Embed Size (px)
Citation preview
Energy Policy 40 (2012) 110–120
Contents lists available at ScienceDirect
Energy Policy
0301-42
doi:10.1
E-m
journal homepage: www.elsevier.com/locate/enpol
Overcoming obstacles against effective solar lighting interventionsin South Asia
Sam Wong
School of Environmental Sciences, University of Liverpool, UK
a r t i c l e i n f o
Article history:
Received 29 November 2009
Accepted 1 September 2010Available online 25 October 2010
Keywords:
Solar lighting
World Bank
South Asia
15/$ - see front matter & 2010 Elsevier Ltd. A
016/j.enpol.2010.09.030
ail address: [email protected]
a b s t r a c t
Basing on our devised World Bank’s ‘Design Principles’ for effective renewable energy projects in
developing countries and an in-depth analysis of our two solar lighting projects in Bangladesh and
India, this paper explores three key obstacles that constrain poor people from obtaining solar lighting:
financial exclusion, weak governance, and passive NGO and customer participation. The low take-up
rate has a social and psychological impact. This paper recommends creating easy access to credit,
establishing a robust complaint system, and developing strategic partnership to overcome the
obstacles.
& 2010 Elsevier Ltd. All rights reserved.
1. Introduction
The World Bank has been, and will continue to be, aninfluential player in shaping renewable energy policies indeveloping countries. It has spent US$11 billion on renewableenergy since 1990 (World Bank, 2008a: 10). The rate of financingnew renewable energy and energy efficiency schemes increasesby an average of 30% per year (World Bank, 2009a: 9). 2.5 millionhomes in the developing world are now linked to solar lighting asa result (World Bank, 2008a: 8). After working on the ground fordecades, the Bank has accumulated huge experience about whatmakes renewable energy projects work and what does not.Lessons drawn from the Bank’s policies and strategies, therefore,merit great attention.
This paper focuses on the World Bank’s ‘Design Principles’ forrenewable energy projects in developing countries. The three keyelements of the Design Principles are: getting the prices, theinstitutions and the contexts right. The first principle touches oncost management: making renewable energy cheap in order towiden access. The second principle stresses good governance:strengthening institutions to ensure effective delivery of renew-able energy services in a transparent and accountable manner.The third principle underlines the appropriate choice of technol-ogy that meets local needs. However, the World Bank has neveractually used the term ‘Design Principles’; we derive it from theBank’s ten key policy documents and publications over energy(e.g. World Bank, 2009a, 2008a), infrastructure (e.g. World Bank,2008b), climate change (e.g. World Bank, 2008c), and povertyreduction (e.g. World Bank, 2000). (The criteria for selecting these
ll rights reserved.
publications for deriving the concept will be discussed in nextsection.)
These Design Principles offer a clear understanding of the keyfactors affecting effective implementation of renewable energyprojects in developing countries. We use the Design Principles asa reference point and guide to analyse two solar lighting projectsin South Asia. These are an individual solar home system inBangladesh and a solar lantern system in India. They are both off-grid, small-scale interventions for poorly served rural populations.
In this paper, we will analyse the project designs and theimplementation process of these case studies, and highlight threekey obstacles that constrain poor people from obtaining solarlighting: firstly, financial exclusion: poor people cannot afford toown or rent solar light systems because of the heavy down-payment and monthly subscription. Using universal subsidypolicies to reduce costs creates unfairness and benefits only thenon-poor. Secondly, weak governance: a lack of both robustmonitoring and efficient technical support reduces service qualityand leads to the break-down of the self-regulatory relationshipsbetween customers and service providers. Thirdly, passiveparticipation: customers and NGOs are not involved in thedecision-making process of the project designs, and they do notfeel a sense of ownership of the projects. Our study will suggestthat solar lighting’s selective take-up rate adds a new layer ofsocial divides in the communities, between those who couldafford the solar lighting and those could not.
To overcome the obstacles, we will make a number ofrecommendations. Innovative financial arrangements, such asproviding easy access to credit, are promoted to create financialinclusion. Increasing transparency about how rules are made andthe setting up of a sound complaint system to monitor servicescould achieve governance strengthening. Genuine partnership is
S. Wong / Energy Policy 40 (2012) 110–120 111
developed to engage NGOs and customers at the early stage ofproject design to promote ownership.
This paper will first outline the rationality and strategies of theWorld Bank’s ‘Design Principles’ for renewable energy projects indeveloping countries. It will then discuss the qualitative researchmethodology that guides the selection of the case studies and theresearch process. Highlighting the background, implementationprocess and outcomes of the two solar lighting projects in SouthAsia, the next section will explore the obstacles that hindereffective solar lighting interventions, such as cost management,social influences, governance and participation, and technicalsupport. It will conclude with recommendations that help over-come the obstacles.
2. Design Principles
It is estimated that 1.6 billion poor people are denied access toclean and reliable energy (World Bank, 2008a: 8). How to addresssuch a grave scale of energy poverty, by making renewable energyaffordable and attractive, has become one of the biggestchallenges to the World Bank. In its recent energy strategyconsultation paper, the Bank asks: ‘what are the ways in whichthe [energy] strategy’s focus on the poor could be enhanced?’(World Bank, 2009b: 8).
We have closely examined the World Bank’s policy documentsand publications over renewable energy and related issues, andderived the World Bank’s ‘Design Principles’ for achieving
Table 1The World Bank’s ‘Design Principles’ for effective renewable energy projects in develop
Source: Our own table, derived from World Bank publications.
Principles Rationality
(1) Cost effectiveness:
getting the prices right
� Price is the primary determinant of fuel switch
reducing the end-user prices of renewable energ
crucial (World Bank, 2009a: 1)
� Produce a level-playing field for renewable ener
existing energy market is distorted by heavy subs
non-renewable fuels (World Bank, 2000: 103)
(2) Good governance: getting
the institutions right
� Sound and credible regulatory and monitoring
structures provide clear and predictable signals t
customers and service providers, and that impro
management efficiency and quality of services (W
Bank, 2008b: 14).
(3) Appropriate technology:
getting the contexts right
� Renewable energy innovations work better if m
local conditions and priorities (World Bank, 2000
effective renewable energy projects in developing countries. Thereasons why we coin this phrase are that: the World Bank hasbeen working closely with developing countries over renewableenergy for decades, and there are some lessons we can learn fromthe Bank’s experience and engagement. The three key elements ofthe Design Principles, summarised in Table 1, aim to providepolicy-makers, NGOs and practitioners with some ideas aboutwhat factors shape effective renewable energy projects. TheDesign Principles can be interpreted as the favourable conditionswhich need to be created, before the projects are started, in orderto enhance the access to clean technology for the poor. The DesignPrinciples are not prescriptions for ‘sure-win’ interventions. Theylay stress on context specificity and allow flexibility, and theirrelevance is subject to local circumstances and the nature of theinterventions.
There is a need to stress that the World Bank does not usethe phrase ‘Design Principles’ in its documentations; we deriveit from the Bank’s ten key documents touching on issues,such as renewable energy and energy efficiency (2009a, 2008a,2007, 2006, 1999), infrastructure (2008b), climate change(2008c), and energy–poverty links (2004, 2001, 2000), publishedover the past decade. The criteria to select the literature arebased on their relevance, significance and year of publication. Theliterature shows a strong consistency of the Bank’s position-ing and strategies in renewable energy. We provide detailedsources of the literature to enhance our research validity. Weare aware that the Bank has proposed the ‘Framework for theEnergy Strategy’ in its recent energy consultation document
ing countries.
