Solar Energy

Published on June 2016 | Categories: Documents | Downloads: 37 | Comments: 0 | Views: 7313
of 62
Download PDF   Embed   Report



Solar Energy and Rural Development:
Constraints and Insights from the Developing World

April 30, 2004 Anta Gueye Joshua McNary Jungo Okai

2 OUTLINE Foreword………………………………………………………………………………….3 Acknowledgement………………………………………………………………………...3 Executive Summary…………………………………………………………………….…4 Abbreviations……………………………………………………………………………..5 CHAPTER 1 INTRODUCTION Introduction…………… ………………………………………………………...6 Argument……….. ………………………………………………………………..7 Critical Questions………………………………………………………………...7 Methodology……………………………………………………………………...8 CHAPTER 2 FINANCING MECHANISMS Framework of Financing Mechanisms for PV Dissemination.……………...11 Analysis of Financing Mechanisms for PV Dissemination…………………..15 Key Financing Findings from the Kenyan Site Visit…………………………17 CHAPTER 3 CAPACITY BUILDING Capacity Strengthening for PV Dissemination in Developing Countries…...22 Analysis of Capacity Building………………………………………….………22 Notable Best Practices in Capacity Building……………………………….…27 Key Capacity Strengthening Findings from the Kenyan Example …………31 CHAPTER 4 QUALITY ASSURANCE Quality Assurance………………………………………………………………36 Analysis of Quality Assurance…………………………………………………39 Notable Best Practices in Quality Assurance…………………………………44 Key Quality Assurance Findings from the Kenyan Example Case Study…..48 CHAPTER 5 CONCLUSION Conclusion………………………………………………………………………52 Invalidated Assumptions…………………………………….…………………52 Recommendations………………………………………………………………53

1.1 1.2 1.3 1.4

2.1 2.2 2.3

3.1 3.2 3.3 3.4

4.1 4.2 4.3 4.4

5:1 5:2 5:3

Appendix A: Site Visit Research Questions

3 Foreword The Mater’s of Art in International Development Studies (IDS) at the Elliott School of International Affairs (The George Washington University) is a multidisciplinary program that is designed to prepare students for a career in development. In their final semester, students form teams and engage in a consulting assignment for a client of their own choosing, in the field of development. These teams work with their clients on a specific project, and clients this and previous years have included organizations such as Winrock, Africare, and Interaction. This report was compiled by a team of three MA candidates (N. Anta Gueye, Joshua McNary and Jungo Okai) from The Elliott School of International Affairs in their graduating semester (spring 2004). This represents their capstone project, for which the client was the Solar Energy Industries Association (SEIA), based in Washington DC. Formed in 1974, the Solar Energy Industries Association is the national trade association of stakeholders in the solar industry and acts as a lobbying group in Washington DC. In addition, SEIA provides its over 400 members with networking and financing opportunities and has formed close relationships with the World Bank, the Export-Import Bank of the United States, and several foreign governments through their Washington delegations. As a domestic NGO, SEIA coordinates the efforts of 14 state organizations in supporting local and regional advocacy efforts. SEIA’s membership comprises NGOs and commercial organizations including British Petroleum, Sharp, the Solar Energy Corporation, Shell Solar, BP Solar and Kyocera. SEIA has recognized the dissemination of rural solar electrification in developing countries as an institutional objective and seeks to facilitate the efforts of those member organizations currently operating or seeking to expand into the developing world. Acknowledgement During these several months of research, the team has received invaluable support, referrals, advice, feedback and/or information from different individuals and groups. We would like to thank: Glenn Hamer (Executive Director, SEIA), Colin Murchie (Legislative Assistant, SEIA) and the entire staff at SEIA, our program director Dr. David Gow, John Rogers (SOLUZ Inc.), Shelton Davis (The World Bank Group), Anil Cabraal (ASTAE/World Bank), James Onsare (Kentan Safari General Manager, Kenya), Mark Hankins (Energy for Sustainable Development AFRICA, Managing Director), Robert Muriuki (Brand Leader, Chloride Exide), Festus Irimu (Area Manager, Chloride Exide), Mbatha Mbithi (Advances Manager, K-Rep Bank Limited), Charles O. Rioba (Solar World Ltd.), Elizabeth Musyoka (Sales and Marketing Executive, Solagen), Fred Sokoro (Sales and Marketing Executive, Solagen), Jamleck K. Gicharu (Branch Manager, Solagen), Daniel and Mary Jocombo, Steven Waley and the children of the Mawaka Highway Junior Academy, our colleagues of the IDS class of May 2004, and the Elliott School of International Affairs and Dean Harry Harding.

4 Executive Summary Rural Photovoltaic (PV) dissemination initiatives in developing countries are subject to three primary constraints. Limited credit facilities and the lack of investment capital among Solar Home System (SHS) customers represent significant financial barriers. Poor information transfer, recurrent costs and discrepancies in donor-supported programs serve to hinder capacity strengthening efforts. Market failures associated with information and geographical distance undermine quality assurance. These conclusions were derived from case-study analysis and interviews. The findings of the Kenyan case study demonstrate the weaknesses inherent in efforts to apply a blueprint solution to a development challenge. In particular, the informal sector was identified as a primary component of market vitality and we recommend demand-side capacity strengthening as a means of increasing the skills of informal technicians. Rural PV dissemination was found to have limited application in poverty alleviation efforts beyond electrification projects in rural schools and health centers due to issues of pricing, financing and low demand. This document recommends several strategies for supporting rural PV dissemination. Many PV projects take a “market first” approach emphasizing the role of PV within the power sector. We have chosen to emphasize the “development first” approach. Thus we argue that projects must be framed in the context of addressing the development priorities of a community and/or region and must endeavor to create sustainable endogenous institutions. For the former, we recommend that development practitioners establish a clear picture of community specific development needs and tailor PV accordingly. For the latter, we recommend that project designers commit to involving indigenous stakeholders in all levels of the project and seek to support the emergence of viable assessment and planning capacity. We argue that although the cash-sales model has several drawbacks, measures aimed at expanding financing options for consumers should focus upon improving, rather than replacing, the cash sales model. We recommend that raising consumer awareness of the capacity and compatibility of PV components is a key strategy in improving the technological capacity of informal technicians. Similarly, we argue that efforts to train and standardize informal technicians in the Kenyan PV market are misdirected and must be reconsidered. Within the development context, PV technologies represent essential tools in supporting poverty alleviation initiatives. PV dissemination alone is insufficient in meeting the needs and demands of rural communities. Rather, small-scale renewable technologies, such as PV, are best utilized in support of existing poverty alleviation strategies. PV technologies hold significant potential for rural electrification efforts and their dissemination in the developing world represents a positive movement away from expensive, infrastructureintensive electricity grids. By overcoming the primary constraints to rural PV dissemination in developing countries, such potential may be fully realized. This process will require long-term support and can provide unparalleled opportunities for supporting the flowering of indigenous capacities. PV represents a decentralized, small-scale technology and may prove both affordable and of high quality given appropriate initiatives. From an academic perspective, PV projects represent a cross-sector strategy and are thus a useful source in the consideration of the roles of public, private and nonprofit institutions in the development process.

5 Abbreviations CBK CIDA EC ESCO ESMAP G8 GEF GWU IEA IFC JAICA Jua Kali KCB Ksh K-REP kWh MDG NGO PRSP PV PVPS REEF SACCO SEIA SHS USAID W WB The Cooperative Bank of Kenya, Ltd The Canadian International Development Agency European Commission Energy Service Company Energy Sector Management Assistance Programme Group of the 8 Leading Industrialized Countries Global Environment Facility George Washington University International Energy Agency International Finance Corporation The Japanese International Cooperation Agency Informal Sector Technician (means “Hot Sun” in Swahili) Kenya Commercial Bank Kenyan shilling Kenya Rural Enterprise Programme Kilowatt hour Millennium Development Goals Non Governmental Organization Poverty Reduction Strategy Papers Photovoltaic Photovoltaic Power Systems Renewable Energy and Energy Efficiency Fund Savings and Cooperative Credit Organization Solar Energy Industrial Association Solar Home System The United States Agency for International Development Watt The World Bank group

6 CHAPTER 1: INTRODUCTION 1:1 Introduction Consider the daily experiences of two women: Mary, an American and Ayan, a Kenyan. Upon waking in her climate-controlled apartment, Mary strolls into the bathroom and, after turning on the light, washes her face with warm water from the tap. She cooks her breakfast on an electric stove while watching the morning news on her counter-top television. Her refrigerator blew a fuse during the night and, although the milk is lukewarm, she suffers no inconvenience. Waking up eight hours later, Ayan experiences quite a different morning. She washes in the same water that her three sisters and two brothers have used; hauling water from the local well is an arduous task. She moves about the dark kitchen and revives the smoldering charcoal in the hearth. As the smoke rises and the air turns grey, she coughs and opens a window. Over time her cough will worsen and she will suffer from respiratory ailments. But this morning she considers how her youngest brother, ill from an infection, has just returned from the local clinic to inform her that, during the power outage last night, the refrigerated vaccines spoiled. In Mary’s case, reliable electricity is a given reality and when the power does fail, she is confident that it will soon return. For Ayan, the lack of electricity in her rural community serves to exacerbate existing poverty. Indeed, the extension of reliable, clean energy sources to rural areas remains a formidable challenge facing the developing world and is a necessary component of long-term poverty alleviation efforts. Particularly in large populous countries such as China and India, the combination of a widespread, predominantly rural population and the high cost associated with the extension and maintenance of suitable infrastructure combine to limit the plausibility of traditional grid-based energy distribution. Furthermore, heavy dependence upon fuel wood and biomass for cooking and heating among rural communities has been shown to increase the incidence of respiratory ailments while contributing to environmental degradation. Particularly in rural areas, solar-related technologies represent viable small-scale sources of renewable energy. Photovoltaic systems are portable, increasingly affordable and require minimal maintenance. Aside from pollutants expelled during the manufacturing process, photovoltaic systems did not create a waste stream. By converting a free and abundant source of energy into direct current electricity, photovoltaic technologies may be used to power a wide variety of appliances from basic lighting to refrigerators. Photovoltaic systems may be installed by an individual household or may be linked together to form a grid with sufficient energy production to power an entire community. Finally, when linked with appropriate financing mechanisms, photovoltaic systems represent a cost-effective tool for securing needed electrical capacity. Photovoltaic technologies hold great potential for extending electrification into rural areas of developing countries. Rural electrification that does not degrade human and environmental well being is a fundamental aspect of sustainable international development. Sustainable development refers to meeting the needs of people today without inhibiting the ability of individuals in the future to meet the same needs. In this manner, sustainable development may be considered a type of cautionary principle akin to the saying “a frog does not consume the pond in which he lives.” If international development initiatives are to have any lasting success, they must support the formulation of

7 sustainable institutions within a country. Ideally, such institutions gain the capacity to identify needs and appropriate responses and form the stable foundation upon which new organizations may be conceived. Rural electrification supported by renewable resources allows individuals to benefit from electricity without polluting their environment. Indoor air quality may be improved as solar stoves replace traditional indoor fireplaces. Women and children can dedicate less time to foraging for fuel wood while evening lighting permits weaving or studying at night. Rural health centers can refrigerate vaccines and PV powered water pumps can permit families to draw water from deeper wells. In short, PV technologies offer a broad range of solutions to the challenges of sustainable development. 1:2 Argument We will address a primary and secondary argument. In our primary argument, we will contend that although rural PV dissemination blueprints often outline meaningful, case study supported strategies, implementing institutions must develop an extensive understanding of project specific needs and realities. In our secondary argument, we will contend that PV dissemination should be viewed as a tool to support poverty-alleviation strategies, rather than as a development priority.1 In formulating this argument, we will demonstrate the following thesis: The consistent barriers to rural PV dissemination are associated with financing mechanisms, capacity buildings and quality assurance and are exacerbated by informational and financial market failures and geographical limitations. Efforts to address these barriers may be identified but are site-specific and require intimate knowledge of the program and market contexts. In the case of Kenya, functioning financing arrangements, technical expertise and consumer awareness differ from the preferred models identified among leaders in the field of PV within the development community. Issues of financing, capacity strengthening and quality assurance often overlap. For example, a consideration of the barriers facing the emergence of appropriate financing measures in a developing country should include an assessment of awareness raising initiatives and consumer satisfaction with existing options. Although we have chosen to discuss the above-mentioned challenges in an independent fashion, assigning each barrier to a separate chapter, we recognize that these issues are interrelated. 1:3 Critical Questions What are the primary barriers to project design in rural PV dissemination in LDCs? Through literature review, interviews and in-country research, we have identified three fundamental constraints to rural PV dissemination in developing countries: financing, capacity building, and quality assurance). In addition, we have identified three major market failures associated with these constraints: financial, geographical, and information. What measures may be identified that address these constraints?


