59204 Environment Notes Climate Change RENEWABLE ENERGY Number -- 06 Quantifying Carbon and Distributional Benefits of Renew- able Energy 1 Programs-- The Bangladesh Case Study on January 2011 Solar Home Systems World Bank Carbon Finance Unit's Photo Library Limin Wang, Sushenjit Bandyopadhyay, Mac Cosgrove-Davies, and Hussain Samad Scaling-up adoption of renewable energy technology-- such as solar home systems (SHS)--to expand electric- ity access in developing countries can accelerate the transition to low-carbon economic development. Using a national household survey, this study quantifies the carbon and distributional benefits of SHS programs in Bangladesh. Three key findings are generated from the study. First, dissemination of SHS brings about significant carbon benefits: the total carbon emissions avoided from replacing kerosene use for lighting by Authors SHS in non-electrified rural households is equivalent Limin Wang, Consultant, Environment Department; Sushenjit Bandyopad- hyay, Senior Environmental Economist, Environment Department; Mac to about 4 percent of total annual carbon emissions Cosgrove-Davies, Lead Energy Specialist, South Asia Sustainable Develop- in Bangladesh in 2007. This figure increases to about ment Energy and Mining Unit; and Hussain Samad, Consultant, Develop- 15 percent if grid-based electricity generation is used ment and Research Group Agriculture and Rural Development--all of the World Bank. as the energy baseline to estimate the carbon avoided from SHS installation. Second, SHS subsidies in rural Acknowledgements Bangladesh are progressive when the program is geo- This Note benefited from the helpful advice and suggestions of Richard graphically targeted. Third, SHS has market potential Hoisier, Sameer Akbar, Doug Barnes, Raihan Elahi, Kirk Hamilton, Shahidur in many rural areas if micro-credit schemes are made Khandker, Glenn-Marie Lange, Subodh Mathur, Zubair Sadeque, and work- shop participants in the World Bank's Washington and New Delhi offices. available. The propensity to install SHS is very respon- sive to income, with a 1 percent increase in per capita This study was supported by the Swedish International Development income increasing the probability of installing SHS by Cooperation Agency (SIDA). 12 percent, controlling other factors. Climate Change RENEWABLE ENERGY Introduction This means that the bulk of the additional investment Number -- 06 for climate change mitigation, in particular in the clean Development agencies place a high priority on provid- energy sector, should flow into developing countries. ing electricity to over a quarter of the world's population Following the Bali Action Plan--which calls for mitiga- currently without access. This is largely motivated by the tion actions by developing countries to be supported and increasing recognition of the broad range of economic enabled by technology, financing, and capacity building and social benefits associated with electricity access from developed countries--a large number of multilat- (World Bank 2002, Wang 2003). Advances in an ar- eral and bilateral funds and financial mechanisms such 2 ray of renewable energy technologies--including wind, as the Clean Development Mechanism (CDM) were solar, biomass, and hydroelectricity--present an op- established (Doornbosch and Knight 2008). Rapidly portunity for developing countries to increase electricity increasing financial resources, both from the public and access while accelerating the transition to low-carbon private sectors, have been channeled into the clean en- economic development. ergy sector,2 thus providing an enormous opportunity to integrate climate change mitigation with economic With a range of off-grid options--in particular solar home development. systems (SHS)1--it is possible to provide the basic elec- tricity needs of households, local communities, and small Maximizing the carbon mitigation and development im- businesses in rural areas where grid-based electricity is not pact of expanded carbon finance depends on the efficient an option in the foreseeable future. The dissemination of and equitable allocation of these resources, as well as SHS over the past two decades has benefited many people better implementation of projects through targeting and in remote areas, providing better quality lighting, enabling coordination with poverty projects. Renewable energy extended working hours, and powering small appliances programs--in particular dissemination of solar home such as mobile phones. These benefits have been achieved systems--have been implemented by international in- with near zero carbon emissions, while also reducing the stitutions and NGOs in many developing countries over use of fossil fuels, such as kerosene for lighting and diesel the past two decades. But few studies have employed fuel for charging batteries. large-scale household surveys to quantify the carbon benefits and the distributional impact of SHS programs. Despite technological maturity and the constant decline This study aims to fill this gap by using the first available in SHS prices, the current level of SHS dissemination national household survey that collects SHS installation among rural populations is low. According to the Global data in rural Bangladesh, which is one of a few countries Status Report 2009 (REN21 2009), out of the 400 mil- that have made significant progress in providing elec- lion households who lacked access to grid electricity in tricity access to rural people through the installation of 2007, only about 2.5 million received