Strategies
ing, so
y is
� To reduce the market prices of renewable energy by continuous
technological improvement (World Bank, 2009a: 10)
gy since
idies on
� To improve efficiency by competition and by providing more
energy suppliers (World Bank, 2001, quoted in World Bank,
2009a: 5)
� To fill the investment gaps by attracting more investments from
the private sector and by introducing the user-pays principles
(World Bank, 2006: 20)
� To promote innovative financial arrangements to make access to
credits easier (World Bank, 2000: 5)
� To provide more and better information for fuel-switching
(World Bank, 2008a:21)
� To remove broad-based subsidies and to promote the principle
of ‘subsidising connection, not consumption’ to increase clean
energy penetration (World Bank, 2000:103)
o
ves
orld
� To decentralise decision-making process (World Bank,
2008a:10)
� To enhance transparency and accountability to foster trust and
to combat corruption (World Bank, 2009a: 18)
� To develop clearly defined commercial mechanisms between
renewable energy service providers and customers (World Bank,
2000: 3)
� To strengthen voices of consumers and communities by public
participation (World Bank, 2006: 17)
� To promote public–private partnership in investment,
monitoring and service provision (World Bank, 2008b: 13; World
Bank, 2009a, World Bank, 2009b:11)
eeting
: 105)
� To locate clean technology where renewable energy resources
are available and plentiful (World Bank, 2008a: 10)
� To tailor services to meet requirements of poor communities
and households (World Bank, 2008b: 4)
S. Wong / Energy Policy 40 (2012) 110–120112
(World Bank, 2009a: 11), which is similar to our ‘DesignPrinciples’. We both touch on similar issues, such as financialperformance and governance strengthening in improving theaccess to renewable energy. Yet, we argue that our concept of‘Design Principles’ has some advantages as it highlights theappropriate choice of technology and the importance of localcontexts and culture in shaping effective renewable energyinterventions. It also underlines the links between cost manage-ment, governance reforms and social contexts. We are also awarethat the list of the key elements in the Design Principles is notexhaustive; we underline the three most important factors thatinfluence successful renewable energy projects implementationon the ground.
2.1. Cost effectiveness: getting the prices right
Calling for ‘appropriate energy pricing policies’ (World Bank,2006: 17) or ‘getting prices right’ (World Bank, 2000: 103) hasbecome the central principle in the Bank’s clean energy policies.The Bank suggests that price is the primary determinant of fuelswitching (World Bank, 2009a: 14). While there is an urgency tolower end-user prices of renewable energy, it is equally importantto end the broad-based subsidies on non-renewable fuels. TheBank acknowledges that renewable energy technologies are ‘notcurrently financially viable’ (World Bank, 2006: vii) and ‘notalways the lowest-cost options’ in many remote areas (WorldBank, 2009a: 15), and there is a need to ‘bridge the gap[s]between service cost and households’ ability to pay’ (World Bank,2000: 3).
There is a sense of optimism that the costs of clean technologywould gradually be reduced by technological improvement andeconomies of scale (World Bank, 2006: 22). More funding fromdonors and new investments from the private sector are, there-fore, needed to ‘provide financing for promising new and cleanenergy technologies’ (World Bank, 2006: 12). The Bank alsosuggests that more energy suppliers and strong competition help‘reduce price and improve service quality’ (World Bank, 2001: 5).
The Bank also makes clear that customers are expected to makefinancial contributions to clean energy (World Bank, 2006: 17).The user-pays principle is based on two observations: first, poorpeople are already paying for their daily consumption of non-renewable energy; second, people are willing to pay for goodservices, as suggested in the Bank’s economic analysis ofsolar photovoltaic home systems in the Philippines in 2003(quoted in World Bank, 2008a: 27). To assist the poor to pay, theBank advocates ‘innovative financial arrangements’, such ascommunity saving schemes, in order to facilitate easier access tocredit (World Bank, 2000: 104). To encourage more people toswitch from non-renewable to renewable fuels, the World Bank(2006) underlines the importance of information provision. Byproviding customers with more and better information aboutthe beneficial impact of renewable energy on their livelihoods, itwould make them less risk aversive about taking the new energyoptions.
The Bank acknowledges that subsidies are still needed to makeclean energy affordable, but the direction of the subsidy policiesneeds to be changed. The Bank promotes the removal of broad-based subsidies because the policies have distorted the market,exerted heavy financial burdens on developing countries, esti-mated at US$250 million per year, and failed to target the needypeople (World Bank, 2009a: 18, 2006: 8). The Bank advocates theprinciple of ‘subsidising connection, not consumption’ whichstresses the need to ‘award subsidies to suppliers based on thenumber of new households they connect and serve’ (World Bank,2000: 103).
2.2. Good governance: getting the institutions right
The Bank defines institutional strengthening as ‘reshaping theinstitutions that determine the nature and cost of energy servicesdelivered’ (World Bank, 2000: 104). The good governance agendais a response to the poor management and performance of thenon-renewable energy sectors in developing countries, includingpoor pricing policies, inefficiency and corruption. These problems,caused by ‘an incomplete legal and regulatory reform’ (WorldBank, 2006: 11), result in financial losses and reduce ‘the ability tofinance new investment’ (World Bank, 2009a: 10). To promotebetter governance in renewable energy programmes, the Bankadvocates decentralisation policies, making the decision-makingprocess at the local level (World Bank, 2008b). It suggests thatdecentralisation helps make ‘better ways of meeting consumerneeds’ (World Bank, 2000: 102), contribute to ‘local economicdevelopment by creating employment’, and introduce ‘newcapital and innovation’ (World Bank, 2008a: 10).
Establishing sound and credible legal and regulatory frameworksis essential to provide ‘clarity and predictability to private investorsand consumers’ (World Bank, 2008b: 14). This strategy aims toenhance transparency and accountability to improve managementand to reduce corruption. Strengthening monitoring and evaluationidentifies if any progress has made. Developing commercialcontractual relationships between service providers and customersis expected to offer clearly defined roles and to clarify the rights andresponsibilities of each party (World Bank, 2000: 3). Strengtheningthe voice of consumers and communities by public participationmeans including the previously excluded groups, such as women, inthe decision-making process (World Bank, 2006: 17), and toempower them to monitor the quality of service delivery (WorldBank, 2000: 104). Public participation is also intended to ‘increaseownership, lower costs, lead to quicker implementation, enhancetransparency and increase communities’ commitment to sustaininginfrastructure investments’ (World Bank, 2008b: 20).
The promotion of public–private partnership in investment,monitoring and service provision is based on comparativeadvantages of the state, the market and community in co-managing infrastructure investments (World Bank, 2009a: 11,2008b: 20). While the involvement of the private sector is to fillthe investment gaps and to find new business opportunities, theactive roles of NGOs and communities can provide quick feedbackon service quality. The states and local governments are expectedto create an enabling environment that allows multiple players tonegotiate the institutional rules (World Bank, 2000: 104). Anotheraim of the public–private partnership is to achieve bettercoordination of policies, so that crowding out effects can beavoided (World Bank, 2008a, 2008b, 2008c: 14).
2.3. Appropriate technology: getting the contexts right
The success of renewable energy interventions, the WorldBank suggests, lies in a careful adaptation of technologicalinnovations to local conditions and priorities (World Bank,2000: 105). The choice of the locality of the technology dependson ‘where renewable energy resources are plentiful’ (World Bank,2008a: 8). The use of locally available renewable energy resourceshelps ‘reduce reliance on energy imports’ (World Bank, 2008a:10). A deep understanding of local contexts and culture is alsocrucial for raising project and technological efficiency. In order tofacilitate customised interventions, any technological changesshould be tailored to the requirements of low income householdsand communities (World Bank, 2009a: 17).