This second argument will rely heavily upon our Kenya research and will be emphasized towards the end of this text

8 These are meant to be notable best practices, as identified through research and analysis, for each area of challenge. How do the above-identified measures contrast to the practical experience in a LDC (Kenya)? This will mainly express the differences between the research and the realities in the field. 1:4 Methodology An estimated one-third of the world’s population resides in regions that lack access to electricity and in which the construction of a conventional grid infrastructure is not a viable measure. In such regions, contemporary renewable energy sources may hold the answer to rural electrification, pending advancements in system design, reliable finance, delivery, and ongoing support. Several NGOs have developed individual programs in an attempt to address these issues and facilitate the spread of renewable energy technologies as alternatives to traditional, unsustainable energy sources. In addition, a number of different actors (NGOs, International development agencies and energy providers, some governments, etc.) have shown a growing interest in the extension of renewable energy technologies to more developing countries, and the associated investment potentials. However, current efforts are comparatively dissociated form one another and do not operate according to any agreed-upon paradigm of success; they further suffer from a limited ability to replicate or extend results. Consequently, many developing countries either establish or expand a conventional utility grid based on both a lack of alternatives and access to readily available paradigms, or simply leave the need of their rural communities to access better and more sustainable sources of energy unmet. This consultancy, provided to SEIA, aims to explore challenging areas of rural PV dissemination in LDCs. In addition to identifying and further analyzing industry standards, the consulting group executed a country visit for the purpose of gathering practical insights, and possibly exploring main differences between the literature recommendations and the actual field experiences. Limitations This study is subject to a number of limitations. The confined duration of the research, conducted between January and April 2004, limited financial resources and a lack of extensive access to in-country stakeholders during the site visit have all directly affected the scope of our project. It would have been particularly beneficial to conduct additional site visits for the purpose of supporting comparisons. We hope to see further research take place. Research Process

9 The three challenging areas to PV rural electrification in LDCs (i.e. financing mechanisms, capacity building and quality assurance) were identified through primary and secondary source research conducted the semester before the project started, and by using the team’s academic training in program management. These areas were mentioned by different literary sources and leaders in the field, and were later confirmed through more project research, and interviews. The team relied on four categories of sources for this project: literature review, personal interviews, site assessment, and stakeholder surveys. Literature reviews were conducted through Internet searches and text research. Referrals to sources of information were requested and obtained from multi-lateral agencies, multinational corporations, non-governmental organizations, commercial ventures, and published and unpublished university scripts. Among the reports collected and reviewed, the team chose to rely more heavily on manuals produced by task forces mandated by the International Energy Agency. These manuals present a wide variety of topics on PV rural electrification in developing countries, and were compiled through a participatory process that included developed and developing country experts, the UNDP and the World Bank. Additional sources come from institutions such as the World Bank, GEF and various NGOs. Personal interviews with stakeholders in Kenya served as a means of contrasting practical experiences with the literature review. Assumptions Assumption 1: We assumed that perspective that we would research the project design process. However, we found that a discussion of project design was highly dependent upon further discussion of project implementation. We have chosen instead to focus our consideration on overlapping barriers that consider both the design and implementation contexts yet remain barriers nonetheless. Assumption 2: We believed that rural PV dissemination efforts would prioritize the poor. In particular, we assumed that access to PV would be facilitated through credit-related financing schemes. We also assumed that the poor would acquire PV systems for the purposes of income generation and improved lighting (for nighttime studying, etc) as these relate to povertyreduction. Assumption 3: PV projects would be large scale and in part directed to poverty reduction. We expected to see PV projects to target educational and health institutions (i.e. powering boarding schools, vaccine refrigeration, powering rural health centers, etc). We also expected to see multiple-households sharing a large capacity panel. Assumption 4: Given the fact that Kenya has the largest unsubsidized PV market, we assumed that there would be considerable NGO involvement with PV dissemination and development related projects. Kenya Field Research

10 The field research in Kenya was conducted March 14-26, 2004. The purpose of the trip to Kenya was to visit project sites, collect additional written documentation and meet with stakeholders to gather practical insights and experiences. Prior to the trip, we secured interviews with actors in the PV market including financial institutions, NGOs, universities, suppliers and an international aid agency. Questionnaires were prepared for each of these stakeholders and were presented in semi-formal and informal interviews. The intended number of interviewees per site was kept to a minimum in order to extract as much information as possible in the limited amount of time allocated to site visits. Background information on Kenya, along with our methodology for the site visit, can be found in Appendix A. The site visit proved essential, as we were able to speak with a wide variety of stakeholders. We maintained a flexible schedule and, when several previously identified contacts were not available to meet, relied heavily upon referrals in accessing additional interviewees and site visits. The site visits were successful, in large part, due to the support of our in-country contacts, particularly our translator and host. The questionnaire was revised several times during the field research. We found that we were able to learn more from interviewees by asking them to tell us the story of their PV experience, rather than asking pointed questions. Furthermore, as the emphasis of our research shifted from financing mechanisms to capacity strengthening and quality assurance, we altered our questions to reflect emerging research priorities.

11 CHAPTER 2: FINANCING MECHANISMS 2:1 Framework of Financing Mechanisms for PV Dissemination One of the main obstacles for the sustainable application of Solar Home Systems (SHS) is the lack of suitable financing options for consumers. SHS typically targets rural families with no access to a formal electricity grid. In order for such consumers to acquire PV electrification, mobilization of financial instruments is critical. Installation of SHS typically costs between US$500 to US$1200 depending upon size, components, and country-specific taxes and duties.2 Also, although an appropriately installed SHS incurs minimal maintenance costs, typical monthly payments may cost between US$5 and US$10.3 Thus the customer must recover the cost of both the installation and of the operation and maintenance of the system. The lack of affordable credit schemes and/or other forms of financing mechanisms is regarded as the main barrier to widespread introduction of SHS in rural areas of the developing world.4 Although the establishment of appropriate financing mechanisms is a primary challenge to PV dissemination, dialogue within the international community tends to focus primarily on this issue. Indeed, the majority of the finance related documents that we reviewed focus on this issue. However, in practice, the financing issues of PV dissemination require additional factors including (1) economic stability, (2) trade mechanism, (3) financing mechanism and (4) willingness to pay.5 These factors will be explored separately in section 2.2. Financing Mechanisms of PV dissemination This section will consider the primary financing mechanisms that may be applied within a PV market. These will include: (1) cash sales, (2) dealer credit, (3) end-user credit, (4) hire-purchase and (5) fee for service.6 We will discuss the basic concepts behind each financial model and provide an analysis of their benefits and weaknesses. It should be noted that the aim of this section is to compare and contrast different financing models and that no single model will be applicable to every situation. Model 1 Cash Sales The cash sales model is the most basic approach to financing SHS dissemination due to its simplicity and limited number of actors. According to this model, the consumer purchases a SHS with some type of physical capital. The supplier may be contractually bound to provide technical assistance, although such an arrangement is not an inherent aspect of this model. Advantages

IEA-PVPS T9-01:2002. “Financing Mechanisms for Solar Home Systems in Developing Countries.” p5 Ibid 4 Kenya Implementation Manual: Financing Mechanisms for Solar Electric Equipment, Joint UNDP/World Bank Energy Sector Management Assistance Programme, (ESMAP), July 2000 5 These issues will be discussed in the analysis portion of this chapter. 6 IEA PVPS T9-02:2002, p1


12 Due to the limited number of actors associated with the cash sales model (the supplier and the consumer), the transaction costs can be minimal and the model is relatively easy to adopt within varied economic contexts. In addition, this model does not require the supplier to maintain large sums of capital. As a result, access to SHS markets can be faster and easier compared to other models of dissemination. Disadvantages The most significant limitation of this model is the difficulty associated with dissemination of SHS to the rural poor. Because consumers pay cash to purchase SHS, buyers must maintain a certain level of economic capability to invest the initial capital costs of SHS. As a result, this model often does not extend beyond urban areas and fails to reach the rural poor. In addition, this model provides poor incentives for suppliers to honor service and maintenance contracts once the purchase has been completed. In terms of product quality, this model also tends to promote the cheapest and lowest-quality SHS due to poor product quality standards and high-initial investment costs. Model 2 Dealer Credit Whereas the cash sales model requires high-initial capital investment, the dealer credit model enables end-users to purchase SHS with low initial costs. In this model, SHS vendors provide a consumer credit or installment payment facility. Dealer credit may be characterized by comparatively short payment terms, high down payments and high interest rates. Depending upon the dealer, payment terms are often six months to a year, a down payment may reach 50 percent and interest rates may reach 20 to 25 percent.7 Because most SHS suppliers lack the financial capacity and/or the know-how to provide credit, larger institutions such as microfinance and donor organizations often support dealer credit. The dealer credit model involves various stakeholders and generally requires a credit institution to supply financial support for dealer credit activity. This often results in dealer re-financing and increasing interest rates. Also, this model often requires government supported regulations, certification, and standards for sustainable dealer credit. Dealer credit may also serve to expand the market through improving credit related processes. Advantages This model serves to reduce the high-initial costs of SHS. With accessible credit, end-users are able to purchase the system while maintaining ownership. Also, financial support from larger institutions allows dealers to provide not only SHS sales but also maintenance, training, and other after-sale services. Thus, if PV rural dissemination is to target the poor, introduction of dealer credit may prove more appropriate than the cash sales model. Disadvantages While micro-finance and dealer credit are accessible in many developing countries, these financing options remain relatively unknown, particularly among the poor. Though dealer credit has the potential to reduce the high-initial costs of SHS, re-financing mechanisms associated with larger financial organizations may result in higher interest rates. Thus, even though the rural poor have easier access to purchasing SHS, high interest rates may be a barrier for SHS

Ibid, p7

13 dissemination. In this context, dealer credit suppliers are required to have extra skills and knowledge about credit administration, collection, and management.8 Model 3 Third-Party Credit Unlike the dealer-credit model, when consumers utilize end-user credit, they do not receive credit from a dealer, but rather from another credit provider, such as micro-finance organizations or NGOs. The SHS dealer thus avoids transaction costs of delivering credit and can operate his/her business with fewer risks. This model also permits users to maintain ownership of SHS and be responsible for maintenance and repair. In several developing countries, credit institutions directed towards rural populations, such as rural development banks, savings and credit co-operatives, NGOs, and other rural development organizations already exist. However, because PV systems are often classified as consumer products and non-income generating opportunities, many of these credit institutions do not provide credit for SHS purchasing. If purchasing of a SHS was viewed as an acceptable credit target activity, these credit institutions could significantly contribute to SHS dissemination, both as credit providers and as SHS promoters and extension networks.9 Advantages The user-credit model effectively removes and/or reduces initial investment costs for SHS purchasing. End-users can receive support from those financial institutions that maintain specialized knowledge of, and experience with, the incentives of rural financing. SHS dealers are able to secure their profits with fewer risks and worries about collection of payment. With more rural financing institutions operating as credit providers and promoters, dissemination of SHS can be significantly enhanced. Disadvantages While initial purchasing costs may be minimized, this model tends to support high interest rates, except in cases where governments or donor agencies actively implement programs to keep interest rates low. High interest rates and loan payments may prove a disincentive for the rural poor, thus, without government or other donor intervention, this model may limit dissemination of SHS for the rural poor. Model 4 Hire Purchase The hire purchase model requires SHS dealers to have a high degree of financial resources and administrative and technical infrastructure. In this model, SHS dealers lease the system to endusers for a predetermined period (usually two to three years). When the payment is complete, the ownership of the SHS is transferred to the consumer. This model enables consumers to receive installation, maintenance and after-sale services as well as financial support from PV companies.10
Designing and Financing Solar Energy Usage Notes, World Bank, ( Institutional Framework and Financial Instruments for PV Deployment in Developing Countries, International Energy Agency IEA-PVPS T9-06:2003 10 Ibid
9 8


Advantages With dedicated technical and financial support from PV companies, this model can effectively reduce the initial costs of SHS in addition to lowering interest rates. This model also assures the quality of the system through securing professional maintenance services for the leasing period. Disadvantages Despite its potential for SHS dissemination among the rural poor, the hire purchase model has not spread in developing countries. This is due to consumer confusion and the inability and/or unwillingness of many PV suppliers to serve as hire-purchase model providers. In order for a company to implement this model, economic conditions within developing countries must permit the PV market to mature. If the end-users stop their payment during the lease, PV companies may not be able to sustain their profits. Also, because end-users benefit from a PV system that they do not own, the incentives for end-users to maintain the system may be low.11 Model 5 Fee for Service The fee for service model treats SHS dissemination as a utility service, somewhat similar to a grid system. In this model, there is an energy service company (ESCO) that provides electricity to rural customers. Because ownership of SHS stays in the ESCO, the company provides installation, maintenance, repair, and additional services to end-users in return for monthly fees for connection and service. While end-users may never own an SHS, they are able to receive guaranteed services from the company. Clearly, the fee-for-service model requires strong financial capacity on the part of the ESCO in addition to commitment for long-term investments. In order to support the ESCO’s financial requirements, credit institutions that supply long-term capital to the ESCO must be established. Even with the support of financial institutions, in many developing countries, application of this model is difficult due to economic fluctuations, an unorganized energy sector, lack of regulation and standards for energy and necessary rural infrastructure. The government plays a significant role in the processes of legal framework, economic sustainability, and other pre-conditions for the establishment of the ESCO.12 Advantages Once application has been achieved, initial cost may be quickly recovered and permit the widespread application of services to rural areas. This model also provides relatively highquality systems and services because of the long-term agreement. Disadvantages This model produces monthly fees and transaction costs and the rural poor may not be able to access the services. This model also requires monthly payments from end-users on an at-will payment plan. Although such a structure permits consumers to pay when savings are available, the lack of a payment schedule may constrain the cost recovery process. In addition, electricity

Institutional Framework and Financial Instruments for PV Deployment in Developing Countries, International Energy Agency IEA-PVPS T9-06:2003 12 Ibid

15 is a difficult medium to markets any individual with a rudimentary understanding of wiring can easily tap into the system. Despite its potential, the fee-for-service model is the most difficult and time-consuming model to implement. 2:2 Analysis of financing mechanisms for PV dissemination Commercialization is one of the primary goals of SHS dissemination projects. While most of the existing literature only deals with issues specific to financing mechanisms for PV dissemination, commercialization, in reality, is much broader and more complex. Just within the financial sector, commercialization of SHS involves issues such as (1) macro economic stability, (2) trade mechanisms, (3) financing mechanisms, and (4) consumer’s willingness to pay. We believe that there is a need for a stronger understanding of all these factors in order to grasp the complexity of commercializing SHS. In addition, the nature of the financial sector differs from country to country, so the international aid community must emphasize country-specific approaches, rather than advocate blueprint-type strategies to PV dissemination. Discussing each of these abovementioned issues in this section will hopefully contribute to a better understanding of the complex nature of financing PV rural dissemination in developing countries. (1) Macro Economic Stability: Inflation, market regulations, and financial sector capacity Establishing appropriate financing mechanisms heavily relies upon a country’s macro economic stability. In order for a market to function well, fundamental market regulations must exist and be enforced, inflation rates have to stay stable, and there should be a strong supporting financial sector. Without these conditions, the majority of the financing mechanisms discussed earlier are not applicable. Macro economic stability is especially important because almost all PV equipments used in developing countries are imported from developed countries, thus macro economic instability would jeopardize the sustainability of imports and discourages foreign investment. This point can be illustrated with the case of Zimbabwe, where the World Bank and JICA have been conducting PV dissemination projects that have been failing because of the country’s hyperinflation rate of over 300%.13 Also, in many countries, banks and micro-credit institutions have failed to reach rural consumers or have lacked the capacity to support PV dissemination. Without a stable economy with appropriate regulations and financial sector capacity, it would be pointless to discuss which financing scheme is best. Rather, it would be more appropriate to identify which model is most applicable to the market conditions. (2) Trade Mechanisms: Value-added Tax for PV equipments Because the vast majority of PV equipment is imported from Italy, the United States and Japan, a developing country government’s policy on international trade, especially on value-added taxes for PV equipments, will have a significant impact on the pricing of PV related equipments and services. Indeed, in the case of Kenya and many other countries, the cost of PV equipments is high because the national government imposes high value-added taxes on PV related products. Since 1994, the duty and value-added taxes in Kenya resulted in a cumulative tax of 55 % on PV