electricity from so- solar home systems.3 It focuses on (1) the quantification lar home systems. Scaling up the adoption of low-carbon of the carbon benefits, particularly on kerosene displace- energy technologies in developing countries must be part ment; (2) SHS affordability; and (3) the distributional of the global effort to reduce the devastating risks posed consequences of SHS subsidies. by climate change. IEA projections suggest that devel- oping countries' share of global carbon dioxide (CO2) emissions from energy use will increase from 38 percent in 2002 to 52 percent in 2030, while developed coun- Rural electrification and the SHS tries' share will decline from 60 percent to 47 percent program in Bangladesh (IEA 2007). Clearly, reducing emissions in developed countries alone will not be sufficient to achieve the goal While Bangladesh has made impressive progress in of limiting the global average temperature increase to no expanding rural electrification, its electrification rate more than 2o C (OECD 2008). still lags behind other countries in South Asia. Between Climate Change RENEWABLE ENERGY 2000 and 2008, the rural electrification rate increased projects to electrify village markets and small enterprises Number -- 06 by only about 8 percentage points--to 28 percent. By (World Bank 2009). comparison, South Asia experienced an increase of 18 percentage points--to 48 percent--over the same pe- A World Bank project--Rural Electrification and Re- newable Energy Development (RERED)-- started in riod (Figure 1). 2002 with total funding of $298 million and aims to The Bangladesh government has set a target of provid- expand rural electrification through both grid exten- ing the entire country with electricity by 2020, with sion and renewable sources. The project's success, in improved reliability and quality of electricity supply. particular with the dissemination of solar home systems 3 However, achieving this target requires effective mea- in rural areas, had led to a request from the government sures that can overcome two major constraints currently for additional financing of $130 million in 2009, with facing the electricity sector. First, while per capita elec- $100 million earmarked for scaling-up SHS installation and other renewable-energy-based mini-grids in rural tricity generation capacity in Bangladesh is among the areas. The implementation of SHS programs was carried lowest in the world (at about 165 KWh per year), the de- out using two different delivery models. The first model mand for electricity is growing at a rate of over 500 MW was implemented by the state-owned Rural Electrifi- per year due to population growth, the rapid increase in cation Board (REB). It disseminated SHS through a demand for electrical appliances, and industrialization. fee-for-service program, whereby the system would be As a result, power outages are common occurrences. installed and owned by REB and households would pay Second, the majority of the rural population lives in areas a monthly fixed fee for using the system. The second approach is through a private agency, the Infrastructure that are distant from the national electricity grid. Even if Development Company Limited (IDCOL), which these households were connected to the grid, insufficient received financial support from GEF to implement generation capacity would lead to disproportionate load the dissemination of solar home systems through vari- shedding in rural areas. Realizing that grid electrification ous private agencies, such as Grameen Shakti, using a is not an economically feasible option, the government micro-finance scheme.4 has taken a dual-track approach to expanding rural elec- trification: (1) expanding the electricity distribution grid While REB was able to provide solar home systems to connect new consumers, and (2) making solar home for about 12,000 households, IDCOL reached over systems available to households and promoting biomass 320,000 households over the same period (accounting for about 1.6 percent of non-electrified Figure 1. Rural electrification prog- rural households). The success of the micro-credit scheme by the private sector 2000­08 ress, Rural Electri cation Progress between 2000­2008 in SHS dissemination lies mainly in the 60 fact that these private delivery agencies, 50 in particular Grameen Shakti, have more practical knowledge in providing micro- Rural electri cation rate % 40 finance and greater reach at the commu- 30 nity level (Asaduzzaman and others 2008). 20 The key lessons from the Bangladesh 10 experience is that while scaling-up the adoption of renewable energy technol- 0 Bangladesh India South Asia ogy depends critically on private sector 2000 2008 participation, public support is critical at Climate Change RENEWABLE ENERGY the initial stage, in particular financing. Lessons from Using the carbon emission factor for kerosene (2.45 Number -- 06 the Bangladesh case study can possibly be transferred kg CO2/liter), the avoided CO2 emissions for the most to other countries while taking account of local condi- commonly purchased solar home system (40­50 Wp) is tions. about 76 kg CO2 per year in the context of rural Ban- gladesh. The estimate from the Bangladesh survey is significantly smaller than estimates obtained from other sources, as summarized in Table 1. This is possibly due Carbon benefits, affordability, and to two factors: (1) the estimate for Bangladesh controls 4 distribution impact of RERED for confounding factors; and (2) the Bangladesh study focused only on kerosene; it did not include