The three key elements of the ‘Design Principles’: costs,governance, and contexts, are inter-dependent. Access to clean
S. Wong / Energy Policy 40 (2012) 110–120 113
energy depends, not only on financial management, but also oninstitutional arrangements. The success of governance reformslies in careful adaptation to local circumstances. The right choiceof technology may not bring about any impact if issues ofaffordability are not addressed.
3. Case studies and methodology
Our two case studies were: an individual solar home system inChar Kajal, Bangladesh and a solar lantern project in Rajasthan,India. Both cases were both off-grid technological interventions,intended to offer clean, reliable and affordable lighting to improvethe well-being of poor rural villagers (company annual report;interview with Indian project manager, 14/03/09). We acknowl-edge the narrowness of our case studies since they shared similartechnology and were applied in the same geographical region.However, by contrasting the differences between the case studiesover cost management and institutional arrangements, our paperintends to offer some contextual understanding of how technol-ogy shapes social interactions and how people respond to thechanges. By exploring different types of obstacles that hindereffective implementation of solar lighting in our case studies,it will also help identify some research areas that need furtherin-depth empirical studies.
The investigation of these case studies was part of our 3-yearresearch project ‘Making Sustainable Technology for LivelihoodImprovement’ (2007–2010), funded separately by the BritishCouncil and the Engineering and Physical Science ResearchCouncil (EPSRC) in the UK. The general objectives of the projectwere to understand how sustainable technology made an impacton poor people’s lives. The British Council was concerned withgovernance issues—how transparency, accountability and repre-sentation brought fair distribution of the costs and benefits oftechnological changes into poor communities. EPSRC, in contrast,was more interested in engaging local scientists for technologicaldesigns. In the research consortium, our social scientist teamexamined factors facilitating, as well as hindering, effectivestakeholder participation from a full project-cycle perspective,and from the choice of technology to the application of thetechnology on the ground.
The choice of solar lighting for investigation was based on awidely accepted belief that it made a positive impact on poorpeople’s lives in developing countries, such as raising productivityfor home-based crafts and improving education opportunities(World Bank, 2008a; Kirubi et al., 2009; Cherni and Hill, 2009;Martinot et al., 2002). It also had a strong gender impact: womenfelt safer if places were lit up, and solar lighting was crucial toincreasing their physical mobility (Wong, 2009a). Literature,however, also mentioned challenges around solar lightingsystems, such as access, affordability, technical and maintenancedifficulties (Mondal, 2010; Pode, 2010). By taking a holisticperspective into understanding the obstacles that hinderedsuccessful solar lighting interventions, our project was intendedto come up with practical solutions that could maximise thepotential of solar lighting. The focus of our investigation on ruralSouth Asia was due to the concern that the region was the secondlowest in electrification after Sub-Saharan Africa, only achieving60–65% penetration (World Bank, 2009a: 13). Chaurey andKandpal (2009) suggest 68 million households in India still relyon kerosene as a fuel for domestic lighting, while the grid-basedpower supply in India was highly unreliable, and power shortagesduring peak hours ‘averaged 17%’ in the first half of 2009 (Chaureyand Kandpal, 2009). According to the World Bank’s research,‘4 out of 5 without access to electricity live in rural areas’ (WorldBank, 2006: 3). Using stand-alone solar lighting interventions,
therefore, had huge potential for livelihood improvement. Theregional specificity of our funders was our another concern. OurEPSRC grant, for instance, was directly related to the UK–Indiaresearch strategic partnership.
3.1. Qualitative approach
The two specific aims of our research were: firstly, to explorewhat factors facilitated, as well as constrained, public participationin solar lighting projects; secondly, to examine the strategies toovercome the barriers. Since our investigation was exploratory innature, rather than relying on statistical analysis, qualitativemethodology, with the intention of understanding the diverse socialmeanings (Mason, 2002), would prove useful to us in examining thecomplexity of the human–technology interactions. We also wantedour research process to be empowering—to get the voice of themarginalised social groups heard. Our study also examined howproject managers interpreted, and applied, concepts, such asparticipation, governance and cost management, in the projects.
To meet the criteria of our case study selection (i.e. off-gridsolar lighting projects in rural contexts), we contacted differentNGOs in Bangladesh and India for assistance. We were particu-larly interested in the individual home system in Char Kajalproposed by our Bangladeshi NGO partner and the solar lanternproject in Rajasthan proposed by our Indian NGO partner. Thecontrasting features of these two case studies made an interestingcomparison: firstly, the former case study was based on a fishingcommunity whereas the latter is an agricultural community. Wesuspected that solar lighting might play different roles as themain economic activities differed. Secondly, customers wereexpected to own the solar panels in the Bangladeshi case studywhile people simply rented the lanterns in the Indian case study.These contrasting ownership practices would touch on issues ofaffordability and cost management. Thirdly, the Bangladeshi casestudy adopted a universal subsidy policy which might createunfairness. Fourthly, there was a wide disparity of success ratebetween villages in renting out solar lantern systems: between45% and 75%. Fifthly, local people, including women, wererecruited as ‘entrepreneurs’ to run and maintain the solar lanternsystems in the Indian case study. This raised concerns about whatimpact these rural entrepreneurs would make on local govern-ance. Research by Bond et al. (2010) suggest that villagers in EastTimor prefer solar home systems to solar lanterns because of ‘aperception of better light quality, ability to illuminate the wholehouse, reduced risk of damage to the PV equipment and longerduration of nightly operation’ (p. 1076). We were curious ifvillagers in our Bangladeshi case study had similar perceptionsbased on their own experience.
3.2. Interview groups and questions
Each case study took three months for the preparation andinformation gathering, such as project proposals, budgets andannual reports. We spent two weeks respectively collectingempirical data (3–17 January, 2008 in Bangladesh and 3–15 April,2009 in India). To contact poor villagers for interviews, wedesigned a systematic approach to select our interviewees. In thefirst few days of our site visits, we randomly asked villagers andNGO fieldworkers what they meant by poverty and who theyconsidered as poor and very poor. Both case studies characterisedpoor people similarly: people with limited savings, limitededucation, no sons in the families, widows with small familiesto look after, the elderly, the ill and the disabled. The very poorwere villagers without savings, no education, no sons in thefamilies, widows with big families to take care of, elderly people
World Bank
Dhaka-based solar lighting company
NGOs
50 watts
4 bulbs, 1 TV, 1 mobile charger Down-payment:
5,200 taka. Monthly
subscription fees: 1,000 taka
40 watts
3 bulbs, 1 TV, 1 mobile charger Down-payment:
4,200 taka. Monthly
subscriptionfees: 800 taka
20 watts
2 bulbs, 1 TV, 1 mobile charger Down-payment:
1,000 taka. Monthly
subscriptionfees: 400 taka
Diagram 1. Governance structure of promoting individual solar home systems in
Bangladesh.
Source: Author’s own diagram, data from interviews with NGOs.
S. Wong / Energy Policy 40 (2012) 110–120114
with poor health, the severely ill and the disabled. In the fishingcommunity of Char Kajal, villagers considered households with-out fishing equipment, such as boats and fishing nets, as poor. Incontrast, community members in Rajasthan considered the land-less or those owning little or marginalised land as poor. Twointerviewees in the Bangladeshi case study mentioned that peoplewithout solar panels were also poor (interview with a sickvillager, 05/01/08; interview with the elderly, 06/01/08).
We approached our interviewees differently in Bangladesh andIndia. In the Bangladeshi case study, we divided our interview groupsinto three: the households without solar home systems, the house-holds owning solar home systems, and the professional stakeholders.We identified the first group of interviewees by villagers’ self-definition of poverty; they were: a widow with a big family to lookafter, a disabled person, an elderly person, a sick man, and a womanwith no son and six daughters. The second group of intervieweescould afford the solar panels. We asked them about their experiencein technical support and reliability, quality of service and social statusthat the ownership gave. We randomly selected a father who receivedremittances from his son (an engineer living in the capital) and abusinesswoman who sold salted fish. Amongst the professionalstakeholders, we interviewed the chief executive of the solar lightingcompany, two fieldworkers from different NGOs, a local village leader,and an official from the Ministry of Rural Development. In total, weinterviewed seven males and five females (see Appendix 1 for details).
