also be noted that even with a move to fossil fuels, it is highly unlikely that members of this part of humanity will become large emitters of GHGs, given their economic situation and low energy consumption levels. Thus the climate implications of moving these 2 billion people up the energy ladder though the kind of scheme suggested above are quite favorable. Additionally, the health and social benefits of implementing such a mechanism would be enormous. It would mitigate the health burden of indoor air pollution caused by the combustion of traditional biomass sources. It would reduce the time and drudgery required for gathering firewood and other biomass resources—some of the saved time could be used for other economically productive applications in the case of adults, and for education in the case of children. Decreasing the reliance 8

Such a mechanism could also be used to raise funds by imposing a levy on the sale of other fossil fuels such as natural gas. 9 This ignores the slightly higher energy density (on a mass basis) of kerosene and LPG as compared to oil (see IEA, 2003). 10 This first-order calculation ignores secondary increases in emissions resulting from some inter-fuel substitution to natural gas and coal that might take place because of the price increase in oil. It also ignores the lower carbon coefficients for kerosene and LPG as compared to oil (see EIA, 2002). 11 Smith et al. (2000) indicate that in India, wood accounts for about 63% of the biomass burnt in cookstoves, crop residues for about 20%, and dung for the remaining 17%. Although the total energy efficiency (TEE) of various biomass/stove combinations varies significantly, for a first-order calculation, we assume that the aggregate TEE of wood/ stove combinations as used in India is about 0.2, of crop residue/stove combinations as used, about 0.11, and of dung/stove combinations as used, about 0.09 (based on the data in Smith et al., 2000). We also assume that the calorific values of wood and crop residues are roughly similar, and dung about 20% lower (based on the data in Smith et al., 2000). This would indicate that wood supplies about 79% of the energy delivered, crop residues about 14%, and dung about 7% in Indian poor households. Even if we conservatively assume that all this wood is harvested renewably, the global warming commitment of the ‘basic’ GHGs emitted by the mix of traditional fuels in the Indian context would be about the same as the GHGs emitted by kerosene use for an equivalent amount of energy delivered. Note that if other PICs such as non-methane hydrocarbons are also included in the calculation, then the shift to kerosene is even more favorable. Note also that LPG is more attractive than kerosene since it burns more cleanly.

ARTICLE IN PRESS A.D. Sagar / Energy Policy 33 (2005) 1367–1372

on biomass for meeting the energy needs would lessen the pressure on the environment. And it would also make these groups less vulnerable to climate change by reducing their dependence on natural resources. 3.3. Feasibility and implementation issues It should be noted here that OPEC has been greatly concerned with the imposition of carbon taxes on fossil fuels (as have other major oil exporters) because this would depress the demand for petroleum in the long term, and also because the taxes would be imposed and collected by individual countries and hence this would cost these exporters revenue.12 But by applying this levy themselves, and then using the funds generated to provide kerosene or LPG for poor households, OPEC members or other oil-exporting countries will not lose any revenue. On the contrary, they gain some additional revenue compared to the base case and moreover add a large fraction of humanity as new customers for their products at a time when the long-term value of their resources may be in jeopardy as climate constraints start becoming more pressing. Furthermore, such an approach may also have significant public relations value for OPEC members: instead of being singled out for opposing progress on environmental issues, they can instead claim credit for making an immense contribution to sustainable development. It would also make allies of other developing nations, and hence strengthen the ‘‘Group of 77’’ (G-77) coalition. It should also gain OPEC countries the support of environment, development, and human rights NGOs as well as multilateral development agencies. There is another major advantage of this approach. Since the deal does not need to be struck globally, it reduces the number of extraneous actors who can shape the final outcome or even play the role of ‘spoiler.’ If, for example, OPEC members follow such an approach in cooperation with other G-77 countries, they together can work to design an effective mechanism rather than having a global discussion that would much more likely be shaped by major industrialized countries and/or multilateral development agencies. An alternate fund for the specific developmental purpose of energy-poverty alleviation that puts developing countries in charge will be a welcome change and enhance the participation in agenda-setting and implementation by their governments as well as NGOs and other groups. Of course, there will probably need to be oversight mechanisms to ensure that the funds get used for the purpose for which they are collected, i.e., helping provide cleaner and 12

Already, national taxes on petroleum generate higher revenues for the governments of consuming countries, as compared to the full export income of oil producers (OPEC, 2001).