The information is taken by informal interviews with people involved in the project

16 panels.14 Indeed, several Kenyan PV venders identified the value-added tax as one of the major concerns in terms of pricing PV products and expanding their market to the rural poor populace.15 The potential reduction of the value-added tax requires the national governments involvement. While there are strong grassroots voices advocating change in the national tax policy, this local movement is unlikely to affect such policies. In order to improve the situation in favor of PV dissemination, there is a need for stronger intermediary organizations that can effectively lobby before the national government. The site visit to Kenya suggests that while NGOs and PV associations do not currently have the capacity and political influence to affect the national government’s tax policy, they are the most appropriate organizations to play such lobbying and advocacy roles in the long run. In order to develop the potential and capacity of these intermediary organizations, capacity strengthening and strong supporting efforts will need to emanate from stakeholders, including local communities, PV venders and technicians, multinational corporations, and international NGOs and PV advocacy groups. (3) Financing Mechanism: Appropriate Models for Sustainable Market Growth Establishing the appropriate financing model(s) to use should come after careful examination of the country’s political, economic, and social conditions and from an understanding of the pros and cons of each financing model. Although there is not an “ultimate best” financing model, according to our research and discussions, we feel that the fee-for service model is the preferred practice for financing PV dissemination. The model guarantees high quality of PV products and services and end-users do not have to invest a large amount on the initial cost of the PV system. Thus, the model can effectively remove the high-initial cost market barrier, while it promotes better quality assurance of PV products and services and significantly expands the market. However, we also feel that the fee-for service model can be an unrealistic model to apply to most developing countries. Considering the requirements that the model calls for, i.e. strong technical and financial capability of PV companies, a stable economy, an organized energy sector, regulations and standards for energy and necessary rural infrastructure, the international aid community should be extremely cautious about the applicability of this model to target developing countries. Implementing agencies will need to determine which model would be the most appropriate and applicable, depending on the specific political, economic, and social conditionality of each target country. Attempting to replicate successful experience from one developing country to another without exploring these issues could lead to failure. For instance, many countries do not have well-functioning market regulations that support healthy market competition and prohibit illegal activities. Plus, financial institutions often do not exist in rural areas and hesitate to provide credit to those lacking collateral. Domestic technicians may lack the appropriate skills and knowledge about PV systems, and end-users may not have sufficient information about the usage and potential of PV system. To establish a viable financing mechanism in a target country, aid-


The figure is taken from an informal note by Mark Hankins, 15 Information is taken in informal interviews with PV venders during site visit to Kenya

17 agencies must effectively explore these conditions to determine which model is the most appropriate for the market.16 (4) Willingness-to-Pay One financing-related challenge of SHS dissemination projects is determining the consumers’ willingness to pay. As we noted earlier, the initial purchase and installation cost of SHS serves as a barrier for most rural customers. End-users also become aware that SHS requires some degree of maintenance. For the consumer, system failure early in the ownership period can cause critical financial problems. Most of these failures are caused by the poor quality of the system and the lack of technical standards. These issues consequently have substantial influence on the customers’ acceptance of the technology and their willingness to pay. Improving the quality of PV products will result in higher durability but requires higher cost. Thus, customers have to make a choice between buying a cheaper low-grade system and facing the risk of recurring high maintenance fees, or buying a more costly high-grade system and have low maintenance fees.17 Experiences from the site visit to Kenya show customers are generally interested in spending as little as possible even though they know that the quality of low-grade products is poor.18 In a demand driven SHS market, consumer’s willingness to pay is a critical factor in determining the expected level of quality assigned by the consumer to the technology. In countries where consumers are willing to pay the extra cost for better quality products and services, SHS suppliers are able to market higher quality products with increased lifetimes. However, in those countries where consumers are not willing to pay, the supply side has limited choices as to their marketing strategies relative to their different products. Because this aspect of the market would differ depending on consumers’ willingness to pay, it is important to identify the consumer’s priority about SHS. 2:3 Key Financing Findings from the Kenyan Site Visit The previous section introduced different financing mechanisms for PV rural dissemination in developing countries. While the section provides general information on financing mechanisms, it does not consider how those mechanisms work in the field. In a way, the previous section serves as a blueprint for financing mechanisms. This section uses the Kenyan market as a case study to provide practical experience. As noted earlier, financing mechanisms used in each country may differ according to the conditions on the ground. Thus, the intention of this case study is not to offer an in-country analysis that can be replicated throughout the developing world; rather, this section tries to show how recommendations from a blueprint may differ from realities in the field.


The Photovoltaic Market Transformation Initiative (“PVMTI”) is an initiative of the International Finance Corporation and the Global Environment Facility, which aims to promote the sustainable commercialization of PV technology in the developing world by providing examples of successful and replicable business models that can be financed on a commercial basis. More info and contacts available at 17 IEA-PVPS T9-01:2002, ”Financing Mechanisms for Solar Home Systems in Developing Countries.” 18 Ibid

18 By the year 2000, about 120,000 solar home systems (SHS) had been installed in Kenya. A joint UNDP/World Bank Energy Sector Management Assistance Programme (ESMAP) conducted a two year extensive study on the financing mechanisms.19 The study found that these were limited. Most of the SHS sales in Kenya follow a cash sales model and require end-users to arrange for installation, maintenance, repair, and other after-sales activities. As a result, the quality of SHS products and maintenance standards in Kenya is also poor. Many end-users fail to invest in recommended components, resulting in high incidence of failed performance. There is evidence that such performance problems have resulted in lowering the confidence and understanding of people about SHS products and limiting the dissemination of the technology.20 While most of SHS sales have followed a cash-sales model in Kenya, there are increasing numbers of financing organizations that provide other types of financing mechanisms. We identified three leading financing institutions: Kenya Rural Enterprise Programme (K-REP), Cooperative Bank of Kenya Ltd. (CBK) and Savings and Cooperative Credit Organizations (SACCO). This section will discuss the major findings about financing mechanisms in Kenya. Emphasis will be directed at current conditions of financing mechanisms, a major challenge for PV dissemination in rural areas, pros and cons of the cash-sales model, the impacts of internationalaid intervention efforts, and future challenges for the market. Finding 1: PV Systems and Consumer Priority Many international development aid projects are concerned about reaching the poor. In one way, targeting the poor makes sense since the majority of poor populations in most developing countries live in rural areas. It is also safe to assume that most poor people in rural areas most likely do not have access to grid electrification systems. Thus, PV rural dissemination projects may often target the poor. However, once in the field, our team quickly discovered that securing PV is not a high priority among the rural poor. While middle-class and lower middle-class people are indeed purchasing PV, almost all the end-users utilize PV systems to watch TV, listen to the radio, and light a house. For the lower-class poor, watching TV is not a priority when they face issues of securing shelter and food. Thus, even if the supply side (development-aid agencies) would like to target the rural poor by providing credit schemes or subsidies, the demand side (lower-class poor) may not want it. As a result, the majority of lower-class poor do not purchase PV systems, regardless of whether or not they have access to credit. Finding 2: Pricing of Products The major challenge for PV dissemination in Kenya is the high initial cost of PV products. Rural lower and middle class populations, in general, cannot afford the high incremental costs of solar home systems with their income. While cheaper products are introduced in the market every year, almost all the PV products are currently imported goods from Europe, U.S.A, and Japan. Value added taxes on these import goods, high dealer margin, and high cost of business
19 20

Duke et al. “Photovoltaic module quality in the Kenyan solar home systems market.” ESMAP The Kenya Portable battery Pack Experience: Test Marketing an Alternative for Low-Income Rural household Electrification, Joint UNDP/World Bank Energy Sector Management Assistance Programme (ESMAP), May 2001

19 operation result in high costs of PV products in the Kenyan market. Due to the fact that many rural PV buyers are engaged in agricultural production and do not have sufficient cash in hand, most buyers go after the cheapest PV equipment available in the market. A typical small affordable solar panel in Kenya is a 14w panel costing about 4, 000 Ksh (US$ 55), and an entire SHS system can cost over 10,000 Ksh (US$ 130).21 Affordability of PV equipments very much relates to quality assurance. Because they cannot afford a large PV system with high quality standards, people generally buy smaller size panels (14 – 20 w) that can power four to six 60-watt light bulbs and a black and white TV. Also, because of its high cost, most people do not purchase a charge controller, which is critical to the prolonged lifetime of a PV system. In the Kenyan market, end-users, especially small system owners, lack the appropriate knowledge about the importance of a charge controller. Pricing of PV services is another major issue for PV dissemination in Kenya. While the cost of services varies by company, types of service, and region, it is generally not affordable by main target groups. In reality, most SHS buyers hire local “Jua Kali” technicians to install and maintain their system. While local “Jua Kali” technicians are more affordable and available, using them has some negative impacts on the quality of PV products. End-users and potential buyers are interested in minimizing their cost while receiving the “good enough” service they expect. While many end-users face poor quality of products and services, they would rather pay less than disburse the additional cost for better product and services. Their main purpose for getting SHS is to have TV, radio, and some light. Their logic is that these, to them, are preferences, not necessities, and therefore should not require huge amounts of investments. Finding 3: The Cash-Sales Model As mentioned earlier, cash sales models tend to lower the quality of PV products and services. This is true in Kenya, where as we have just explained, many consumers buy “piece-meal” (buy different equipment at different time and place in order to save money) and use informal technicians in an attempt to save money. There are also many counterfeit goods available in the market, and they are naturally much cheaper than legitimate products. There are attempts made by some PV dealers to educate customers about appropriate systems and counterfeit goods but the issue of financial limitations convinces the customers to save some money today and worry about problems when they arise later. Another consequence of the cash sales model is the low quality of services by PV sellers and/or local “Jua Kali” technicians. Because customers in general live far away from PV venders, installation and maintenance service by PV venders tend to be far more expensive than customers can afford. Also, many venders do not have enough capacity to provide sufficient after-sales service to individual customers. As a result, most purchasers either do everything themselves or use a local “Jua Kali” technicians to install and maintain the system. From a financial perspective, there is a potential to reduce these quality risks by introducing other types of financing mechanisms that are believed to guarantee better quality of products and services.

The figure is referred from informal interviews with PV venders

20 However, in the case of Kenya, we believe that, since the market is so advanced and adapted to the cash-sales model, introducing other financial mechanisms may not improve the quality of products and services. We also feel that no matter how much credit rural customers receive, the budget they are willing to allocate to a PV system, in terms of money they have today, will not change. What customers want to purchase is the cheapest possible product with “good enough” quality, thus quality improvement provisions need to be made considering the current financing models used. Finding 4: International Attempts to Establish Credit Mechanisms in Kenya Since the early 1990s, there have been a number of internationally driven attempts to intervene in the Kenyan PV market and create alternative financing mechanisms that promote further dissemination of PV to rural areas and improve the quality of products and services. However, most of these attempts did not achieve the expected success or a sustainable outcome. The major constraint was the lack of understanding of PV technology in the Kenyan market. For example, in 1996, the World Bank ESMAP program introduced a method of financing SHS through rural credit groups. The program provided a complete credit package with minimum installation standards, inspection procedures, and maintenance agreements with local technicians. The projects created a high demand for the SHS loans and over 100 SHS were installed through this program.22 However, while this project created an alternative loan program for the SHS market and experienced a large demand for the loan, the successful practice was not replicated into the Kenyan SHS market after the World Bank project ended. 23 The main reason why this program did not continue was because of the lack of institutional capacity. In Kenya, we had an opportunity to look at a brochure of this type of credit program at a financial institution. While the program offers a set of credit packages for end-users with guarantee of high maintenance services, warranties and quality of products, we did not see any financial institutions with the capacity to actually implement the program with the degree of support they promise to provide. The World Bank’s project provided excellent support for finance and quality assurance, but such support was not replicable by lower-capacity local institutions. The Bank’s program was surely good, just not sustainable for the Kenyan market and its institutions. Currently, the World Bank, IFC and GEF are conducting a Photo Voltaic Market Transformation Initiative (PVMTI) project to provide low-interest finance to local consortium to develop innovative financing packages that would ultimately increase the consumer market. While the project offers a minimum of US $500,000 for companies to implement these projects, almost all the local companies end-up not contracting a project with PVMTI because of its strict project development procedures and qualification processes.24 We found that this type of projects usually tries to transform the financial market with top-down approaches. To us, the main reason why PVMTI has not been able to achieve concrete success is because the project is not tailored

Hankins, Mark. “The Kenya PV Experience.” November 2001. 23 The Kenya Portable battery Pack Experience: Test Marketing an Alternative for Low-Income Rural household Electrification, Joint UNDP/World Bank Energy Sector Management Assistance Programme (ESMAP), May 2001 24 Hankins, Mark. “The Kenya PV Experience” November 2001.

21 enough to the context of Kenya’s economy. This is evident from the fact that almost all Kenyan local companies chose to withdraw from PVMTI. To these companies, the project requires too much complex procedures and restrictions and does not reflect the specific needs of Kenya’s market. Considering the fact that mostly small-scale PV venders and rural informal technicians drive the Kenyan PV market, there is a need for PVMTI to be more flexible in its procedures. Despite the fact that these two major international attempts have failed to attain their goals, these projects have created some positive impacts on the PV market in Kenya. Today, there are some financial institutions (KCB and Barclays) that provide financial support for PV companies and end-users. Even though these financing programs are not widely known by venders and endusers yet, they have the potential to develop into more substantial mechanisms that would allow more venders and consumers to enter the PV market. Finding 5: Government Change and Future Potential of the PV Market In order for the PV market to function well and grow steadily in the future, stable political conditions and government support of the market are necessary. In the case of Kenya, the current national government does not have the capacity to provide sufficient support for the PV market. For instance, in order to prevent counterfeit goods from entering the market, there has to be proper legal regulations. However, the Kenyan Bureau of Standards currently does not have the capacity to enforce the regulations. Kenya experienced a smooth political leadership transition in the year 2002. While it is too early to tell how the introduction of this new government will influence the market, there is some optimism among stakeholders. Since the new government came in, interest rates have been kept low and more regulations have been enforced in various sectors of the economy. Long-term stability of the government and improved governmental capacity will be critical for further PV rural dissemination in Kenya.