diesel and A nationally representative household survey was col- dry cells due to data limitations of the survey. lected in 2005 for monitoring and impact evaluation of the RERED project. The survey collected information While displacing kerosene use for lighting and diesel on 20,913 households from 6 divisions, 47 districts, 268 for battery charging are the most direct carbon benefits, subdistricts (i.e., Upazila), and 1,350 villages in rural SHS dissemination can also avoid emissions from new areas, including 1,000 households that had purchased electricity connection through fossil-fuel-based electric- solar home systems under the financing scheme, which ity generation (Kaufman and others 2000). This is par- provides a micro-credit loan and cash subsidies to ticularly applicable in the context of Bangladesh, where households living in non-electrified villages. This survey the government has determined to use SHS as one of data provides an opportunity to (1) quantify the carbon the alternatives to the grid option in its efforts to achieve benefits, (2) assess SHS affordability, and (3) analyze the the target of universal access to electricity by 2020. Us- distributional consequences of SHS programs. ing grid-electricity generation as the energy baseline, the carbon emissions avoided from SHS are equivalent How much CO2 emissions can be avoided from to about 269 kg CO2 per SHS per year, which is about SHS? 3.5 times that estimated using the kerosene displace- The total amount of CO2 avoided depends on how much ment baseline. kerosene consumption is displaced due to SHS instal- The scale of the carbon emissions avoided from SHS lation. The United Nations Framework Convention on adoption can be better illustrated by putting these es- Climate Change procedure defines the fuel consumption timates in the national context of the total number of of the technology/device in use (or that would have been households currently without electricity access (about used) in the absence of project activity as the energy 24 million households in Bangladesh in 2008, according baseline for the quantification of carbon benefits of re- to IEA statistics). If all non-electrified households were newable energy projects (Ybema and others 2009). The provided with solar home systems, the carbon emissions impact of SHS on kerosene displacement is estimated avoided from kerosene displacement per SHS per year using the propensity score matching method, which would be equivalent to about 4 percent of total annual controls confounding factors such as income, house- carbon emissions in Bangladesh in 2007. This figure hold size, and location effects. The results show that, on will go up to about 15 percent if grid-based electricity average, about 2.7 liters of kerosene are displaced per generation is used as the energy baseline to estimate the SHS per month, after controlling for household socio- annual carbon benefit from SHS.6 economic factors.5 The scale of kerosene displacement increases with household income: about 2.3 liters per Is scaling-up SHS possible in rural month for the bottom two income groups, and 3 liters Bangladesh? per month for the top two groups, holding other factors constant. The cost of SHS is significant relative to household incomes in rural Bangladesh. The price of the most Climate Change RENEWABLE ENERGY Number -- 06 TABLE 1. Summary of estimates of CO2 emissions avoided per SHS per year Emissions reduction Country Founding scheme SHS model SHS model (Wp) (kg CO2/yr) Argentina Global Environment Facility 50­400 504 Honduras Activities implemented jointly 30­60 246 India Commercial carbon offset funding 20­53 373 Indonesia World Bank/GEF 50 448 Nepal Government of Nepal 35 79 5 Kenya Commercial cash sales 12­50 205 South Africa Shell/Eskom fee for service 50 230 Swaziland IVAM/ECN triodos commercial credit 50 125 Bangladesh World Bank/GEF 25­85 76 Note: The Bangladesh figure is estimated using the 2005 household survey, but all other studies are not survey-based estimates. Source: Ybema and others 2000. commonly installed system with a 40­50 Wp capacity The spatial analysis also shows that there is market po- was about $556 in Bangladesh in 2002, which was more tential in many rural areas to scale up SHS adoption. than three times the average rural household's annual Among the 42 districts in the sample, 17 districts have an expenditure. Therefore, the major barrier for SHS adop- affordability rate above 25 percent, as well as a relatively tion is the large up-front cost. If micro-credit schemes high proportion of households living in non-electrified are made available, SHS is likely to be an attractive op- villages; the average is about 45 percent among the 17 tion to many households in rural areas, given that grid districts, compared to the national average of 38 percent. electrification is unlikely to be an option for many years The econometric analysis shows that the propensity to to come. purchase SHS is very sensitive to household incomes, with a 1 percent increase in per capita expenditure in- We assessed the affordability of SHS using the average creasing the probability of installing SHS by about 12 energy budget share as the benchmark. A budget share percent, holding other factors constant. of 8 percent, which is estimated based on the existing micro-credit schemes, was used to define affordability. Are SHS programs progressive? That is, households are able to afford SHS under the The general belief based on anecdotal evidence is that existing micro-credit scheme if their budget share for