In the Indian case study, not all poor people failed to get accessto solar lantern systems because the rental services enabled somepoor households to hire the lanterns. We divided our interviewgroups into three: households that never rented the solarlanterns, households using the services intermittently, andhouseholds which had continuously subscribed to the system.The professional stakeholders included NGO field workers, projectmanagers and rural entrepreneurs. In total, we interviewed sixmales and seven females (see Appendix 1).
We adopted semi-structured interviews as our researchmethod. Our interview questions were divided into six sections:(1) changing energy practices before, and after, the projectinterventions; (2) community sharing and conflicts over energyuse; (3) involvement of the stakeholders in the implementation ofthe projects; (4) stakeholders’ decisions over, and motivations for,the subscription; (5) obstacles constraining people from renting/owning solar lighting; (6) measures to overcome the obstacles(see Appendix 2 for detailed interview questions). Each interviewlasted for an hour. We relied on NGO fieldworkers to interpretduring the interviews for us. We recorded and transcribed theinterviews, but some interviewees did not want their conversa-tions to be recorded, so we took notes during the interviews. Inthe process of organising of our empirical data, we used our ownWorld Bank’s ‘Design Principles’ as a reference point. We madeour analysis based on themes, such as cost, management,technical support, governance, gender, and participation.
We acknowledge that our sample size of our research wassmall (25 interviewees in total) because of our tight fieldworkschedules. That said, our research was designed to explore thediverse meaning and impact of the solar lighting on poor people.The diverse viewpoints of different stakeholder groups enabled usto compare and contrast their social positioning and theirinteractions with the solar lighting.
1 Currency rate: £1¼114.9 taka (on 10/11/09).
4. Case studies
4.1. Individual solar home systems in char Kajal, south Bangladesh
The World Bank-funded project adopted a decentralisedapproach to promote the solar home systems in rural villages in
Bangladesh. A Dhaka-based solar lighting company was set up bythe Central Government to oversee the individual solar homedevelopment in Bangladesh (see Diagram 1). Three models: 50, 40and 20 W, were designed for different customer groups. While thedown-payment of the 50 W panel cost 5200 taka1 and supportedfour bulbs, one TV and one mobile phone charger; the down-payment of the 20 W panel was 1000 taka which supported twobulbs and one mobile phone charger. In addition, the owners ofthe first option needed to pay 1000 taka for monthly subscriptionfees while the owners paid 400 taka for the third option.
The chief executive of the company explained that the donorwas in favour of an individualised, rather than a communal, solarsystem because this would ‘create clearer sense of ownership [forindividual households] to look after their solar panels’ and ‘avoidfree-riding behaviour’ (interview, 16/01/08). He revealed that thesubscription fees did not reflect the full financial costs of thepanels because the World Bank had subsidised US$100 for eachinstallation in order to make the systems less costly. He suggestedthat imposing charging, rather than offering the solar panels free-of-charge to poor households, was based on concerns that poorpeople might sell the panels when they were desperate, and thatwould violate the principle of using solar lighting to addressenergy poverty (interview, 16/01/08).
The Dhaka-based solar lighting company sub-contracted theprojects to 15 NGOs in the country which were responsible forselling the panels to individual households. The contracts statedthat the NGOs would provide ten years’ maintenance servicesupport for the PV panels, five years’ for batteries and three years’for electrical circuits. They were also responsible for collectingmonthly subscription fees from each household. In order toincrease the penetration of the solar systems in rural villages, thesolar lighting company offered NGOs financial rewards of US$80for each successful installation. The chief executive also arrangeda number of NGOs to work in the same areas. He explained thatthe company needed to get the NGO–household ratio right inorder to provide proper services. He argued that the arrangement
STAKE HOLDERS ROLES
* to initiate and fund the project
* To build battery-recharging facilities * To select entrepreneurs and provide them with training
* To find customers to use solar lanterns * To collect fees * To provide basic maintenance services
* To pay for services * To monitor service quality * To make complaints to entrepreneurs & NGOs if dissatisfied with services
Delhi-based research organisation
Entrepreneurs
Customers
Rajasthan-basedNGO
Solar lantern manufacturers
Diagram 2. Governance structures of the solar lantern projects in my case studies.
Source: Author’s own diagram, data from interviews with NGO.
S. Wong / Energy Policy 40 (2012) 110–120 115
could create a sense of ‘positive competition’, and that wouldempower customers by making the NGOs more responsive totheir needs (interview, 16/01/08). In order to increase the trans-parency, each NGO was required to submit the numbers ofinstallations to the company every 6 months. The company wouldthen submit their annual report to the World Bank for monitoring.
Our case study, Char Kajal, was located in South Bangladesh. Ithad a population of 35,000 populations, comprising 5300 house-holds. Six NGOs were working in the area. They set up booths instreet corners to display the solar panels and to explain how thesystems worked. They also paid visits to individual households topromote the systems. A NGO fieldworker told us that, owing to theirlimited manpower, they collected household information, such ashouse and family size and availability of fishing equipment, byobservations and word-of-mouth before they decided which house-holds were worth visiting (interview, 10/01/08).
As the solar lighting project progressed, the number forpotential customers who had not installed the systems was fewer.Since the NGOs were not allowed to reduce the subscription fees toattract new customers, they needed to provide better services. Thedownside of the competition, as one NGO fieldwork pointed out,was that: ‘two fieldworkers of different NGOs nearly end up in afight just for one customer’ (interview, 11/01/08). Totally 1100households in Char Kajal had bought the solar systems in 2008.This covered roughly one-fifth of the total households in the area.The low take-up rate could be explained by heavy cost of thedown-payment and high monthly subscription fees.
2 Currency rate: £1¼77.1 rupees (on 10/11/09).
4.2. Solar lantern systems in Rajasthan, west India
This solar lantern project was initiated and funded by a Delhi-based energy research organisation. The project manager explainedthat solar lantern systems were cheaper to install and to operatethan the solar home systems (interview, 14/04/09). The projecttargeted nine rural communities in Rajasthan and adopted adecentralised approach: the energy research organisation con-tracted out the projects to a Rajasthan-based NGO. Apart fromoffering 50 free solar lanterns and building solar battery rechargingfacilities for these communities, the NGO was also responsible forrecruiting one local villager from each community as an ‘entre-preneur’ to run the daily operation of the solar lantern system. Theidea of promoting entrepreneurship through solar lighting inter-ventions was to create job opportunities, to make use of the
entrepreneurs’ local knowledge and social relationships to enhancethe efficiency of the projects, and to advocate women empower-ment by recruiting female entrepreneurs (interview with projectmanager, 14/04/09). The entrepreneurs would provide services,such as running the battery-recharging shops and offeringtechnical maintenance, for their local users. If there were anytechnical faults, the entrepreneurs had the responsibility tonegotiate with the solar lantern manufacturers (Diagram 2).
The NGO chose the nine communities based on their pastworking experiences. To find female entrepreneurs, the NGOprivately approached the women who had participated in itssocial groups and asked if they were interested in becoming‘entrepreneurs’. Only two women were eventually willing toaccept the post. The manager of the NGO explained that manyobstacles held women back from taking up public roles, such asstrong resistance from their husbands, heavy family duties, andstrong negative stereotypes that women were not good managers(interview, 06/04/09).