22 CHAPTER 3 CAPACITY BUILDING 3:1 Capacity Strengthening for PV Dissemination in Developing Countries Capacity strengthening remains a critical element of any PV program.25 This refers to “the combined efforts to support the development of knowledge, competence, and well functioning organizations and institutions.”26 It is necessary to note that the key word in this definition is support. As such, the capacity strengthening process requires implementing agencies to establish an intimate understanding of program related institutions and the policy environment. In addition, because capacity strengthening initiatives seek to support improvements in institutional and individual capacity, these initiatives represent a long-term commitment and must continue beyond the time frame of a specific program.27 The following chapter will discuss how capacity strengthening relates to rural solar electrification in developing countries and identify key stakeholders. An outline will be presented of the capacity strengthening frameworks for PV dissemination proposed by the International Energy Agency and the United Nations Development Programme. This will be followed by an analysis of primary capacity strengthening concepts. In conclusion, notable best practices will be derived from this analysis.∗ Capacity Strengthening and Constraints Capacity strengthening initiatives face an array of constraints when implemented in developing contexts. The donor, recipient relationship involves varying expectations and values and these are often poorly communicated. Information dissemination is often costly and, particularly in rural areas and among various ethnic groups, may prove difficult to target. Monitoring and evaluation have become infamous terms within the development community as these activities are often outlined but rarely undertaken. In addition to these factors, the highly qualitative nature of capacity strengthening initiatives requires innovative evaluation measures and dedicated commitment. 3:2 Capacity Strengthening Analysis Through identifying stakeholders and conceptualizing a three-step initiative, the IEA and UNDP frameworks for capacity strengthening represent a useful blueprint for the formulation of a

We have chosen to refer to Capacity Building as “Capacity Strengthening”. This represents our belief that all individuals and institutions maintain a degree of capacity and thus the role of implementing agencies is not to build capacity, but rather, to support capacity strengthening. The March 16, 2004 phone interview with Shelton Davis reinforced this notion. 26 Swedish International Development Cooperation Agency (SIDA). “Capacity Development as a Strategic Question on Development Cooperation. Policy and Guidelines for Sida”. Stockholm, Sida 2000. 27 For the purposes of this chapter, short-term will refer to a period of 1-5 years, medium-term will refer to a period of 6-10 years, and long-term will refer to a period beyond 10 years. Capacity strengthening initiatives represent a supra-long-term as they seek to address individual and institutional needs over one or more generations. ∗ Training programs are a predominate them of both this chapter and the following chapter entitled “Quality Assurance.” In this chapter we will discuss training as it relates to the informal sector and consumers. We will discuss training of formal technicians and certification processes in the next chapter.

23 market specific capacity-strengthening program. These frameworks will also serve as a foundation from which more specific capacity strengthening constraints may be determined. Government capacity and formal technician training are emphasized in these documents and the following analysis will elaborate upon additional points of consideration. As will be argued below, capacity-strengthening efforts may be hampered by a limited scope of application, poor managerial capacity, recurring costs and market failures associated with information transfer.∗ The Scope of Capacity Strengthening Initiatives If capacity strengthening may be conceptualized as seeking to empower existing networks and support the endogenous formulation of domestic organizations, it must be considered along an extensive timeline. Capacity strengthening initiatives also require an acutely localized and dynamic set of priorities. The highly qualitative nature of capacity strengthening has led one researcher to remark that: “[it] is more akin to social experimentation than to social engineering.”28 In this context, programs involved in PV information dissemination, industry coordination and technical capacity must emphasize support over intervention. According to Horton, the donor funded development projects of the 1970s and 1980s may be characterized as planned interventions with specified timeframes and targeted resources intended to achieve a series of objectives. As such, these frameworks relied upon the following notions: (1) Operational models which have proven useful in one context may be exported to another context with similar results, and (2) Outside experts with little country-specific knowledge can provide meaningful technical and capacity strengthening support. This strategy of “organizational development” lost credibility during the late 1980s as the call for local participation and the development of existing capacity gained prominence within the development community.29 In contrast to strategies of intervention, capacity strengthening initiatives require dedicated resources and flexible, long-term timetables to achieve objectives which may not be measurable. As Edoho suggests, “the end purpose of capacity building is the development of human beings”. He points to skill upgrading, procedural improvement and organizational strengthening as key components; the language suggesting areas of focus that lack a definable end product (as

A note on definitions: Formal technician will refer to an individual who is employed, trained and uniformed by an established, registered PV component supplier. Informal technician will refer to any individual who lacks all three characteristics of a formal technician but who provides technical services to a consumer. An informal technician may be sub-contracted for installation or maintenance purposes by a vendor and he/she may receive PV specific training. Vendor will refer to an established, registered PV component supplier who provides packaged PV systems and maintains a network of loosely associated nation-wide dealerships. Dealer will refer to an independently owned store in which PV components are sold. Dealers may carry the components of a particular vendor. 28 Horton, Douglas. “Building Capacity in Planning, Monitoring and Evaluation: Lessons from the Field”. Knowledge, Technology, & Policy, winter 1999, Vol. 11, No. 4, pp 152-188. p 182 29 Horton, 157

24 compared to the construction of a school).30 In addition, due to the ongoing nature of the training and coordination processes, funding and human resource allocations must eventually be borne entirely by the recipient organizations if initiatives are to continue beyond the donor participation period. In practice, capacity strengthening for PV dissemination must address issues of customer and technician training and public/private coordination within the overall context of supporting the emergence of local expertise. Managerial Capacity Strengthening Successful coordination among stakeholders is highly dependent upon the ability of each organization to operate efficiently within the business environment. Managerial capacity strengthening refers to initiatives aimed at supporting three aspects of human resource development: entrepreneurial, technological and administrative.31 Entrepreneurial capacity is a fundamental component of successful markets and relates to the willingness of businesses to accept risk, identify opportunity, and manage daily operations. Technological capacity involves the ability of institutions to operate and maintain technical inputs while formulating appropriate technologies in response to changing market conditions. Administrative capacity is associated with the day-to-day operation of an organization including accounting and record keeping, human resources management and information technology.32 In his discussion of managerial capacity in Sub-Saharan Africa, Edoho suggests that USAID, World Bank and CIDA technical cooperation strategies aimed at enhancing managerial capacity have incurred high costs while producing few positive outcomes.33 Such strategies incorporate foreign technical assistance in enhancing indigenous capabilities and rely upon a measure of donor funding. In addition, donor and foreign technician involvement is intended to decrease over time to promote project sustainability. A survey of outcomes in 385 projects revealed that approximately 20 percent of projects delivered substantial outcomes. Inadequate training and poor incentives for information transfer were identified as primary constraints. In particular, researchers found that expatriates were often unwilling to pass on managerial skills for fear of eventually being displaced by local personnel.34 Recurring Costs A primary financing issue relating to capacity strengthening initiatives is that of recurring costs. Such costs include salaries, input maintenance and replacement and continuing education. Due to the extended commitment often involved in capacity strengthening initiatives, any sustainable donor-supported program must include a provision for the gradual shift from of donor support to

Edoho, Felix. “Management Capacity Building: A Strategic Imperative for African Development in the TwentyFirst Century.” In James, Valentine Udoh. Capacity Building in Developing Countries: Human and Environmental Dimensions. Praeger: London, 1998. p. 235 31 Edoho, 237. This author further notes that managerial capacity strengthening may be further divided into cognitive (competency and information transfer) and noncognitive (values, attitudes, and beliefs) dimensions. 32 Ibid 238 33 Such strategies relied upon expatriate technical assistance to support indigenous skill enhancement and called for the removal of foreign experts within a given timeframe. 34 For additional information on managerial capacity strengthening, including recommendations, please consult Edoho.

25 recipient responsibility. However, the application of this paradigm has consistently suffered from the inability of recipient organizations to maintain quality following the departure of outside support. Institutions involved in capacity strengthening are at a particular disadvantage as such initiatives hold little potential to generate income and prove self-financing. As a result, organizations may become aid dependent as donor or government funding is utilized to prevent complete failure.35 Market Failures Capacity strengthening relies upon the timely dissemination of accurate and pertinent information. Stakeholders at all levels are constrained by poor information transfer and technological inputs require supplemental training efforts. Demand-side information market failures in PV and SHS dissemination include issues of misinformation, lack of awareness and consumer education. Misinformation remains a primary constraining factor to capacity strengthening initiatives. A World Bank case study of a PV program in Sri Lanka found that dealers were misrepresenting the capabilities of their solar components and were neglecting to address recurring costs within the system.36 The broad variances between size, shape, capacity and compatibility among solar components from differing manufacturers and the unlimited combination of electronic devices serve to further confuse the consumer.37 Within this context, the consumer likely relies upon word of mouth, NGOs and/or vendors for information regarding PV and SHS systems. However, each of these sources presents drawbacks. Word of mouth relies upon the information held by others within the consumer’s social network and thus may be constrained in the quality of knowledge.38 Traveling NGO training programs reach a limited number of consumers and must regularly update consumers as technologies are altered. Product vendors remain a questionable source of consumer information as the quality of expertise among individual dealerships may vary. Dealerships also have an incentive to sell as many components as possible to the consumer regardless of need or compatibility; if they do not, the consumer is likely to take his/her money elsewhere. Public demonstrations of the capabilities of a technology may be undertaken for little cost. During the in-country research, a vendor was observed organizing daily displays in the market place in front of the store. Most notable among these displays was a solar panel powering a water pump. Awareness raising is another information related capacity strengthening issue. An appreciation for the role that PV dissemination may play in supporting rural electrification efforts must be

For a complete discussion of this matter including lessons learned, please consult Christer Gunnarsson’s, Capacity Building, Institutional Crisis and Issue of Recurrent Costs. 36 IBRD “Best Practices for Photovoltaic Household Electrification Programs.” p. 68 37 Although PV systems require a degree of technical understanding, it is necessary to note that rural communities do not lack the capacity to gain installation and maintenance skills. According to a well-informed source, if a rural farmer can master the workings of a cow, than he/she can easily operate and maintain a PV system. Interview 3-2304 38 Acker and Kammen’s study of the Kenyan PV market found word of mouth to be the most common source of introductory consumer information, as identified by 75% of respondents. Acker, Richard H. and Daniel M. Kammen. “The Quiet (Energy) Revolution: Analyzing the Dissemination of Photovoltaic Power Systems in Kenya.” Energy Policy, Vol 24, No 1 pp81-111, 1996.

26 established among electrification, development and financing related government agencies. Trade facilities and universities provide electrical technicians and must recognize the implications of a growing PV market through offering PV specific training. Most importantly, demand for PV will be greatly enhanced through end-user targeted marketing campaigns. A 1997 SHS leasing system in Swaziland sought to increase consumer awareness in rural areas through the following methods: - Traveling road shows with a drama group - Media-supported public service messages including television, radio, newspaper, brochures in strategic places (electronics shops, general stores) and leaflets dropped from airplanes. - Door-to-door sales - Working demonstrations at community gatherings Among these methods, media supported advertising combined with professional recommendations were reported as having the greatest impact on sales. Working demonstrations were also noted as a useful tool in awareness raising. However, although these efforts proved useful in transferring information, their effectiveness in increasing sales was difficult to determine.39 The cost of such efforts may prove a significant constraint, particularly among programs with a limited budget. In the above-mentioned case study, high-cost mediums such as airplanes and television spots were available due to a high degree of investment from PV installation and maintenance companies. Consumer education is a necessary component of information related capacity strengthening in rural PV dissemination. The consumer has a high incentive to maintain and repair his/her own system due to geographical conditions, inadequate dealer networks and technician costs. The supplier benefits from improved consumer skill levels as the costs of providing technical expertise to remote areas are decreased. In addition, through extending adequate after-purchase training, the supplier ensures that the consumer has appropriate expectations for the system and thus is more likely to be satisfied with overall performance. In a 1998 survey assessing SHS consumer satisfaction in Namibia, failure to address inaccurate expectations was found to be a primary constraint.40 Supplier directed end-user education efforts will likely require support in formulating appropriate training materials and suppliers may be less likely to fund such programs in communities with nearby technical expertise. In addition, if the supplier has little interest in honoring warranties or service contracts, there exists no incentive to invest in appropriate training efforts.41 Finally, such efforts require a trained and monitored body of technicians, must be directed at those most likely to be responsible for maintaining the system, and, in the case of troubleshooting manuals, must be available in the range of local languages and include clear diagrams for those with little reading comprehension.42

IEA-PVPS T9-03:2003. “PV for Rural Electrification in Developing Countries-A Guide to Capacity Building Requirements.” p. 24 40 Ibid p. 25 41 In financing arrangements with little after-purchase engagement, such as cash sales, the supplier has little incentive (aside from word of mouth) to provide appropriate customer service. 42 Acker and Kammen, p.107


27 3:3 Notable Best Practices in Capacity Building Conceptualizing Assessment-Implementation-Evaluation The capacity-strengthening framework as outlined by the UNDP and IEA advocate similar formulas. Both agencies assign the government the roles of formulating and implementing regulatory policies and enforcement, thereby nurturing a business environment free of counterfeit goods while supporting product and technical standards. NGOs are viewed as holding technical and managerial expertise and, as such, are assigned the responsibilities of project implementation and information transfer among the informal sector in addition to supporting cooperation within and between sectors.43 Private firms are responsible for capacity strengthening initiatives within the formal sector.44 The IEA has identified six stakeholders in capacity strengthening for PV dissemination: 1) Government Bodies 2) Utility Sector 3) Financial Community 4) NGOs 5) Service Delivery Chain and 6) End-Users. An assessment of existing capacity is necessary prior to program implementation and must consider state, regional, and local contexts. In particular, assessments must identify existing local infrastructure in addition to recognizing the social values and norms held by participants. Assessments must take into account the flexibility necessary to turn a blueprint into a project and consider the potential challenges presented by a locality. An assessment framework as outlined by both organizations seeks to consider the following: a) The overall context for capacity strengthening: A holistic consideration of local conditions including social, political, economic, and cultural. Such inquiries may include: an assessment of political commitment, the likelihood of “brain drain” as technical skills increase, the capacity of the labor market to support trained technicians, existing technical and training standardizations, local health conditions (particularly the prevalence of HIV/AIDS), local climate conditions, political stability, local power relations and the role of women, and local perceptions of educated individuals. b) The policy and institutional environment of the public sector: An emphasis on financing, judicial, and regulatory bodies within the public sector, economic management, and performance incentive systems. In addition, the viability of certification and product standardization efforts should be evaluated. Such a review may indicate necessary alterations in public sector/donor practices. As with all issues of public policy, government involvement is a function of political will and capacity-strengthening strategies may challenge vested power relationships. As such, the assessment should consider the current political environment and strategies directed accordingly. c) The framework of implementing institutions: An assessment of inter-organizational communication and efficiency and the management structures within each body. Instances of prior coordination within and across industries are of particular importance. d) The availability of technically trained personnel: An assessment of personnel qualifications, training programs, and incentive structures. To what extent will foreign technical capacity be relied upon?
43 44