SHS subsidies are not pro-poor because better-off monthly SHS financing is below the level of 8 percent. households disproportionately capture the subsidies. Admittedly, this level of budget share is high, so the Consequently, expanding rural electrification from SHS estimated affordability rate should be regarded as an dissemination is often not considered to be an effective upper-bound estimate. Under this criterion, the total policy choice for addressing poverty issues, although number of households in rural areas that can afford SHS little empirical evidence exists to validate such claims. is about 76,000 households under the existing micro- The Bangladesh national survey data present an oppor- credit scheme, plus a $50 cash subsidy representing about tunity to empirically assess the distributional impact of 24 percent of non-electrified households in the sample SHS programs. districts. This number goes up to about 45 percent if the cash subsidy increases to $90. The distributional consequences are illustrated using a policy simulation exercise under two assumptions. Climate Change RENEWABLE ENERGY First, all households currently without grid electricity from the strong spatial correlation between the concen- Number -- 06 are assumed to be entitled to the micro-finance scheme. tration of non-electrified households and the poverty Second, the private sector agencies who are responsible rate, as shown in Figure 3. for SHS dissemination decide program locations at the upzila level. The choice of location is driven by the objec- This finding indicates that targeting, as well as better tive of maximizing SHS dissemination while minimiz- integration of renewable energy projects with existing ing operational cost. Two indicators that are important development projects at the local level, is critical to en- from the perspective of delivery agencies include (1) the hancing the synergies between carbon mitigation and affordability rate, and (2) the proportion of households development. For example, placing SHS programs in 6 living in villages without grid electricity, both capturing localities where the affordability rate is sufficiently high the market potential of SHS as well as the scale of the and where poverty alleviation programs are in place operational cost. The distributional consequences are can avoid duplication, while maximizing the impact essentially determined by the location choices of the through resource pooling and coordination. With the private sector delivery agencies. rapid increase in carbon finance for climate mitigation projects in developing countries, policy makers should The distributional impact of the SHS program is analyzed now focus on how renewable energy projects should be using the concentration curve, which plots the cumula- targeted and integrated with poverty programs. tive percentage of SHS subsidies received by households against the cumulative percentage of household popula- tion, ranked by per capita income in ascending order.7 Figure 2 shows that targeting SHS programs based either Conclusions and policy messages on affordability rate or the proportion of non-electrified Using the Bangladesh national household survey, this households will be progressive. As expected, the loca- study provides three key findings. First, dissemination tion choice based on the latter is more equitable, with of SHS for rural electrification can generate substantial the bottom 30 percent of households receiving about 55 carbon benefits in the context of rural Bangladesh. The percent of total subsidies, while they only receive about annual carbon avoided from kerosene displacement as a 45 percent if the targeting is based on affordability rate. result of SHS installation would be equivalent to about 4 The positive distributional consequences result mainly percent of total annual carbon emissions in Bangladesh in 2007 if all non-electrified households were pro- Figure 2. The distributional vided with solar home systems. This figure will go impact of location choices up to about 15 percent if grid-generated electricity is used as the energy baseline to estimate the carbon benefit from SHS. 1 Choose upzilas by non-electrification rate Second, under the assumption that the existing .8 Cumulative Subsidy Proportion micro-credit scheme plus a cash subsidy is made available to all non-electrified households in rural .6 Choose upzila by affordability rate Bangladesh, the affordability assessment indicates that scaling-up SHS adoption is possible in many .4 rural areas. Among the 41 districts in the survey, about 17 districts have an affordability rate over .2 25 percent, as well as a high concentration of non- electrified households in comparison to the national 0 0 .2 .4 .6 .8 1 average. This means there is a potential market for Cumulative Population Proportion solar home systems. It is also possible to reduce the Climate Change RENEWABLE ENERGY Figure 3. Upzila level poverty and non-electrification rates Number -- 06 in Bangladesh 7 costs of SHS dissemination due to economies of scale to Second, the rapid increase in financial resources chan- promote profitable participation of the private sector in neled to climate mitigation, in particular in the clean the SHS market. energy sector, presents an opportunity to integrate re- newable energy projects with development projects to Third, contrary to the commonly held view that subsi- reinforce synergies between climate change mitigation dies to promote SHS dissemination in rural areas ben- and development. This means that efforts must focus