The project funder and the NGO agreed to fix two rupees2 aday as the daily rental fees because, according to their estimation,households on average were paying the same amount of moneyfor their daily consumption of kerosene for lighting (interviewwith NGO, 06/04/09). However, in order to reduce transactioncosts and to provide entrepreneurs with financial certainty, onlythose households who could pay a lump-sum 60 rupees for amonthly subscription could enjoy the rate at two rupees. Ifvillagers opted for daily rental, they had to pay 5 rupees a day. Theentrepreneurs were allowed to expand the market by servingcustomers living in other communities, and could charge the‘outsiders’ flexibly. The NGO estimated that, if all 50 lanterns weresuccessfully rented out, each entrepreneur would receive 3000rupees income a month. However, each entrepreneur wasrequested to set up a maintenance fund by reserving up to 1000rupees of their income (or one-third of their total income) to bepaid into a bank at the end of each month. In case of systemfailure, the entrepreneurs would draw on the maintenance fundto finance repair or replacement. If the maintenance costs becametoo much, the NGO would fill the financial gaps. The entrepre-neurs were not expected to contribute any costs upfront or to bearany of the running costs of the project. This policy was intended topersuade the entrepreneurs to focus on the service delivery in
S. Wong / Energy Policy 40 (2012) 110–120116
the knowledge that the financial stability was guaranteed. Theentrepreneurs were expected to report ‘regularly’ to the NGOabout the operation. Before the scheme was implemented, theNGO called for a community meeting and explained the pro-gramme and the charges.
The NGO told us that on average 55% of households subscribedto the scheme. We deliberately selected two communities:Banganga and Chhota Kantrana, for comparison because of theircontrasting features and results. The former village was managedby a female entrepreneur while the latter by a male. The formersystem was regarded by the NGO as a ‘failure’ because of a low 40%subscription. The latter, in contrast, had a 75% subscription rate.The NGO suggested that the former received an older-generation ofthe solar lanterns which had suffered from many technicalproblems, thus causing unreliability and reducing the take-up rate.
5. Obstacles
In this section, we use our devised World Bank’s ‘DesignPrinciples’ as the framework and analyse the two solar lightingcase studies from the perspective of cost effectiveness, govern-ance management, and technology and contexts. We also examinethe impact of exclusion from the projects on poor people’s lives.
5.1. High costs reduce affordability
Much research has been conducted on issues of affordabilityand has suggested that the high cost of solar lighting systems wasone of the key issues prohibiting poor people from obtaining solarlighting (for latest discussion, see Pode, 2010; Mala et al., 2009).Our research has made similar findings on costs and affordability,but by comparing and contrasting the cost management andpayment schemes of our two case studies, our discussion willenrich the existing debate.
Both case studies have demonstrated the project designers’awareness of cost management and attempts to adjust prices inorder to enhance poor people’s access to solar lighting. In theBangladeshi case study, three models of solar home systems weredesigned, with different capacities and costs, to meet needs ofdifferent customer groups. A universal subsidy policy wasimplemented to reduce the actual costs of the systems. In theIndian case study, solar lanterns and solar battery chargingfacilities were given to the communities free-of-charge. The rentalfees were set at the competitive price level of the non-renewablefuels. Furthermore, both case studies had defined end-users ascustomers and encouraged them to make financial contributionsto the solar lighting systems. Their motivations for adopting theuser-pays principles were similar: to generate a sense of owner-ship, to promote an efficient use of facilities, and to fill thefinancial gaps in the operation and maintenance costs (interviewwith the chief executive, 16/01/08; interview with Indian NGO,06/04/09). Another similarity was that strong financial incentiveswere given to the ‘middlemen’ with the intention of raisingproject efficiency. The NGOs in the individual solar home systemswere awarded US$80 for each successful installation3 whereas theentrepreneurs were allowed to make profits from service delivery.Some attempts were also made for providing information for theircustomers to enhance their knowledge about the systems, such ascalling for community meetings, making home visits, setting upstreet booths and distributing leaflets.
3 Providing US$80 for each successful installation might sound a lot, but the
NGOs relied on these financial awards to pay salaries and to finance their own
transport and other logistic costs.
There was, however, a contrasting rate of subscription in thecase studies: while the individual solar home project inBangladesh only had a 20% subscription rate, the solar lanternsystem in India achieved 55% on average. The statistics suggestedthat the majority of the villagers in Char Kajal were not able toobtain the solar lighting for a variety of reasons.
Owning the solar home systems in the Bangladeshi case studymarked the biggest policy difference of the solar lantern project inIndia which adopted the strategy of renting. Heavy down-payment and high subscription fees for the solar home systemsseemed to be the reason for the high level of unaffordabilityamongst the villagers. However, the high costs of the solar homesystems could easily be reduced if the World Bank had offeredmore subsidies. The crux of the problem actually lay in the factthat operation and maintenance costs of the solar home systemswere very high. The World Bank’s US$100 subsidy for eachinstallation, according to the chief executive of the Dhaka-basedsolar lighting company, could only cover 65–70% of the total costs(interview, 16/01/08). As a result, customer contributions weredesperately needed to fill the financial gaps.
Raising subsidies might help the customers pay less, but theBangladeshi case study has suggested that this was not a desirablesolution. The universal subsidy policy of the individual solar homesystems was a heavy financial burden. In Char Kajal alone, theWorld Bank had provided more than US$200,000 by way ofsubsidy for just 1100 households.4 The subsidy policy has alsocreated unfairness since it had only encouraged the better-offcustomers to use the clean energy. 80% of the community in thecase study remained under-served.
Offering NGOs US$80 for each successful installation providedNGO fieldworkers strong motivation for promoting the solar lightingin the Bangladeshi case study. It also gave the organisations a senseof financial security. The financial incentives could, however,become pervasive incentives. The NGOs only targeted the less-poor customers who could afford the down-payment and pay for themonthly subscription fees. Our interviews with the NGO fieldworkworkers have shown that they had put a lot of resources and effortsinto targeting ‘affordable’ customers. They were, however, lessinterested in helping the poor gain access to solar lighting. Beingasked why mirco-credit systems had not been introduced to helpthe poor to pay for the down-payment, one NGO fieldworkerresponded that he had not been told to promote micro-credit(interview, 11/01/08). Another NGO fieldworker suggested thatsetting up a proper micro-credit system was time-consuming andhis NGO lacked the manpower to achieve it (interview, 10/01/08).
5.2. Social factors affect access
Discussion about the role of social factors in affecting the accessto solar lighting is not new (see Wong, 2009b; Jacobson, 2007), butour findings, highlighting the linkages between affordability andsocial relations as well as villagers’ perceived convenience ofenergy-saving, touch on different perspectives of the debate.
Although the solar lantern systems in India had a highersubscription rate than the case study in Bangladesh, the Delhi-basedresearch organisation was not satisfied with the overall subscriptionrate. The manager suggested that, given such a high investment, hisorganisation expected at least 75% subscription rate in the targetedcommunities (interview, 14/04/09). In reality, however, on average45% of the solar lanterns on average were not successfully rented out
4 This was based on the fact that the World Bank had subsidised the
installation of 1100 sets of solar home systems in Char Kajal, and each installation
cost the World Bank at least US$180 (i.e. US$100 subsidy for the manufacturing
costs of each solar panel, plus US$80 for NGOs for each successful installation).
S. Wong / Energy Policy 40 (2012) 110–120 117
in the communities. Even for the best-performing village, ChhotaKantrana, recorded a 25% ‘idle’ rate. Our research suggests that socialfactors have played a role in shaping people’s willingness and abilityto pay for solar lighting.