IEA-PVPS T9-03:2003, p. 1 United Nations Development Programme. “Capacity Development: Lessons of Experience and Guiding Principles.” 2001

28 e) The prioritization of training needs and measures: Determining where capacity strengthening is necessary and the types of training conducive to each organization. f) The formulation of a training program: An assessment of in-country training organizations and technical expertise and a framework for incorporating education programs in the implementation process. Following the assessment of the development context, strategies for capacity strengthening among actors and institutions must be designed and implemented. Policy and regulation efforts must be coordinated with the implementing institutions. Depending on the project, the organization and the staff, different training approaches may prove appropriate. Training may vary from basic packages to more elaborate seminars, practicum, guidebooks or extensive courses. Having local professionals among the trainees will also prove to be very practical for the future maintenance of the system. Commercial bodies, particularly as they posses technical knowledge, may be included, although care should be taken to ensure that multiple commercial interests are represented to prevent a monopoly on information. Commercial interests also represent possible service alternatives and may serve to reinforce a competitive environment among providers. Through case-study analysis, the UNDP has identified three key strategies for personnel capacity strengthening.45 a) Recruitment: Open and competitive recruitment, including interview and technical skill requirements is clearly associated with improved performance. Individual perceptions of ownership in the institution were higher among recruits who had undergone stringent hiring processes and who had been recruited by the organization itself. b) Utilization: The case studies further demonstrated that personnel with advanced training are often underutilized through poor management systems, failure to recognize experience and performance, and abundant administrative duties. c) Retention: Performance recognition, competitive salary, and job satisfaction were identified as key aspects of personnel retention. Institutional loyalty and a sense of ownership were shown to successfully motivated personnel in instances where salaries remained comparatively low. In comparison, environments in which factors other than performance determined advancement were shown to have negative consequences on employee retention. Strategies may be directed at reinforcing or re-inventing institutions, should provide a flexible long-term framework, and must involve stakeholders at all levels. Financial resources for initial and subsequent training must be secured and existing seminars, re-certification requirements, and licensing programs may provide existing training forums. Monitoring and evaluation are particularly difficult as capacity strengthening is highly qualitative in nature. Measures must be assigned in the assessment phase for determining the extent to which strategies have increased, decreased, or did not alter institutional and individual capacity. Due to the dependent nature of the implementing organizations, evaluation efforts should encompass all bodies within the institutional framework of a project. Private and public performance must be evaluated alongside one another, as their partnerships are mutually

UNDP, 2001

29 dependent. Evaluation should also include a consideration of unintended externalities and recommend policy changes within the program.46 Although this framework outlines a valid project design/implementation process, several weaknesses remain. First, this model clearly emphasizes assessment and implementation while providing little insight into the monitoring and evaluation process. This is an all-to-common practice among international development initiatives. Due to the failure of many projects to secure the commitment and funding necessary for detailed evaluation, lessons and insights of incalculable value remain unrecorded. Second, this framework recommends that the implementing agency gather a herculean array of contextual information prior to organizing an initiative. Although such data collection efforts would be ideal, they will prove costly and time consuming. Alternatively, an organization wishing to gather such information should consider partnering with those organizations already operating within country. Too often, development practitioners seek to reinvent the wheel and forming operational partnerships is a viable tool for decreasing information costs. Third, the role of AIDS in reshaping the social, cultural and economic contexts in Africa cannot be underestimated. Fourth, improving individual stakeholder perceptions of ownership in a project is a low-cost method of gaining dedicated local support. Employees and volunteers are more willing to devote time and energy to an initiative when they feel as though their opinions and efforts are valued and are moving the project in a positive direction. Finally, efforts to coordinate among agencies or stakeholders will invariably be stymied by existing power relations. Securing the support of local elites (or operating in a non-threatening manner) is necessary for project longevity. Supporting Sustainability Action-inquiry refers to a design process intended to support the sustainability of an organization through a cyclical process of planning, acting, observing and concluding. In particular, actioninquiry assumes the existence of a continual challenge and seeks to formulate a responsive and dynamic problem-solving environment. Recognizing the value of such characteristics for capacity strengthening initiatives, many in the development community have attempted to apply action-inquiry to practical development challenges with varying degrees of success. This section will consider the lessons learned from one such action-inquiry based initiative that succeeded in creating self-supporting capacity strengthening structures. The initiative was begun in 1990 and supported by the Consultative Group on International Agricultural Research (CGIAR)47. The underlying lesson of this case study is that action-inquiry strategies, although labor intensive, are a highly useful method of capacity strengthening.48 1) The project design process requires considerable negotiation. In the process of formulating a project design, concessions will be necessary to gain stakeholder support and ownership.
46 47

UNDP, 2001 For the sake of brevity, the particulars of the case study will not be included and the reader is recommended to consult the source. Horton, Douglas. “Building Capacity in Planning, Monitoring and Evaluation: Lessons from the Field”. Knowledge, Technology, & Policy, winter 1999, Vol. 11, No. 4, pp 152-188 48 Although we have argued above that strategies must be program specific and that what works in one program is not likely to work in another, we feel that these recommendations provide a general direction for capacity strengthening initiatives and do not constitute a rigid framework.

30 2) Project design is a continual process and extends beyond implementation. In the CGIAR case, following the implementation of the project, local professionals and project managers were invited to attend workshops and discussed how capacity-strengthening efforts could be tailored to meet their needs. In addition, these professionals expressed interest in preparing and delivering training materials. To support these wishes, complementary funding was secured from another donor. The workshops continued on a regular schedule to review progress and plan accordingly and are noted by the case-study author as being a primary source of innovation. 3) Capacity-strengthening should emphasize flexibility. Workshop participants requested that theoretical frameworks of capacity strengthening be overlooked in the training curriculum in favor of proven, responsive methods for dealing with complex issues in an environment of uncertainty. 4) In-country collaborators must be viewed as equals. According to Horton, a local professional remarked, “We don’t want you to “help” or “train” us. We want to work together as colleagues, as equals.”49 5) Assessments do not occur in a value-free environment. Due to the inquiring nature of assessment efforts, stakeholders may feel threatened and prove unwilling to cooperate. Care must be taken to collect information in a tactful manner and perceptions must be tempered by additional sources of information. 6) Beneficiaries must have a meaningful role in designing capacity-strengthening initiatives. This is an obvious recommendation but is rarely given due consideration. In the CGIAR case, some expatriate and local specialists assumed that this process would be a waste of time. Post-project analysis revealed that beneficiary involvement in the design process provided motivation to attend training classes and was instrumental in empowering participants to continue the assessment/planning process. The CGIAR project provides an excellent example of a capacity strengthening initiative that emphasizes community participation, sustainability and local knowledge. Through identifying endogenous priorities and promoting native experts, this project secured local ownership and addressed valid needs. The formulation of critical management skills was supported through empowering native stakeholders to assess and redirect the program. As a whole, the project allowed those involved to determine the underlying expectations and values. However, the weakness behind such a process lies within the funding vehicle. Rarely is aid funding provided in an expectation-free environment. A process that prioritizes participant identified outcomes will likely prove undesirable to a donor with a set expectation for how the funding is to be spent. Also, care must be taken to ensure that, when participants are interviewed, the interviewer contacts an appropriate cross-section of various stakeholders. This must include marginalized and elite populations.


Horton, 168

31 3:4 Key Capacity Strengthening Findings from the Kenyan Example The field research has demonstrated that capacity strengthening is the key to furthering the Kenyan solar market. In particular, we will argue that raising customer awareness of technological capacity and compatibility will lead to an improvement in the skills of informal technicians.50 Contrary to prior capacity strengthening projects in the region, we do not advocate directing training programs towards informal technicians as the incentives for participation will remain marginal until customer understanding improves. We believe that correcting demand side information market failures is a necessary first-step in expanding this market.

Finding 1: Efforts Aimed at Training Informal Technicians are Misdirected Kenya benefits from an extensive network of informal technicians, the Jua Kali. These laborers are a necessary component of both the formal and informal markets and PV dissemination has relied heavily upon Jua Kali technicians, particularly in rural areas.51 Due to the prohibitive costs of operating an extensive network of certified technicians in remote areas, many vendors chose to sub-contract regional and village electricians for installation and maintenance services. In addition, our research suggests that most customers forego utilizing vendor-associated technicians in the installation process, preferring to save money by instead paying the near-by handyman. The primary drawback to the prominence of Jua Kali labor in the PV market is their lack of PV specific knowledge. In particular, our research suggests that there exists a poor understanding of the compatibility and capacity of solar specific electrical components. We found that when the consumer purchases all system components from a vendor, the informal technician is highly likely to perform a satisfactory installation and such systems tend to operate according to expectations. However, when the technician is responsible for providing inputs such as light fixtures and wiring, cost is often prioritized over quality and/or compatibility and system performance suffers.52 A number of capacity strengthening initiatives have been directed at increasing the PV specific skills of Jua Kali technicians and these projects have shared similar constraints. First, Jua Kali technicians had little incentive to undergo skill specific training because the majority of

In this context, we will rely upon the terms “capacity” and “compatibility” in reference to technological inputs. Capacity will refer to the electrical loads that a component is designed to carry. Compatibility refers to the design parameters of a component in relation to other PV components. For example, a charge controller must be used with an appropriate grade of wiring to ensure optimal performance and longevity. In this context, we will say that the charge controller is compatible with the appropriate wiring. When incompatible components are connected, system longevity is threatened. In addition, when the components fail, they often fail catastrophically and cannot be repaired. Hence, incompatibility carries system-wide costs. 51 For a discussion of how Jua kali technicians have supported the dissemination of new technologies, please consult: Kammen, Daniel M. “Cookstoves for the Developing World.” Scientific American, 1995. 52 During the 1980’s the Jua Kali played an active role in the Kenyan government’s efforts to disseminate improved efficiency cook-stoves (known as “jikos”). Recognizing the capacity of the informal technicians to meet consumer demand for easily replicated goods, implementing agencies distributed leaflets explaining how to fabricate the stoves from basic materials. In this manner, informal craftspersons were able to supply the market surge in consumer demand for the improved efficiency stoves.

32 consumers lack the understanding of PV components to demand compatible and appropriate capacity inputs. Second, efforts to organize demonstration sites and formal training opportunities face significant geographical barriers as informal PV technicians operate throughout the country. Finally, although vendors may provide employment incentives to informal technicians with PV specific training, the market for inexpensive installation and maintenance is sufficient to support independent actors. Recognizing the constraints facing training initiatives directed towards informal technicians, we recommend that such initiatives should focus on demand side capacity strengthening. Through awareness raising and education initiatives, increases in consumer understanding of the capabilities of SHS and the technologies involved will support a corresponding increase in the skill levels of informal technicians responding to market conditions.53 This process could be conceptualized as having two phases. The first would seek to expand consumer awareness and product familiarity and would require the development of measures for the quantitative analysis of these issues. We believe the increase in consumer information will create an incentive for informal technicians to provide improved services. Technicians will then require, and seek out, additional training. However, the tendency to share information in informal circles could lead to the spread of misinformation and the second phase would seek to provide NGO and vendor supported training opportunities. Such initiatives could be as informal as distributing leaflets and brochures or as formal as a classroom environment. Regardless of the initiatives utilized, we believe that standardization of the informal technicians in the Kenyan PV market should be avoided. Informal technicians provide a necessary service and have demonstrated the ability to install and maintain systems when provided with the appropriate materials. Although larger scale SHS programs (<200W) require a standardized installation process to facilitate maintenance and repair, the average rural SHS system does not require extensive installation. To standardize their practices or require certified training programs would not only ignore the existing incentive structure, it would also require a monitoring process well beyond the government’s current capacity.54 In contrast, the capacity strengthening literature recommends industry-wide technical standardization and training programs.

Although our findings suggest that customers lack a high incentive to increase their knowledge of SHS, they also lack a disincentive to do so. Consequently, if such information were to be made easily accessible, non-biased and user friendly, the consumer may prove more receptive. We recommend that viable demand side capacity strengthening initiatives are a necessary topic for further research. Color-coding components according to compatibility and capacity may prove a viable method for overcoming the confusion associated with a wide-variety of technological inputs. 54 For an excellent consideration of informal capacity strengthening in Kenya, please consult: Jacobson, Arne. “The Market for Micro-Power: Social Uses of Solar Electricity in Rural Kenya.” Energy and Resources Group, University of California, Berkley. 3-15-04


Informal Intuition: The case of Daniel Jocombo Daniel Jocombo is an elderly farmer who lives with his wife in rural Molo. Mr. Jacombo spent time in formal schooling and worked for a period as farm supervisor. He has learned many skills over the years including electrical, carpentry, and metal work. Before his children moved to the U.S., they purchased him a cell-phone. However, in order to charge the phone, he had to travel several kilometers and eventually he purchased a car battery so that he might recharge it at home. Carrying this car battery to the nearest recharging facility proved a weighty task. He created a pedal powered machine to recharge the battery. His children didn’t like the idea of their farther pedaling this contraption in his old age so they sent him funds to purchase a solar panel. Once installed, his system was able to power his television, two light bulbs, and could recharge his phone. However, during the rainy season, he spent the majority of his time indoors watching television and his system kept running down. He still recharges the battery with the pedal powered machine when necessary.