efit mainly better-off households, our policy simulation on improving the efficient and equitable allocation shows that SHS subsidies in rural Bangladesh are pro- of carbon finance to projects that generate the largest gressive when programs are targeted to localities based carbon and development benefits. At the project level, either on the affordability rate or the concentration of the design and implementation of renewable energy non-electrified households. projects should focus on key issues--such as targeting, integration, and coordination with poverty alleviation Two policy messages emerge from this study. First, the programs--in order to maximize the carbon mitigation Bangladesh experience shows that while the potential and development benefits. for scaling up SHS in rural areas exists if the up-front cost of SHS can be addressed through improving ac- cess to micro-credit in combination with cash subsidies, References the real challenge lies in how these programs can be Asaduzzaman, M., D.F. Barnes, and S. R. Khandker. 2008. implemented on the ground. The success of SHS dis- Restoring Balance: Bangladesh's Rural Energy Realities. semination in rural Bangladesh depends critically on ESMAP Report. Washington, DC: World Bank. active policy support, in particular at the initial stage Chaurey, A., and T. C. Kandpal. 2009. "Carbon abate- of the operation. This support should include financing, ment potential of solar home systems in India and technical assistance, SHS information dissemination, their cost reduction due to carbon finance." Energy and the development of institutional capacity and hu- Policy 37: 115­125. man resources at the community level. Climate Change RENEWABLE ENERGY Doornbosch, R., and E. Knight. 2008. "What role for ity standards in many countries. The price has been Number -- 06 public finance in international climate change mitiga- declining rapidly over the past few years; the average tion?" OECD discussion paper. Paris: OECD. price for a 40­50 Watt peak system (Wp) is about IEA (International Energy Agency). 2007. World Energy about $200­$300. Outlook. Paris: IEA. 2. The World Bank Group has seen a steady increase in IFC (International Finance Corporation). 2007. Selling the share of energy sector financing committed for Solar: Lessons from more than a decade of IFC's experi- low-carbon renewable energy and energy efficiency, ence. Washington, DC: IFC rising to about 40 percent in 2007 from 13 percent in 8 Kaufman, S., R. Duke, J. Rogers, R. Schwartz, and M. 1990­94. By 2007, about 650,000 solar home systems Trexler. 2000. "Rural electrification with solar energy had been installed across 23 countries. as a climate protection strategy." Research Report No. 9. 3. The Bangladesh SHS programs that were financed by Washington: Renewable Energy Policy Project. the Global Environment Facility (GEF) and Interna- OECD (Organisation for Economic Co-operation and tional Development Association (IDA) had installed Development). 2008. Environmental Outlook to 2030. over 300,000 solar home systems (about 1.6 percent Paris: OECD. of rural households without electricity) in rural areas REN21. 2009. Renewables: Global Status Report 2009 Up- by 2009. date. Renewable energy policy network for the 21st 4. Under this scheme, households were provided with a century. loan for a period of 4­5 years, at an annual interest Wang L. 2003. "Determinants of child mortality in LDCs: rate of 12 percent, plus a $50 cash subsidy. House- Empirical findings from demographic and health sur- holds who received the loan were required to make a veys." Health Policy 65 (3): 277-299 down payment of 10 percent of the total cost of the World Bank. 2002. Rural Electrification and Development system. in the Philippines: Measuring the Social and Economic 5. The estimated reduction is 3.9 liters/month from a Benefits. Energy Sector Management Assistance Pro- smaller scale household survey (441 households) con- gram, Report No. 255/02. Washington, DC: World ducted by Grameen Shakti in 2009. Chaurey and Bank. Kandpal (2009) provide an estimate of 9.6 liters/ World Bank. 2009. International Development Association month for rural households in India, and the World Project Documents: Bangladesh Rural Electrification and Bank project report finds 19.6 liters/month displace- Renewable Energy Development Project. Washington, ment for rural households in Indonesia. DC: World Bank. 6. The above estimated carbon benefit is not measured in Ybema, J.R., J. Cloin, F.D.J. Nieuwenhout, A.C. Hunt, terms of life cycle emissions. and S.L.Haufman. 2000. "Towards a streamline CDM 7. If the concentration curve (CC) lies above the 45 de- process for solar home systems." The Netherland Energy gree line of equality, the allocation of subsidies is pro- Research Foundation ECN report no. ECN-C-00-109. gressive, with the poor households capturing the to- Petten: Energy research Centre of the Netherlands. tal subsidies disproportionately, and vice versa if it lies below the equality line. The CC can also be used to compare different policy options. If one CC lies ev- Endnotes erywhere above another one, the first curve is said to 1. A typical SHS that consists of a PV module with a dominate the second one, in the sense that the first 20-year life cycle, a controller, and a rechargeable lead- curve represents a more progressive policy option than acid battery can be easily manufactured to good qual- the second one. Environment Department 1818 H Street, NW The World Bank Environment Notes Washington, DC 20433 USA www.worldbank.org/environment