One of the issues in the solar lantern project was the design ofthe monthly subscription policy. Villagers in Banganga and ChhotaKantrana generally agreed, in interviews, that setting the daily rentalfees at two rupees was reasonable, and some of them were happy topay. Yet, they needed to become monthly subscribers and paid 60rupees in a lump-sum in order to enjoy the two-rupee-a-daytreatment. The NGO explained that using this strategy to attractmonthly subscription was intended to reduce the transaction costsand to provide the entrepreneurs with financial stability (interview,11/04/09). This policy had, however, become an obstacle. Villagers ininterviews explained that farming was a seasonable business, andtheir incomes fluctuated over a year. They found it difficult to pay forthe monthly subscription fees particularly after sowing and beforeharvesting. For the poor villagers living at a subsistence level, theirlimited savings prohibited them from paying such a lump-sum. Evenfor the less-so-poor villagers, they were reluctant to pay because, asone interviewer suggested, ‘you never know what happenstomorrow. My husband may be sick, or we need money foremergency’ (interview with a farmer, 12/04/09). Life’s uncertaintiesand many other social priorities had made the monthly subscriptionpolicy less popular.
Daily rental should be an attractive alternative to monthlysubscription as it gave a high level of flexibility for villagers todecide when they wanted to rent the solar lighting. Yet, chargingfive rupees, rather than two, had caused dissatisfaction. Under thispayment scheme, solar lighting became 2.5 times more expensivethan kerosene. Some villagers also expressed scepticism aboutbeing ripped off by their entrepreneurs. An interviewee suggestedthat the solar lanterns were given free-of-charge from the donors,and they were ‘communal property’ and the benefits should beshared with everyone. She questioned why the entrepreneurs hadsuch a right to impose the ‘over-charge’ (interview with 12/04/09).Seeing the entrepreneurs use the ‘communal property’ to serveand charge people from other villages also fuelled the suspicion.(We will explain later that a lack of transparency in the process ofrecruiting entrepreneurs could be blamed for undermining thelegitimacy of the entrepreneurs).
The attractiveness of the solar lanterns was also shaped bypeople’s perceived convenience of energy-sharing. It is beyonddoubt that solar lanterns provide better quality of lighting thankerosene. Our research, however, shows that kerosene gavevillagers a stronger sense of control and more flexibility insharing. Solar lanterns provided four to five hours lighting for theusers. Once the households paid the rents, no matter whetherthey used the lanterns or not, they had already paid for theservice. They could not save the solar energy for future use.Kerosene, in contrast, is kept in liquid form and only used asrequired. Our interviewees suggested that they could controlwhen, and how much, kerosene was used. Unlike solar lighting,the left-over kerosene could be kept for future use (interviewswith the widow, 10/4/09). The liquid form of kerosene also madecommunity sharing relatively easy. Although solar lanterns werealso mobile, and the users could carry them around the villages,solar lanterns, unlike kerosene, did not help two households livingsome distance apart to get the lighting at the same time.
5.3. Weak governance
Both of the energy programmes in our case studies adopted adecentralised approach to delivering services to local populations.Instead of providing services themselves, the Dhaka-based solar
lighting company and the Delhi-based research organisationcontracted out the implementation of the solar lighting systemsto the NGOs. To ensure the decentralisation worked, both projectmanagers stressed the importance of establishing robust self-regulatory monitoring relationships between customers andservice providers. To achieve this, the roles of the players wereclearly defined. As the chief executive of the solar lightingcompany suggested, customers would understand their ownrights and make sure that they got the best from the serviceproviders (interview, 16/01/08). Rules over payment and serviceprovision were also laid down systematically to avoid disputes.
The self-regulatory mechanisms, however, did not worksmoothly in the field. Our research suggests that there was adeficiency of robust monitoring between project leaders, NGOsand entrepreneurs. Villagers in Banganga complained in inter-views that their female entrepreneurs kept changing, or designingnew, rules. For instance, one imposed a new rule that no refundwas allowed if the users did not report any technical faults within2 h after collecting the solar lanterns. The solar lantern usersfound the new rule unreasonable because they often did notcheck the batteries until they used the lanterns in the evening(interview with a farmer, 12/04/09).
The NGO explained that the entrepreneurs needed flexibility tochange rules to meet local circumstances, but our study showsthat a lack of a sound complaint system had disempoweredcustomers. In principle, dissatisfied customers could file com-plaints to the NGO. However, the long absence of the NGOfieldworkers meant that customers could either bear with theentrepreneurs or terminate the solar lighting contracts and revertto kerosene or other non-renewable fuels. The NGO explained thatthey served not only these nine solar lighting-related commu-nities, but they also had many other commitments. They admittedthat such a heavy workload had ‘compromised their monitoringrole’ (interview, 06/04/09). This explained why they were notsurprised when we revealed that a female entrepreneur had failedto set up the maintenance fund, even though the project had beenoperating for a long time.
Lack of a good complaint system in Banganga also causedproblems in our Bangladeshi case study. As mentioned before, oneof the key motivations for the solar lighting company to assign anumber of NGOs to work in the same region was to usecompetition to maintain good quality services because customerscould choose between different service providers. In reality,however, villagers told us that once they signed the contract,they could not change the service provider; otherwise, they wouldlose most of the down-payment they had already paid (interviewwith businesswoman, 07/01/08). The company was based inDhaka, far away from this remote village. The affected customersdid not know how they could make complaints to the company iftheir NGOs did not do the job properly. Neither did they knowwhich persons in the company they could contact.
The weak governance was also demonstrated by inadequateefficient technical support. When the solar lantern projectsstarted in Rajasthan, the NGO arranged meetings for theentrepreneurs and the solar lantern manufacturers. The NGOalso expected that they could sort out the technical problemswithout its interference. Chhota Kantrana was using thesecond-generation solar lantern system, and it provided reliableservices. In contrast, Banganga’s first-generation system wasproblematic. When the female entrepreneur first identified thetechnical faults, she quickly reported the problems to themanufacturers. To her despair, the technician did not come till amonth later. On some occasions, the technicians did not have theright equipment or accessories to fix the system, so the faultylanterns were left idle for another month. The NGO explained thatthe technician lived some distance away, and he only came round
S. Wong / Energy Policy 40 (2012) 110–120118
when he received a number of requests in the same area. Withoutan efficient technical support, our case study demonstrates thatthe high break-down rate reduced the demand for the services.When profits dropped, the female entrepreneur was disincenti-vised, and the systems further deteriorated.
5.4. Limited customer and NGO participation
The World Bank advocates ‘strengthening the voice ofconsumers’ (World Bank, 2006: 17) by ‘involving communitiesin service delivery’ (World Bank, 2000: 2). Nevertheless, there wasa lack of genuine participation by NGOs and customers in bothcase studies, from project designs and implementation toevaluation. The Dhaka-based company and the Delhi-basedresearch organisation were the key decision-makers: they devisedthe intervention strategies, set the goals, chose the technology,and decided where to intervene, who would benefit and whichNGOs to work with. In the whole decision-making process, theaffected communities were rarely consulted, and the NGOsinvolved did not participate at the early stage of the projectplanning. The chief executive defended this by stating that theproject was largely ‘technical and financial’ in nature at the initialstage, and their involvement would be premature (interview,16/01/08). Both project planners gave us an impression that theywere doing something good for the communities, and they knewhow the goals could be achieved.