Finding 2: Limited NGO Support Information dissemination efforts and stakeholder coordination are hampered by the lack of a dedicated network of solar specific NGOs. Indeed, the largest solar market in Africa is supported by one solar-specific NGO. As a result, private vendors are the overwhelming source of consumer education and stakeholder training programs and the research findings suggest that the quality of such programs varies widely. In contrast to the capacity strengthening literature, the Kenyan PV market has grown despite the lack of formal NGOs providing information to consumers. Consumers rely upon several sources for solar-related information. As argued by Acker and Kammen, word of mouth remains the most significant source of consumer information transfer among consumers. Although such a system provides incentives for vendors to provide high quality products and service, misinformation remains a significant barrier to education efforts, particularly as they relate to overzealous expectations. Recognizing this, several vendors have implemented a similar rural village marketing scheme. Upon finding a community with little to no experience in SHS, the vendor provides a prominent family with a complete system for a discounted price. The vendor then ensures that the newly electrified household receives exemplary customer service. In so doing, the vendor relies upon word of mouth to sell the systems. Our research further indicated that although vendor supported consumer education programs exist, they are predominantly directed at the consumer when he/she enters the store to complete a purchase. In this environment, the level of information presented to the consumer depends upon the salesperson, the size of the purchase and the interest of the consumer. Although the major vendors in the solar market maintain urban offices throughout the country, they must contract rural electrical shops to sell and maintain their products. Consequently, rural dealerships often carry name-brand products but lack the technological capacity to educate consumers appropriately.55 Purchase size was also a factor in the quality of maintenance training provided to consumers. Vendors correctly view large SHS projects as a marketing opportunity and seek to provide the consumer with satisfactory service. Research did not indicate that consumers making larger purchases tended to place a higher value on receiving proper maintenance training; rather, the salespersons tended to be more forthcoming with maintenance related

In one instance, solar components were being sold in a grocery store and no supplemental information or expertise was available.

34 information.56 According to the owner of one solar franchise, salespersons prioritize closing a sale and thus may not prove the ideal vehicles for consumer education.57 Finding 3: Poor Coordination The lack of effective coordination within the solar industry and with associated institutions (financing, regulatory and development) represents a primary constraint to information transfer. Through the interview process, we found considerable misinformation regarding micro-credit options. According to several vendors, a prominent financial institution was providing a solarspecific group loan that customers have found to be particularly useful. However, when we inquired at the financial institution, the representative informed us that the loan program in question had been discontinued for a number of years due to lack of interest. Although the solar specific NGO is intended to coordinate activities among private actors in the solar market, there is no effective cross-sector dialogue and additional NGOs are necessary. Although incentives for NGOs to become involved are limited, the high reliance of NGOs with rural offices on PV suggests that the non-profit community is aware of the technology and would benefit from improvements in rural technical servicing and quality control. Another example of poor coordination relates to the government sponsored Rural Electrification Programme (REP). This initiative provides highly subsidized grid-based electricity to rural communities living near existing grid facilities. Although SHS represents a viable alternative to the expensive, infrastructure dependent and unreliable electricity grid, government subsidies deflate the cost of REP supported energy and provide SHS users an incentive to discontinue SHS use. An interview with a former SHS end-user highlighted the benefits of grid-based electricity: it is cheap (due to subsidies), more consistent than SHS during periods of foul weather and has the voltage to power several large appliances at the same time. She did note that blackouts were a common problem, but felt that the advantages of grid-based electricity were sufficient to counter such issues. When one considers the potential for widespread PV dissemination in rural areas and the aggregate cost associated with grid extension and fossil fuel based electricity production, the benefits of supporting SHS dissemination for rural electrification are obvious. Finding 4: Alternative Uses for PV Our research revealed several uses for PV technologies that we had not considered. These examples illustrate the broad applications of such technologies. We were quite surprised to find that the average SHS consumer purchases a system for the primary purpose of powering televisions and radios. In a number of instances, dealers were marketing SHS packages designed expressly for a black and white direct current television set. We also found PV being utilized to power water pumps and to dry food for the rainy season. Finally, we did encounter a number of

In one instance, the researchers were able to interview a salesperson and a farmer who had just completed the purchase of a 14W SHS. When asked if he informed consumers as to the necessary components of SHS, the salesperson claimed that doing so was one of his priorities because it ensures consumer satisfaction. The customer was unaware of the need for a charge controller, a component that protects the battery (Meru interviews. 3-20-04). This demonstrates how the salesperson may alter their sales pitch depending upon their perception of the customers needs. Our research also suggests that consumer mistakes due to inadequate information transfer are a common source of power failure. 57 Interview with Charles Rioba. 3-16-04

35 examples of consumers using PV for income-generating opportunities. These include roadside barbershops, community movie theaters and kiosks playing loud music to attract customers.

36 CHAPTER 4 QUALITY ASSURANCE 4:1 Quality Assurance and PV Rural Electrification Quality assurance is a challenging area, and failing to give it the adequate attention has caused many PV programs to fail in the past. It is a very important area to any rural electrification project, as it affects and/or benefits a large number of different stakeholders; it also has an influence on financing mechanisms, and on the success of the project. Quality assurance spreads across PV project design and implementation, but certainly has to be taken into account early to avoid costly mishaps. All leaders in the field of solar energy recognize the importance of quality assurance in effectively conducting PV rural electrification, and the need to make related and relevant provisions early at the design stage. Quality assurance is more than just making sure that the equipment works. It is also ensuring that qualified personnel is available to train people to properly operate their systems, and create the manpower that will assure maintenance for the life of the project. Quality assurance also encompasses quality management of the project, and adherence to some specific quality standards, whether those are related to quality management systems or hardware and training certification. Quality assurance for PV or any other form of rural electrification encompasses three main areas: quality management, hardware quality and training and practitioner quality.58 Customers are naturally those most affected by the quality component of the project, yet they have the least capability to evaluate it. Low quality equipment and/or installation will all mean one thing to the customer: a non-functioning system; this could cause them to lose revenue if they were using their system for income-generating activities. Ensuring quality in a PV/SHS project will guarantee that customers are not left to choose their systems based on lowest cost, which may very well be synonymous with low quality.59 End-users are not the sole beneficiary of rigorous quality assurance. The International Energy Agency states that all stakeholders in a PV rural electrification project are directly affected and identifies three additional core beneficiaries in the (1) finance sector, (2) government ministries, and the (3) service delivery chain.60 (1) Financial Sector is naturally indisposed to taking risks, and PV rural electrification is mainly directed at rural populations that are generally poorer, and less able to provide collateral for loans. The growth of PV in the face of the role energy is to play in helping achieve the millennium development goals will however put some pressure on the finance sector to get involved in making modern renewable energy services available.61 The use of quality management systems will be practical in assessing risk when appraising loans and investment prospects for PV equipments and projects. The finance sector will help in pushing the PV
IEA-PVPS Report Task 9-04:2003. “The Role of Quality Management, Hardware Quality and Accredited Training in PV Programmes in Developing Countries.” p.1 59 JP Ross, Certification of Off-grid Renewable Energy Systems, Center for Resource Solutions, September 29, 2001. 60 Ibid, p.3-5 61 Energy for the Poor: Underpinning the Millennium Development Goals published by Department for International Development (DFID) May 2002, p.2-3

37 business and relevant government bodies to provide the criteria (quality standards/ certification for hardware, training and practitioners) against which it will consider organizations and/or people applying for the loans or linked to them in some way (installers, maintenance, etc.). These standards, along with use of consulting experts, are the only way the financial sector, which is generally no technical expert in PV, has to assess applicants and approve loans. They will also be helpful to development organizations providing financial services, as their technical experts can additionally rely on certification to select certain contractors.62 Let us make an important note here that this argument, though it makes sense, may not be realistic in the developing world context where weak institutional capacity will render the existence and/or enforcement of such standards unlikely. Therefore, this argument to rally the support of the financial sector may only work in countries were a standardization office is in place and functional. (2) Government Ministries that are involved in the design and/or implementation of rural energy programs usually provide some financial support, and are accountable to both funding entities and beneficiaries of the programs. Ensuring the success of these programs and their long-term sustainability is in their best interest, as they will justify and potentially secure additional/future funds easier if they can show positive results. Incorporating quality standards, including requirements for certifications and accreditations where appropriate in their contracting will help them better guarantee success for their programs.63 (3) At first glance, The Service Delivery Chain will bear added cost by undertaking quality assurance efforts, whether it would be through supporting accreditation and certification processes or conducting consumer awareness-raising. However, taking on such actions will lead to added profits in the long run. Indeed, the use of certified equipment and accredited personnel would lead to more reliable service. The latter will create less forecasted repairs, more satisfied customers and referrals, and thus added revenues and profits.64 Quality Assurance in Management, Hardware and Training Three sectors of quality assurance were mentioned earlier, and similar recommendations are found across various agencies on how to bring quality assurance to each area. Quality management deals with the operational processes of all organizations engaged in the project (from those in charge of PV installation and hardware supply to technical specialists, financial organizations and service providers). IEA-PVPS task 9 and other leader agencies in the renewable energy field (i.e. IEC, IECQS, ISO, PV GAP, etc.) strongly advocate the adoption of quality management systems to help keep quality assurance at the design phase, and quality control through implementation, on track. In brief, a QMS is keeping excellent track of all proceedings and related stakeholder-actions and records, doing so in a cohesive way, embraced and endorsed by all parties within a company, and ideally certified by an external audit from a recognized international body (usually ISO). QMS
62 63

IEA-PVPS Report Task 9-04:2003, p.3 Ibid, p.4 64 Ibid, p.4-5

38 provides a procedure for constant evaluation, which can check processes within an organization, whether these are manufacturing or installation practices, for reliability, quality of outcomes, and success of improvement measures. Moreover, QMS guarantees that the records needed to assess any part of the process are not just accessible but also auditable. Quality management systems (QMS), though often associated with large corporations, are just as valuable for smaller companies, which will however need significant support in capital and know-how due to their size and lack of experience in dealing with such systems. Because implementing QMS can take up to some years, this is a challenging task for development aid projects65. At any rate, QMS are a valuable tool when handled properly, and a good framework within which organizations function and that can be used to evaluate their performance. A complete review of QMS can be read in the PVPS Report Task 9-04 on Quality Assurance for PV Rural Electrification in developing countries.66 The reputable operational reliability and technical maturity of PV technology really only applies to functions in more developed countries, where high-quality products and after-sale services are primed. One main challenge in implementing PV technology in developing countries is in the unreliability of some elements of hardware, particularly when these are made locally.67 It is important that project designers know that an initial lower investment in low quality systems can lead to much higher operational costs down the line, in addition to a decreased consumer confidence level and willingness to pay. The technical issues associated with PV rural electrification projects need to be given just as much importance as those non-technical (i.e. financing, capacity strengthening, etc.) concerns to get a suitable project. Ensuring the quality of the hardware will protect the customers and reduce the potential commercial risks for suppliers. Among the most efficient approaches to lessen technical risk is the adoption and use of nationally or internationally established standards. However, despite the increasing prevalence of some standards for PV components, international standards on testing the balance of system components are by and large lacking. The global efforts do exist however, through agencies such as the IEC and PV GAP, to develop a number of standards for PV systems and parts. In the meantime, various organizations are producing a string of “interim standards” to be acknowledged by the IEC while many national programs are creating national and/or regional standards. Details on these procedures and standards can be found in various IEA task force publications.68 It is project designers’ responsibility to find out what certifications and standards are available for hardware, and require that potential suppliers uphold them as a condition for consideration. Suppliers who fail to offer certified hardware should be penalized, and these penalties should be made clear in terms of agreements worked out at the design phase. Standardization is again a strong recommendation of different IEA task forces when it comes to the quality of training and practitioners in the field of PV rural electrification for developing
65 66

IEA-PVPS Report Task 9-04, p.6 Ibid, available online [] 67 Ibid, p.12 68 IEA-PVPS Report Task 9-04 and IEA PVPS Task 3 Report T3-07: 2000. “Survey of National and International Standards, Guidelines & QA Procedures for Stand-alone PV Systems.” Available from the IEA PVPS website []

39 countries. Indeed, the IEA asserts that due to the decentralized nature of rural electrification, ensuring that the people responsible for installing PV systems, which most likely will include a wide number of SHS in rural settings, are adequately trained to carry out such tasks is vital. Failing to do so would lead to inadequate installations that will undermine the lifetime of the systems, and decrease trust in the technology, causing a serious impediment to a worldwide acceleration in the market development of decentralized technologies such as solar electrification.69 Coordination efforts between government, international and local development agencies are also deemed crucial for the success of such training programs, and such coordination has by and large been lacking in the past, leading to failing or ineffective programs.70 Additional conditions for the success of training accreditation and practitioner certification programs, as identified by the IEA, are: the backing of the industry such programs represent, reliability to funding sources, participating groups and the public sector, a proven benefit to all stakeholders, a solid basis on existing standards, an all-encompassing chain of responsibility that expand from the highest national or international standards and bodies down to the smaller participating groups and endusers.71 Finally, for any attempt to establish a quality system for training accreditation and certification, it will be important to review international standards, assess the requirements based on local needs and also evaluate the effects of executing the quality management system. Furthermore, having adequate standards bodies to ensure the sustainability of practitioner certification programs and to enforce adopted standards is a required step.72 This last point will prove very challenging in developing countries, as it requires substantial institutional capacity, which is often times lacking in these markets. Indeed, realities in the field most times show that standardization bodies are either inexistent or worthless in developing countries, which makes these strong recommendations about standardization, accreditation and certification quite ambitious, and even unrealistic, for developing countries’ markets. Rather, a thorough study of specific conditions in these markets should be done to find a viable alternative to standardization processes, which can take years to implement. 4:2 Analysis of Quality Assurance At first glance, we had a number of questions and concerns about linkages between quality assurance and other areas of the project, and about potential and existing market failures associated with quality assurance in PV rural electrification. It is apparent that poor quality would not only affect customers and other stakeholders, but also financing. Financing can also have some effect on the quality level of equipment. Additionally, some market failures can be identified and linked to quality assurance (i.e. lack of information, geographical access), and affect the project’s outcome or success level. Quality Assurance and links to other parts of PV Rural Electrification Projects
69 70

Ibid. IEA-PVPS Report Task 9-04 71 Ibid. 72 Ibid.