1371

improved energy services for the poor. Such a scheme also avoids bilateral deals of the kind that are the most likely under the emerging climate regime in which developing countries generally are at a disadvantage in negotiating the best deals.13 While this approach mainly targets consumers in wealthy countries (since they consume the bulk of the world’s oil), it also targets petroleum users in developing countries who are often high GHG emitters (relative to their poor compatriots). This is a major departure from current discussions in the climate regime that makes a strict separation between poor and rich countries. In this scheme, the levy would impact consumers globally, but in proportion to their use of petroleum products. Hence, industrialized country consumers would still bear a greater burden on average than their developing country counterparts. They, of course, will likely not be particularly supportive of such a mechanism but the proposal is constructed so as to obviate the need for their cooperation. Furthermore, the price increases under this proposal would be extremely modest, i.e., of the order of a few percent.14 While a levy of 8% would be needed to fully cover the basic energy needs of the world’s poor, even a smaller levy would still make a significant contribution. It should also be highlighted that this price increase would not be to enrich the OPEC but to help the world’s poor.15 Some issues will clearly need further consideration, such as: *

*

*

What would be the best mechanism for distributing kerosene/LPG to rural/urban poor? There is also a need for special attention to ensure that these funds result in additional benefits for the poor (rather than replacing existing programs that allow national governments to utilize freed-up finances for other goals). What would be the best institutional arrangement for imposing this levy as well as for distributing fuels to poor? This will require an examination of the possible roles of both supra- and sub-national institutions. What would be the general-equilibrium implications for the poor of such a levy on petroleum products, and how could these be minimized?

13 There will likely be more ‘sellers’ than ‘buyers’ for GHG mitigation projects. Additionally, developing country actors’ (in the public or private sector) often suffer from relative lack of information as well as analysis and negotiating skills. 14 A price increase by a small percent is quite modest in comparison to the fluctuations that already take place in oil prices. For example, spot crude prices increased by over 50% between 1999 and 2000 (BP, 2003). Also, for reference, oil prices increased by over three times between 1973 and 1974 during the ‘‘oil crisis’’. 15 To avoid any criticism on this front, OPEC members as well as other participating oil exporters could even offer the LPG or kerosene (or oil for that purpose) at somewhat concessionary rates.

ARTICLE IN PRESS 1372

A.D. Sagar / Energy Policy 33 (2005) 1367–1372

4. Conclusion While such a mechanism can proceed outside the UNFCCC, there are a number of reasons why the Climate Convention should be supportive of it. First of all, this should be seen not as a mechanism that switches over biomass users to fossil fuels and hence undercuts the Convention but one that contributes in a major way to the sustainable development of a significant fraction of humanity that is already contributing to the objective of the Climate Convention by virtue of its low per-capita GHG emissions. The mechanism moves the members of this group up the energy ladder, and that may result in some increase of their GHG emissions through the switch to fossil fuels, but this is consistent with the recognition in the UNFCCC of the fact that emissions need to grow to meet developmental needs. Furthermore, in this case, since the mechanism transfers emissions from current oil consumers (and hence GHG emitters) to the 2 billion low-emitters (while also eliminating PICs resulting from the traditional use of biomass), the global result is climate friendly. Such a transfer is also consistent with the equity principle enshrined in the UNFCCC. In order to be successful, the Climate Convention needs to protect the climate system while promoting sustainable development. This mechanism allows substantial movement towards sustainable development while reducing global GHG emissions. What can be a better yardstick for progress?