The NGOs started their involvement at the stage of imple-mentation in both case studies. The NGO in Rajasthan wasassigned to build the battery-recharging facilities, to recruit andprovide training for the entrepreneurs, and to build partnershipsfor the entrepreneurs and the lantern manufacturers. The NGOs inChar Kajal were busy marketing and installing the solar homesystems for the subscribers. The project manager of the Delhi-based research organisation stressed the significance of theinvolvement of local NGOs, acknowledging their local knowledge,networks and personal relationships with the communities wouldenhance project efficiency and help reach the customers (inter-view, 06/04/09). Our study suggests that the project planners tookan instrumental perspective concerning the participation of theNGOs, considering them as merely executors. They made use ofthe legitimacy of the NGOs to get things done and to get thetargets met, rather than build strategic partnership. The problemwas that the NGOs internalised their role as rule-followers. Theybecame less critical of the funders and the project managers aswell as the outcomes of the solar lighting projects. This wasclearly an issue in the Bangladeshi case study: despite the lowtake-up rate of the solar home systems, the NGOs were not keento look for any innovative measure to widen access.
Active customer participation was encouraged at the laterstage of the project. As the chief executive suggested, customerspaid for the solar lighting services, and they would play a key rolein monitoring the performances of the service providers (inter-view, 17/01/08). The idea of ‘They-Pay-Therefore-They-Partici-pate’ was, however, problematic. According to this principle, only20% of the households in the Bangladeshi case study were‘legitimate participants’ while the right of participation by therest of the population in Char Kajal was denied simply becausethey were poor.
We need to stress that public participation does not guaranteebetter outcomes from the projects, but denying poor people’sinvolvement could undermine the effectiveness of the interventions.In the solar lantern project, the NGO engineered the recruitmentprocess of the entrepreneurs. It stressed that they would only choosethe entrepreneurs who had good social relationships in theircommunities (interview, 08/04/09). Without proper community
consultation, however, villagers were not involved in the rule-making process. They did not know how their entrepreneurs wereselected (interview with the disabled, 10/04/09) and why they hadno right to choose their entrepreneurs (interview with teacher,11/04/09). The lack of transparency led to the undermining of thelegitimacy and authority of the entrepreneurs. When the femaleentrepreneur in Banganga were seen as ‘incompetent’ because of theserious technical faults in the equipment supplied, the pressure toask her to step down was mounted.
5.5. Social and psychological impact
The above-mentioned obstacles, over cost management andgovernance designs, have denied the needy the benefits of solarlighting. This exclusion has both social and psychological impacton the communities. Our study suggests that a new layer of socialdivide, between those who could afford the solar lighting andthose could not, was developed in the communities. Solar lightinghas become a symbol of wealth and status in the villages. This wasparticularly the case in the solar home systems in Bangladeshwhere the solar panels, being installed on the roof top, werehighly visible. The solar lighting divide has mirrored the existingasset-based inequalities in the communities where the richcontrol most of the land and the fishing equipment while thepoor lack the basics. Our finding echoes with the study byJacobson (2007) in rural Kenya that solar electrification con-tributes to the ‘deepening of middle class formation’ (p. 158)because solar home systems have facilitated reliable batteryrecharging of mobile phones, and these significantly improved themeans of communication by the rural middle class.
In interviews, women often expressed their guilt of not beingable to provide solar lighting for their children. A primary teacherin Banganga was worried for her students who were deprivedof the opportunities that the lighting could bring (interview,11/04/09). The solar lighting exclusion appears to widen theexisting development gaps within communities in the long-run:while the health conditions of the not-so-poor are improving andeducation opportunities expanding, thanks to the non-universalaccess to solar lighting, the poor would lag behind.
Our research also discovers that villagers had been developinga ‘hierarchy of comparative inferiority’, with reference to differentsources of lighting. When villagers were asked about their feelingsabout not being able to use the solar lighting, their replies wereusually a mixture of guilt and blessing. On the one hand, they feltbad about not being able to enjoy the ‘modern’ lighting. On theother hand, however, they comforted themselves by suggestingthat they were not as bad as those who relied on clay lamps orlived in the darkness (interview with the elderly, 11/04/09;interview with the mother with no son, 11/04/09). According toour interviews, the hierarchy was as follows: at the bottom, thepoorest could not afford anything and lived in darkness. A layerabove was formed by those who relied on clay lamps. The nextlayer was formed by those with access to kerosene, and then gotthe top, those with solar lighting. The ‘best’ form of lighting, manyvillagers would say, was provided by the national grids.
6. Lessons and recommendations
After gaining a deep understanding of the mistakes made, andchallenges encountered, in our case studies, plus the inspirationsof the World Bank’s Design Principles, we come up with threelessons that could make solar lighting projects more effective,inclusive and pro-poor; they are: poverty-sensitive cost manage-ment, better governance, and robust technical support.
S. Wong / Energy Policy 40 (2012) 110–120 119
6.1. Poverty-sensitive cost management
We are aware that research by Rao et al. (2009) and Srinivasan(2007) suggests that, with the proper support of micro-finance,poor people are willing and able to own the solar lighting systems.Our research has, however, shown that, without the support ofany micro-credit systems and where poor people were expectedto pay for the service by their own means, it would be better tohelp poor people rent, rather than own, the solar lighting systemssince renting, as indicated in our case studies, exerted lessfinancial pressure on poor households, and simultaneously,fostered a sense of ownership that was essential for co-managingthe technology. Our study also suggests that daily rental, ratherthan monthly subscription, may be a better fit for poor people’slivelihood strategies. A daily rental scheme may involve highertransaction costs, but it gives poor people flexibility about whenthey want to hire the solar lighting. This way also avoids thempaying a heavy lump-sum by way of deposit or purchase. Toincentivise fuel-switching from the non-renewable to renewablesources, the charges for the daily rental should be set at acompetitive price level with regard to the non-renewable fuels.
Our research has also demonstrated that providing non-discriminatory subsidies to help people install solar lighting isnot a desirable policy option. As the Bangladeshi case study hasshown, the policy would only exert a heavy financial burden ondonors, without providing any benefits to the poor. Offeringspecial financial assistance to poor people by targeting subsidiesis possible, but we are aware of the high transaction costsinvolved and the possibility of rent-seeking (Bjorvatn andConiglio, 2006). We support the World Bank’s suggestion ofmaking credits more accessible to the poor (World Bank, 2000: 5).Jacobson’s research in Kenya (2007) suggests that ‘solar PVsystem can often be viewed more as a middle class consumergood than as a productive investment’ (p. 157). This perspectivesuggests that the ‘consumptive, rather than productive’ nature ofsolar lighting is an obstacle to achieving successful micro-creditsystems. We, however, argue that: instead of setting up themicro-credit systems after the introduction of solar lighting, weshould turn things the way round. When poor people’s financialconditions are improved, thanks to the introduction of the micro-credit schemes, they will be more willing to invest in cleanenergy. Micro-finance institutions could also consider providingfunds under a special window specifically for small solarapplications. To encourage more people to use micro-credits topay for solar lighting, the provision of better information aboutthe advantages of solar lighting is crucial. By promoting aninclusive micro-finance, we believe this could generate a newmomentum for solar lighting projects in poor countries.
6.2. Better governance
Another lesson we have learnt from the case studies is thatdecentralising solar lighting projects requires more, rather thanless, time and resources from project leaders (cf. Havet et al.,2009). It is because stakeholders, such as NGOs and customers,usually lack the opportunity, capacity or confidence to makedecisions or changes by themselves. Effective decentralisedpolicies need empowerment, and empowerment, we argue, comesfrom genuine participation. Defining solar lighting end-users aspurely customers is problematic because they could only exercisepower at the very late stage of the projects. To promote effectiveinterventions, we need to empower the end-users by giving therights back to the affected communities and by getting theminvolved from the start of the projects. While the communitiessend their representatives to attend the project meetings and to
share their experiences and their concerns, the project leaderscould create a level-playing field for equal participation. Oursuggestion about active local participation in solar lightingresonates with the lessons and recommendations of the reportby the International Energy Agency (2008) over its PhotovoltaicServices for Developing Countries that: ‘ensure thorough con-sultation with end-users and implementing agencies duringplanning and implementation’ (p. 28). The study of Adkin et al.(2010) also gives empirical evidence that villagers and vendors inMalawi actively participating in the negotiation process hasbrought the success of the solar lighting promotion into theMillennium Village Project.