Firstly, It seems that the financing option(s) chosen for rural electrification could directly affect certain quality factors. In a cash-and-carry situation, the equipment sold is paid for entirely in one transaction, and rural customers, if not given adequate information on how to choose appropriate systems, will most likely go by prices and choose low cost products. Depending on what type of vendor they dealt with (small, low-capacity vendor or large, high-capacity retailer/store), customers may have real problems down the line should their systems fail and require maintenance. Low cost systems may not even come with a warranty, let alone a pledge to quality, and in that case customers are left to deal with losses and negative consequences alone. Furthermore, privately bought solar home systems, even if they are of high quality, will often be taken home and installed improperly by the customer or other unqualified party (if the vendor in this case only takes care of the sale and not the installation of the equipment), further exacerbating the risk for inadequate performance and sometimes associated hazards.73 However, in a rental/leasing/fee-for-service situation, the company providing solar electrification services owns the equipment. Maintaining the quality and high performance status of the equipment then becomes a vital condition to preserve working assets, and to ensure customers respect of payments. Companies leasing their equipment will most likely put more effort and investments into certified material, accredited staff, and ongoing maintenance; customers of such companies will probably have higher satisfaction, due to the reliability of their systems. John H. Rogers, Vice-President of SOLUZ, Inc.74, confirmed that the rental option, which generally brings enhanced quality assurance, has proven the preferred method for clients in their PV rural electrification projects in Honduras and the Dominican Republic. Clients favor renting equipment while paying for services better because this route warrants less commitment, lower cost, and reliable maintenance. Secondly, quality, or lack thereof, will also affect the future of PV or SHS dissemination in some target areas. Rural communities are usually rather small and interconnected; people know each other and share experiences. If some people got the systems first, and turn out to be satisfied customers, they can be expected to act as a “walking ad” for the technology, and will recommend it to people around them. Unsatisfied customers who have experienced the quandaries of failing, improperly working, and/or low performance systems will have the opposite effect. Sharing their negative experience could create a sense of “distrust” for the technology, when this experience may have only been related to the lack of quality equipment from a specific supplier. The non-users, who are the remaining target population, may thus be reticent to give up whatever alternative energy sources they may be currently using, but which is reliable enough for them, to adopt another that they know for sure has failed its customers, and possibly brought them losses. Geography Market Failure A primary reason for bringing PV electrification to rural areas of the developing world is that they are so remote those conventional grids do not reach them, and PV and SHS technology

Eduardo Lorenzo. “In the Field: Realities of some PV Rural Electrification Projects.” Renewable Energy World, Sept-Oct 2000 Vol.3 No.5 74 SOLUZ, Inc. is a business and technology development company commercializing distributed micro-power, primarily photovoltaic (PV), for rural areas in developing countries []

41 would provide what the grid is depriving them of: electricity. But this remoteness also has its potential problems for PV rural electrification, primarily with upholding the quality assurance of the project. IEA-PVPS task 9 states that equipment sold should be certified and should offer warranty to be deemed quality-equipment.75 The use of warranties and guaranties will also provide further protection to customers.76 The questions that come to mind are: How do endusers contact suppliers to receive maintenance when their system fails? How long does it take the supplier to respond to maintenance requests? Does the supplier have the capacity to satisfy maintenance calls efficiently and rapidly? In other words, what provisions are made at the design level to ensure that the quality assurance advocated is properly upheld, especially for remote customers? These questions also relate to the training issues: who will be trained to maintain systems? Where will they live? Will they be part of the community, in which case their services are more accessible and prompter? Will the supplier have a representing unit within or near the community to facilitate communication? These aspects of delivering quality need to be taken into account and worked in the design of the project to allow for smooth implementation. These are not questions to be dealt with when they arise, but scenarios to be considered and solved with appropriate provisions beforehand, has quality issues left unattended can prove more costly later. When designing the solar rural electrification projects, it is important to assure that remote clients are not serviced once, and then left to deal with potential technical defects, which can negatively affect them. Ways in which end-users will contact suppliers have to be facilitated, and there need to be some serious provisions relating to prompt service and sustained access and communication between suppliers and customers; having intermediaries (sub-units, representatives, technicians, etc.) at proximity of different sites could be a solution. Later, in the section on best practices, we will show how SOLUZ, Inc. has taken into account the possibility of geographical market failure early in designing their PV rural electrification projects in Honduras and the Dominican Republic, and has put provisions into place to ensure smooth implementation, and protect their remote customers. Information Market Failure Information on quality of products and practitioners has a great influence on the market and consumers. It is vital that PV projects are designed to ensure the adequate education of all stakeholders about various parts of the project, and end-users have to be given a certain amount of information, which will make them “well-informed consumers”. In the case of SHS, education on some simple adequate self-installation tips needs to be available, as too often offthe-shelf systems may be of high quality, but become low performing equipment when subjected to poor installation (either by customer of untrained or inadequately trained third party).77
75 76

IEA-PVPS Report Task 9-04:2003 IEA footnote: More direct protection against commercial risk is essentially a function of the deployment model used – cash sales, credit/leasing, service provision etc, as well as the details of the delivery/supply chain. A number of programs have considered ways to minimize the commercial risks inherent in delivery of PV technology in rural areas of developing countries. 77 Eduardo Lorenzo, “In the field: Realities of some PV rural electrification projects.” Renewable Energy World, Sept-Oct 2000, vol.3 No.5


Most economics supply and demand models assume ceteris paribus (all other things held equal) or perfect information to work as predicted, but this is not an assumption that can be made in most developing countries’ PV markets. This is one aspect that makes the SHS/PV market so different. The Kenyan example presented by R.D. Duke et al. in a study of PV module quality in the country’s SHS market in 2002 perfectly illustrates this phenomenon.78 According to this study, the majority of SHS bought in Kenya between the early 1990s and 1999, when the study was conducted, utilized small 10 to 14 W amorphous silicon modules (called a-Si). 79 Three major aSi brands were promoted in Kenya, two of which performed satisfactorily. The third however, despite having critical quality problems, had preserved a considerable market share in spite of having a much higher cost per delivered W80. Rural customers in Kenya often pay for the consequences of not knowing enough to adequately choose quality equipment. What is often their most precious and long-lasting possession, which they can use to generate more income, can suddenly become the cause for trouble and losses. Figure 1 illustrates how this direct effect on consumers can lead to a larger economic problem by damaging consumer confidence in PV modules, which will thus confine the broad Kenyan SHS market below the socially optimal level. Imagine that, on average, bad modules only generate half of their rated power while good modules perform at their full rated power. When customers are given perfect information in terms of $ per rated Wp, the demand for the low-quality brands is half that of high-quality brands. With 50 percent of the market share going to low-grade modules, and assuming perfect information about market shares and performance of both types of modules but incapability to distinguish between the two, the pooled demand for the modules will be the average of the demand for each one. However, when we add the risk-aversion factor, the pooled demand will fall closer to that for low-grade modules. The pooled demand will most likely fall further if we consider the perception of households that only low-grade modules would reach their markets. The market share of the high-grade modules will fall, as suppliers overrate these modules in an attempt to compete with the low prices introduced in the market by low-grade modules. Informing consumers about quality issues would eliminate the incentive to overrate modules, as low-grade modules gradually disappear from the market. Although this will increase prices in terms of $ per rated Wp, the true prices in $ per delivered Wp will decrease as overrated modules recede.81

Richard D. Duke, Arne Jacobson, Daniel M. Kammen. “Photovoltaic module quality in the Kenyan solar home systems market.” Energy Policy 30 (2002) p.477-499 79 The output of a solar panel is rated as peak watts, or W. This is the maximum power in watts that the module produces under standard test conditions of one sun, AM 1.5, and T=25oC.For practical reasons, modules are tested and rated under conditions that rarely occur in real life. Instead of a real sun, a rapid flash of light is used. This means that the cells will have no time to warm up at all. In fact such a situation corresponds to an ambient temperature of -5 oC. It is therefore unlikely that the module will ever produce its rated number of peak watts. It is important to take this into account when determining the module output for any given system. [] 80 Ibid, Table 1 shows a study conducted to assess retail price per W for small amorphous silicon (a-Si) and crystalline (x-Si) photovoltaic panels in the Kenyan market. 81 Richard D. Duke, Arne Jacobson, Daniel M. Kammen, Photovoltaic module quality in the Kenyan solar home systems market, Energy Policy 30 (2002) p.477-499


43 Information market failures do not only happen vis-à-vis consumers. In fact, the viability of quality assurance in the PV market, and its integration in project design through implementation seems to depend largely on the ability to generate and effectively convey strong reliable information on quality standards to all stakeholders.

Consequently, one major recommendation reflected in various numbers of manuals, studies, and papers is the adoption of some set quality standards for management, hardware and training/practitioner certification. This concern for quality assurance has prompted numerous international proceedings for the Table 1: Welfare Analysis of the Product Quality Market endorsement of standardization Failure Source: R.D. Duke et al. / Energy Policy 30 (2002) 477-499 and certification techniques. The German Technical Co-operation (GTZ) published a broad assessment of this activity in the year 2000.82 However, despite the efforts, the organization of commonly accepted certification procedures has been encountering intricacies due in large to the decentralized nature of PV/SHS rural electrification, which makes it challenging to gather the essential feedback from the field during the standards development process.83 In the meantime, new projects will have to depend on themselves to incorporate factors such as a quality management system, and implement other types of quality assurance measures. This could be painstaking but needs to be considered early in the design phase to avoid the problems that could emanate from ignoring quality assurance. Another concern should be to guarantee that vendors actually care to inform their customers about quality variability of different brands, and let them make an informed decision, rather than pushing whichever brand they have more chances of selling, generally the cheapest one, regardless of the price or existing quality flaws. This may prove hard to achieve since the principal motivation of vendors is to sell equipment, not to present customers with the highest value for their money. This phenomenon illustrates another type of information market failure: one where the information is known by one party (vendor/supplier) and just withheld out of interest, which is likely to hurt the other party (customer/end-user).
82 83

GTZ Quality Standards for Solar Home Systems and Rural Health Power Supply, Division 44, 2000 Eduardo


The risk of a “brain-drain” effect? One last concern relates to quality of training, and is not really raised by any design-blueprint. There is the possibility of a brain-drain effect once people in rural areas are properly trained and given a new “skilled worker” status. One way to alleviate geographical market failures would be to train people within the community and surrounding areas to conduct maintenance, replacements and repairs. However, these may be people who have always wanted to migrate to urban areas to get better jobs and opportunities. By having new skills, they will recognize the new competitive edge this confers them, and might be tempted to migrate out of the community to a different place where they may be better off and make more income. However, fixing this potential problem should not be hard. Companies training and hiring these employees can stipulate in their contracts that employee are bound to work within a certain area for a certain amount of time in return for the training provided, and this will also guarantee the presence and availability of an adequate workforce, and knowledge of when it will be necessary to hire and train new workers. 4:3 Notable Best Practices in Quality Assurance Some general quality assurance guidelines A good synthesis of recommendations on incorporating quality into designing off-grid PV rural electrification for developing countries is offered by Task 3 of the IEA, and brings together all the crucial issues in the first part that need to be considered to ensure a strong project.84 In summary, the task force emphasizes the need to define roles and responsibilities early in the project’s life, and ensure the three main roles are assigned: Financer, Implementer (where project managers, logistics managers, project engineers, trainers, suppliers and installers fall under), and User. Larger programs will also need an “expert advisor”. Once roles are determined, system ownership also needs to be defined so the responsibility for operation and maintenance is clear. A model of success in implementing quality assurance processes As mentioned earlier, it is forecasted that establishing quality management systems and quality assurance measures may be quite challenging for new projects. However, this is by no means an impossible task, and one excellent archetype of this is the experience of the Regional Solar Programme (RSP) in the Sahel nations of Africa85. With the support of the European Commission and the CILSS (Comité permanent inter Etats de lutte contre la secheresse dans le Sahel), the RSP set up 626 PV water pumping structures, totaling up to 1.2 MW. It is the only known PV rural electrification case that has exhaustively applied a complete quality assurance process86. The author notes that in many instances, practicing and making mistakes were the way to learn in order to reach the current high level of technical quality. Most of the problems

IEA-PVPS Paper Task 3-021: Quality Management of Stand Alone PV systems: Recommended Practices, Alison Wilshaw, Lucy Southgate, and Rolf Oldach, October 2002. 85 Kabore F., Energy for a sustained and social development in the Sahelian region, The Regional Solar Programme, The Yearbook of Renewable Energies 1994, Ponte Press, Bochum, 1994. 86 Eduardo

45 encountered by the RSP were not linked to the basic hardware used of the installations (PV modules and motor pumps) but rather to other components and aspects of the systems (i.e. protection, grounding, housing, boxes, contacts, etc.). The observations and positive outcomes are best illustrated through quotes from the projects’ contractor and a user’s experience featured in the study: “... The EC office awarding the contract imposed strict quality requirements on components and material for this project ... The material and components selected had to pass severe tests performed at the laboratories officially accepted by the CILSS organization ... Further, the fabrication processes of the inverters’ suppliers were audited ... some possibilities for optimizing the devices come to light and were later integrated during production. For instance, the degree of electrical protection was increased from IP54 to IP55, the mechanical mounting of several components was optimized for greater ruggedness ... Although this raised the supplier’s cost, in the end it was justified by the way the components operated; all devices work perfectly in the field and are highly reliable, which is not typically the case...” [Contractor: Siemens Solar] “With a two to five year distance regarding the functioning of the pumps, it can today be affirmed that a high reliability level was reached due to the technical precautions adopted by the programme” [user’s experience quoted by the CILSS].87

The RSP is a very good illustration of the advantages of implementing quality assurance processes. It demonstrates that comprehending the specific local and decentralized nature of the PV rural electrification is also essential to technical quality. Surprisingly, no equivalent to RSP has been found for SHS, in spite of the large number of ongoing operations. Lorenzo (2000) suggests that this may be due to SHS being historically considered as quite simple when this notion is in fact deceptive. In effect, the adequate design and implementation of SHS encounter most of the challenges of any PV system, in addition to the need for adaptation to the differing conditions of each home. SHS actually requires more technical expertise than suggested by its historical notion of simplicity88. The fact that this project stands alone as a successful example of implementing quality assurance processes may mean several things: the cost of such implementation may have been too high for other projects to afford, and it may be the exception that confirms the rule, and its replication to other areas is most likely very limited. Overcoming geographical market failures We have raised the concern earlier that end-users may not be able to enjoy a fully functional maintenance system that will fulfill warranty and other repair issues, as there may be communication barriers raised by the remote nature of PV rural electrification programs. A great example of how these issues may be handled and the barriers overcame is that of SOLUZ, Inc. and its PV rural electrification projects in Honduras and the Dominican Republic. Recognizing that rural customers may not be able to efficiently contact suppliers for repair calls, SOLUZ, Inc. has technicians making very regular rounds in all sites, in addition to having “micro-centers” in all sites that are responsible for collections of monthly payments from rental customers, but also
87 88

Eduardo Ibid.