Acknowledgements The research for this paper was undertaken as part of the Energy Technology Innovation Project at the Belfer Center for Science and International Affairs, Kennedy School of Government, Harvard University, with support from the Energy Foundation, the Heinz Family Foundation, the William and Flora Hewlett Foundation, the David and Lucile Packard Foundation, and the Winslow Foundation. The author would also like to thank Chella Rajan and Milind Kandlikar for their helpful comments on earlier drafts of this paper.

References Baldwin, S.F., 1987. Biomass Stoves: Engineering Design, Development, and Administration. Volunteers in Technical Assistance, Arlington, VA.

Barnes, D.F., Openshaw, K., Smith, K., van der Plas, R., 1993. The design and diffusion of improved cooking stoves. World Bank Research Observer 8 (2), 119–141. BP (British Petroleum), 2003. BP Statistical Review of World Energy, BP, London, 2003. EIA (Energy Information Administration, US Department of Energy), 2002. Emissions of Greenhouse Gases in the United States, 2001. DOE/EIA, Washington, DC. EIA (Energy Information Administration, US Department of Energy), 2003a. International Energy Annual, 2001. DOE/EIA, Washington, DC. EIA (Energy Information Administration, US Department of Energy), 2003b. International Energy Outlook, 2003. DOE/EIA, Washington, DC. Gately, D., Huntington, H.G., 2002. The asymmetric effect of changes in price and income on energy and oil demand. The Energy Journal 23 (1), 19–55. Goldemberg, J., Reddy, A.K.N., Smith, K.R., Williams, R.H., 2000. Rural energy in developing countries. In: Goldemberg, J. (Ed.), World Energy Assessment: Energy and the Challenge of Sustainability. United Nations Development Program, New York. IEA (International Energy Agency), 2002. Energy and poverty. In: World Energy Outlook, 2002. IEA, Paris. IEA (International Energy Agency), 2003. Energy Balances of OECD Countries, 2000–01 IEA, Paris. IPCC (Intergovernmental Panel on Climate Change), 2000. Special Report on Emissions Scenarios. Cambridge University Press, Cambridge, UK. Maples, R.E., 2000. Petroleum Refinery Process Economics. PennWell, Tulsa. OPEC (Organization of Petroleum Exporting Countries), 2001. Who Gets What from Imported Oil? OPEC, Vienna. Reddy, A.K.N., Williams, R.H., Johansson, T.B., 1997. Energy After Rio: Prospects and Challenges. United Nations Development Programme, New York. Sagar, A.D., 1999. A polluters get paid principle? Environment 41 (9), 4–5. Sagar, A.D., Banuri, T., 1999. In fairness to current generations: lost voices in the climate debate. Energy Policy 27 (9), 509–514. Smith, K.R., 1993. Fuel combustion, air pollution exposure and health: the situation in developing countries. Annual Review of Energy and the Environment 18, 529–566. Smith, K.R., 2003. In praise of petroleum? Science 298, 1847. Smith, K.R., Gu, S.H., Huang, K., Xiu, D.X., 1993. 100 million improved cookstoves in China—how was it done? World Development 21 (6), 941–961. Smith, K.R., Zhang, J., Uma, R., Kishore, V.V.N., Joshi, V., Khalil, M.A.K., 2000. Greenhouse implications of household fuels: an analysis for India. Annual Review of Energy and the Environment 25, 741–763. UNEP and C4 (United Nations Environment Programme and Center for Clouds, Chemistry, and Climate), 2002. The Asian Brown Cloud: Climate and Other Environmental Impacts. UNEP, Nairobi. WEC (World Energy Council), 1999. The Challenge of Rural Energy Poverty in Developing Countries. WEC, London. WHO (World Health Organization), 2002. World Health Report: Reducing Risks, Promoting Healthy Life. WHO, Geneva. World Bank, 1996. Rural Energy and Development: Improving Energy Supplies for Two Billion People. World Bank, Washington, DC.