Our study has also shown that passive participation of theNGOs has failed to offer the policy makers insights into makingstrategies corresponding to local contexts. We propose thatdeveloping strategic partnerships between policy makers, NGOsand local people is crucial to building ‘an active voice of civilsociety’ (World Bank, 2008b: 21). NGOs are more than executors;they bridge the project managers and local communities. Theyshould take up a pro-active role in scrutinising the process of thesolar lighting projects, listening to local communities, andreflecting their needs to the policy makers.
6.3. Train local technicians to fill the support gap
The case study in Banganga has shown that the poor per-formance of the solar lantern system was affected by inefficienttechnical support, but the real problem did not lie in poor internalcommunications between the entrepreneurs and the manufac-turers. The entrepreneur reported the faults swiftly to thetechnicians once she was notified about the problems. Thetechnician wanted to help, but his company was located faraway from the affected community. Neither he nor the companycould offer immediate technical support. This ‘support gap’, weargue, led to the deterioration of the technology and reduced theconfidence of the customers as well as the profitability of thebusiness. To build a sound support system, we propose using localtechnicians. If this is not feasible, providing updated training tothe entrepreneurs would mean they could offer at least sometemporary fix and help avoid further break-down of the systems.Chaurey and Kandpal (2009) make similar comments to supporttheir central charging station model in India (p. 4917).
7. Conclusions
The combination of our devised World Bank’s ‘DesignPrinciples’ for effective renewable energy projects in developingcountries and the analysis of the our two off-grid, solar lightingcase studies in South Asia have exposed some financial, institu-tional and contextual obstacles mitigating against successful solarlighting interventions in poor countries. By comparing andcontrasting the individual solar home systems in Bangladeshand the solar lantern projects in India, our paper has indicatedthat inappropriate cost management, such as the heavy down-payment, paying monthly subscription charges in a lump-sumand the non-discriminatory subsidy policies, might have createdfinancial disincentives that have constrained poor people fromobtaining solar lighting. It has also suggested that weak govern-ance arrangements, resulting from both a lack of robust monitor-ing, efficient technical support and sound complaint systems,would easily lead to a gradual break-down of the self-regulatoryregulatory mechanisms between customers and service providers.The instrumental participation of the NGOs and the passiveinvolvement of the poor people in the decision-making process,shown in our cases, would also hinder fostering a sense of
S. Wong / Energy Policy 40 (2012) 110–120120
ownership and empowerment about the solar lighting projectswithin communities.
In addressing these obstacles, we need more in-depth andempirical research into how to increase the ownership of solarlighting by making access to credit easier, how to improve gover-nance by making the decisions more transparent and participatory,and how to achieve genuine strategic partnerships between donors,NGOs and communities. Solar lighting innovations have becomemore technically advanced and now provide more and betterservices to customers. Without proper strategies and strongdetermination to tackle the barriers, the energy divide, betweenthe rich and the poor, in access to clean energy, would become widerin poor communities. Our ultimate goal, to end poverty byrenewable energy interventions, would then become even harder.
Appendix A. Supporting information
Supplementary data associated with this article can be foundin the online version at doi:10.1016/j.enpol.2010.09.030.
References
Adkins, E., Eapen, S., Kaluwile, F., Nair, G., Modi, V., 2010. Off-grid energy servicesfor the poor: introducing LED lighting in the Millennium Villages Project inMalawi. Energy Policy 38, 1087–1097.
Bjorvatn, K., Coniglio, N., 2006. Policy design and rent seeking: targeted versusbroad based intervention. Review of Development Economics 10 (4), 577–585.
Bond, M., Aye, L., Fuller, R., 2010. Solar lanterns or solar home lighting systems—
community preferences in East Timor. Renewable Energy 35, 1076–1082.Chaurey, A., Kandpal, T., 2009. Solar lanterns for domestic lighting in India:
viability of central charging station model. Energy Policy 37, 4910–4918.Cherni, J., Hill, Y., 2009. Energy and policy providing for sustainable rural
livelihoods in remote locations—the case of Cuba. Geoforum 40, 645–654.Havet, I., Chowdhury, S., Takada, M., Cantano, A., 2009. Energy in national
decentralization policies, United Nations Development Programs.International Energy Agency, 2008. Renewable energy services for developing
countries. In support of the millennium development goals: recommendedpractice & key lessons. Available at: /http://www.iea-pvps.org/products/download/REEEP%20RESDC%20%20MDGs%20v9%202.pdfS; (accessed on 2March 2010).
Jacobson, A., 2007. Connective power: solar electrification and social change inKenya. World Development 35 (1), 144–162.
Kirubi, C., Jacobson, A., Kammen, D., Mills, A., 2009. Community-based electricmicro-grids can contribute to rural development: evidence from Kenya. WorldDevelopment 37 (7), 1208–1221.
Mala, K., Schlapfer, A., Pryor, T., 2009. Better or worse? The role of solarphotovoltaic (PV) systems in sustainable development: case studies of remoteatoll communities in Kiribati. Renewable Energy 34, 358–361.
Martinot, E., Chaurey, A., Lew, D., 2002. Renewable energy markets in developingcountries. Annual Review of Energy Environment 27, 309–348.
Mason, J., 2002. Qualitative Researching, 2nd ed. Sage, London.Mondal, A.H., 2010. Economic viability of solar home systems: case study of
Bangladesh. Renewable Energy 35, 1125–1129.Pode, R., 2010. Solution to enhance the acceptability of solar-powered LED lighting
technology. Renewable and Sustainable Energy Reviews 14, 1096–1103.Rao, S., Miller, J., Wang, Y., Byrne, J., 2009. Energy-microfinance intervention for
below poverty line households in India. Energy Policy 37, 1694–1712.Srinivasan, S., 2007. World review of entrepreneurship, management and
sustainable development. Microfinance for Renewable Energy 3 (1), 79–89.Wong, S., 2009a. Climate change and sustainable technology: re-linking nature,
governance and gender. Gender and Development 17 (1), 95–108.Wong, S., 2009b. Linking climate change and sustainable energy technologies to
develop a ‘people-centred’ intervention framework. In: Filho, W. (Ed.),Interdisciplinary Aspects of Climate Change. Peter Lang Scientific Publishers,New York, pp. 449–468 (Chapter 22).
World Bank, 2009a. Energy strategy approach paper. Sustainable developmentnetwork, Washington, DC.
World Bank, 2009b. Energy strategy consultancy plan. Sustainable developmentnetwork, Washington, DC.
World Bank, 2008a. Issues note of the REToolkit. REToolkit: a resource forrenewable energy development, Washington, DC.
World Bank, 2008b. Sustainable infrastructure action plan. FY2009-2011,Washington, DC.
World Bank, 2008c. Development and climate change. A strategic framework forthe World Bank Group. Technical report, Washington, DC.
World Bank, 2007. Catalyzing private investment for a low-carbon economy—
World Bank Group progress on renewable energy and energy efficiency infiscal 2007, Washington, DC.
World Bank, 2006. Clean energy and development: towards an investmentframework, Washington, DC.
World Bank, 2004. World development report 2004: making services wok for poorpeople, Washington, DC.
World Bank, 2001. The World Bank Group’s energy program: poverty reduction,sustainability and selectivity, Washington, DC.
World Bank, 2000. Energy services for the World’s poor. Energy and DevelopmentReport 2000, Washington, DC.
World Bank, 1999. Fuel for thought: an environmental strategy for the energysector, Washington, DC.