46 serve has efficient communication links between end-users and the mother company. These “micro-centers” can be any small rural store from retailers to small businesses run in the community, not necessarily related to PV. John H. Rogers, V.P of SOLUZ, Inc. observed that with the rapid spread of cellular technology, and the fierce competition for markets (which would drive prices down), it is very likely that more rural areas will be reached by mobile telephone technology soon, and this will become a great asset to facilitating direct communication between urban dealers and their rural customers89. Of course, another way of overcoming this geographical market failure would be to target people within the community for training and hiring, and have them close to end-users, which would allow for prompt service. As mentioned earlier, the only risk with this is the potential braindrain effect training rural workers would entail. However, the recommendation stands and companies would have to figure contractual ways to keep their trainees within the areas where they are needed. Using advertisement and branding to indicate supplier’s reputation Through efficient advertisement, suppliers can strengthen their own reputation and also give customers more valuable information about the quality and relative ranking of their systems. Figure 2 shows one such example where Free Energy Europe, one supplier in the Kenya SHS market, used positive results about its modules in an independent study to push their product, but also gain the trust of potential buyers. However, it is worth noting that advertisement campaigns targeted to rural areas tend to be more expensive due to the remoteness of these customers, and does not necessarily work as effectively when customers cannot distinguish poor quality, as manufacturers of low grade systems may also carry attractive advertisement, based on pricing or other factors.90 Advertising thus is certainly a good way to communicate quality information to customers but is not a guarantee that this transfer of information will indeed happen. What makes advertisement so uniquely effective for PV dissemination is the technology component of these projects, which means that education and efforts to make consumers understand the technology will facilitate their use of the systems. Advertising can help with such information dissemination. Branding is also an effective way for suppliers to signal their reputation in the market. Companies such as Shell Solar and BP Solar for instance could use this strategy, which can prove to be quite influential since these modules will carry a name that customers already know for their fuel products, and may therefore either associate with quality, or buy because they recognize the name and trust it more than an unknown manufacturer brand. For those smaller equipment manufacturers whose name may not have similar international influence, the use of well-known and respected retailers within the market can be a way to market their product.


Detailed case-studies on these SOLUZ, Inc. projects can be found on the organization’s website [] 90 Ibid.


The challenge for manufacturers and suppliers would have to prove the reliability and high quality of their product, for this retailer will want to retain its good reputation. This will in turns work in favor of higher levels of quality. The challenge for the retailer would be the potential need to develop processes to provide the customers with adequate information about the systems’ use and installation.91 The Power of Social Marketing

Fig.2. Free Energy Europe distributed this bumper sticker in Kenya in 2000, advertising the superior quality of their modules as best tested in Kenya based on testing results reported in a quality study conducted by Richard D. Duke et al. and published that same year [R.D. Duke et al. Energy Policy 30 (2002)]

We have established how important information is in the PV market. Though some would assume that buyers usually conduct some research before buying a product, it seems that for monetary reasons, it is more likely that they might go for the cheapest deal, which may not necessarily be the best quality or value for the money they are willing to spend. Besides, PV cash-sellers do not care about giving the customers the best value for their money more than they care about simply making a sale. Once the product is acquired, users need to know how to care for their systems, properly store units, etc. Therefore, educating the consumers about PV technology and quality is vital and can be done in various ways. It is true that social marketing can be expensive when it comes to rural areas due to several factors such as remoteness, lower levels of literacy (requiring alternatives to print material), need for direct contact, to name a few constraints. However, consumer awareness and marketing strategies are necessary for PV projects as a part of the quality assurance efforts needed. Prior to this, it is importance to gather good understanding of the market and target population. Such understanding can usually be garnered during the preparation phase of a project through needs assessments, and market surveys that are likely to provide crucial information on affordability, current level of knowledge, and household revenues. In designing awareness campaigns, teaching consumers about the quality issues and giving them the power and ability to choose the right system (or best value for their money), as well as educating them on proper usage and conservation to guarantee the life of their system are important. This can be done through information dissemination workshops. Workshops similar to the type we have recommended above are conducted by provincial governments in Argentina as a supporting activity to their concession system. Similarly, in Sri Lanka hired marketing consultants organize village-level workshops nationwide to promote SHS; dealers get to demonstrate the products, users are taught about the technology, and

R.D. Duke et al., Energy Policy 30 (2002), p. 485

48 financing organizations (mostly micro credit) are able to measure the local interest and plan for credit programs.92 Another way to ensure consumer awareness is to use their social networks. Rural populations often live in villages were people know each other, and are often involved together in some sort of organization, grouping, cooperative, etc. Rural consumers can use these networks to their benefit and do research that way about the products quality or lack thereof. Communication between users and non-users should be encouraged; those who have systems can recommend (or warn about) their place of purchase to those shopping for one, etc. Furthermore, the network groups can be used as consumers associations that would conduct group purchases and put pressure on vendors to provide better quality products and warranties for fear of loosing several customers at once instead of one or two. 4:4 Key Quality Assurance Findings from the Kenyan Example Case Study Our research has identified a number of findings related to quality assurance. In particular, we will argue that the pervasiveness of counterfeit products in the Kenyan PV market is due to consumer demand and government failure and serves to undermine public perceptions of SHS. In addition, most warranty programs are hampered by poor consumer incentive. We also argue that formal technician training programs fail to extend beyond the urban areas and lack consistency. Finding 1: Counterfeit products are commonplace and limit market effectiveness. Counterfeit goods represent a significant threat to the Kenyan PV market. Such goods represent an information market failure as customers are, in a sense, punished for purchasing seemingly high-quality components. Counterfeiting in the Kenyan PV market takes the from of brand piracy and false advertising. As brand reliability has increased, several brand names and logos have been pirated for use on counterfeit components. Chloride Exide, the largest commercial battery manufacturer in Kenya and a primary player in the PV market, has been a target of such brand piracy. Their red, white and blue logo, famous in Kenya, has been found on PV components alongside the names “Chloride” or “Exide.” Counterfeit components also tend to advertise inflated capacities to increase the price of the unit.93 However, for the SHS consumer, the fact that quality is not highly valued may serve to increase the demand for counterfeit products. Due to seasonal fluctuations in income in rural areas, SHS consumers tend to have little investment capital available. Consequently, when purchasing solar components, priority is given to units of the lowest cost, regardless of quality. The price of higher quality components complicates the cash purchase. Thus, if a consumer has 3000 Ksh (approximately US$ 40) in his/her pocket at the time of purchase, he/she is more likely to purchase the lower quality, and possibly counterfeit, 2700 Ksh solar panel rather than a 4200

92 93

Ibid Interview with Festus Irimu 3-18-04

49 Ksh name-brand panel. Quality is not a priority because purchasing higher-end products is costprohibitive and the panel is needed at the current time.94 The issue of counterfeit goods was identified by all of the PV component suppliers interviewed as a significant threat to the vitality of the Kenyan PV market and many claimed to have undertaken more aggressive consumer education programs. A notable strategy that emerged during the research process involved providing return customers with a discount on their second purchase of a component. In theory, if the consumer purchases a low-quality product and it fails to meet valid expectations, he/she can return to the dealer and upgrade with little additional cost. According to the manager, programs such as this have proven successful in securing customer dedication to SHS following a poor experience. In reference to the tendency of consumers to favor lower-quality components, he optimistically remarked, “We have been able to turn the tide.”95 Turning the tide will also require extensive government support. Although the Kenyan Bureau of Standards has adopted a series of standardization requirements for imported PV components, such actions will remain merely paper measures until the government develops greater enforcement capacity. We recommend component standardization as a means of simplifying and regulating the stream of PV related inputs. Finding 2: Warranty programs are common, but ineffectual As discussed above, a product warranty is a primary component of quality assurance. Although warranty programs exist for the vast majority of components in the Kenyan PV market, a number of constraints serve to undermine the credibility of the warranty process. The most prominent of these constraints are geography and conditional arrangements. Geographical constraints in Kenya mirror those discussed above. In particular, the act of transporting a defective component to a dealer entails transportation costs and requires investing in alternatives to PV during the system failure. Our research indicated that kerosene lamps and fuel were sufficiently inexpensive and customers were unlikely to travel with their unwieldy components when an inexpensive alternative existed. Issues of communication were also identified as complicating warranty services. Although cellular phones are a pervasive aspect of Kenyan culture, even in rural areas, the high airtime costs prohibit regular use. If the consumer cannot contact the dealer, he/she must then devote considerable time to traveling.96 Conditional installation arrangements provide a further constraint to warranty effectiveness. All the vendors interviewed offered a one-year warranty on installation if their technicians performed the service. Similarly, several only offered a warranty on one component if it was used exclusively with components of a similar brand. Rural reliance on informal technicians and the high cost associated with formal installation services clearly provide an incentive for rural

Recognizing this condition, at least one PV component supplier has chosen to sell both high and low quality panels and has been identified by other market players as actively dealing in counterfeit and low-quality goods. 95 Interview with Jamlek K. Gicharu. 3/19/03 96 For the purposes of this section, the term “vendor” will be used in reference to a PV industry that markets component packages of its own design and is associated with a brand name. “Dealer” will be used in reference to an independent retailer that carries the products of several vendors.

50 SHS consumers to forego the one-year warranty on installation. Likewise, because many rural SHS consumers purchase components one-at-a-time and from across brands, uniform systems are uncommon. These issues may be resolved in areas of high SHS dissemination through the formulation of local PV/renewable energy experts who are paid through a community organization and are responsible for providing maintenance and installation services in addition to responding to warranty concerns. Vendors have an incentive to train such personnel if an exclusive partnership agreement is reached. The technician is employed by and resides in the community, the community pays a smaller fee for trained expertise and warranty support and the vendor benefits from having a dealership of sorts within the community. Finding 3: Formal technician training varies widely. The line between formal and informal technicians is difficult to define in the Kenyan PV market. Vendors typically rely upon two types of technicians: traditional certified and uniformed technicians and contracted rural technicians. The former is easily distinguished whereas the latter is often a Jua Kali technician who has undergone a type of PV mentoring program. Consequently, within the PV market, “formal” technician training takes several forms. Interviewed vendors claimed that their company maintained a certification program for uniformed technicians and a less formal program for contracted technicians. Initial formal training programs last on average from 8-12 weeks and applicants are expected to have a BS in electrical engineering or commensurate work experience. In addition to structured classroom lessons, most trainees are required to complete a mentorship program of approximately three months in order to qualify for certification. Each respondent claimed that continuing education was a training priority although none of the programs surveyed required technicians to re-certify after a given period of time. Reported training among contracted technicians varied considerably. In particular, a trend emerged suggesting that when a respondent had little managerial responsibilities, he/she tended to report that training initiatives were, in practice, less structured than had been reported by their superiors. According to one vendor, all contracted technicians must complete a mentorship program prior to receiving contracts. When we interviewed a contracted technician working for this vendor, he/she was uncertain as to the existence of this program as was a regional manager.97 This demonstrates the low priority many vendors place upon regimented training initiatives. Consequently, quality of maintenance and installation vary widely among formal technicians, and this incongruity serves as an additional incentive for consumers to rely upon less-expensive, informal expertise. Employee training efforts are further frustrated by the prevalence of independent retailers who carry PV components. Although vendors may prefer that the employees of independent retailers receive PV specific training, one vendor noted that many dealerships have little incentive to meet conditional arrangements associated with carrying a brand. In response to the broad variance among vendor employees and dealer employees, one vendor has instituted a “Solar Franchise” program. Under this initiative, a dealership agrees to maintain personnel certification

Interview 3-19-04

51 requirements in return for subsidized advertising and lower component costs. Such dealerships account for roughly 80% of the vendor’s sales.

52 CHAPTER 5 CONCLUSION AND RECOMMENDATIONS 5:1 Conclusion Our research practical experience has demonstrated that the consistent barriers of rural PV dissemination are associated with financing mechanisms, capacity buildings and quality assurance and are exacerbated by geographical, informational and financial market failures. Efforts to address these barriers may be identified but are site-specific and require intimate knowledge of the program context. In the case of Kenya, functioning financing arrangements, technical expertise and consumer awareness differ from the preferred models identified within the development community. The barriers we have identified have been validated through research and practical experience. Our research shows that the barriers to financing in PV dissemination are high initial cost of the system and access to credit options. The field research further demonstrates that the consumer financing preference may vary from suggested paradigms. Capacity building in the PV dissemination context faces barriers in information transfer, organizational cooperation and technical training. The field research supported these findings and suggests that consumer misinformation and geographic constraints represent additional barriers. Quality assurance in PV projects is a barrier due to most developing countries’ lack of institutional capacity to adopt and enforce standardization and accreditation procedures for hardware, training and practitioners. Thus, the consumer must receive accurate and timely information on appropriate quality provisions. 5:2 Invalidated Assumptions Assumption 1: We assumed that perspective that we would research the project design process. However, we found that a discussion of project design was highly dependent upon further discussion of project implementation. We have chosen instead to focus our consideration on overlapping barriers that consider both the design and implementation contexts yet remain barriers nonetheless. Going into the research process we found that the areas we identified as most challenging were overlapping between project design and implementation and we could not adequately explore them by strictly confining them to the design process. This led us to consider these areas (financing, capacity building, quality assurance) within the overlapping context. Assumption 2: We believed that rural PV dissemination efforts would prioritize the poor. In particular, we assumed that access to PV would be facilitated through credit-related financing schemes. We also assumed that the poor would acquire PV systems for the purposes of income generation and improved lighting (for nighttime studying, etc) as these relate to povertyreduction. Although these assumptions may have held in other areas, our site visit to Kenya demonstrated otherwise. Indeed, PV turned out not to be a high priority for poor communities in rural Kenya. They were more concerned with issues such as shelter, food, employment, etc. In addition, micro

53 finance in PV market is rare, and cash sales constitute the most widespread financing model. Contrary to our belief that rural consumers of PV electrification would use it to acquire revenue, and may be improve their education and health status, end-users’ principal use of PV was for entertainment, i.e. television. Assumption 3: PV projects would be large scale and in part directed to poverty reduction. We expected to see PV projects to target educational and health institutions (i.e. powering boarding schools, vaccine refrigeration, powering rural health centers, etc). We also expected to see multiple-households sharing a large capacity panel. In the case of Kenya, individual SHS rather than large scale PV constitutes the largest part of the market. Very few PV powered health and education facilities were identified, this may be due to the fact that PV marketing in Kenya is unsubsidized. In the site we visited as a team, we did not find any multiple-households PV sharing systems. Assumption 4: Given the fact that Kenya has the largest unsubsidized PV market, we assumed that there would be considerable NGO involvement with PV dissemination and development related projects. We found only one NGO in Kenya that specialized in PV dissemination activities. While PV market is functioning without NGO support in Kenya as of today, there are number of areas NGO can help improve the situation include coordination among different actors in the market and information dissemination to end-users. 5:3 Recommendations PV in the Context of Development and Poverty Reduction. Going into this project, we had an additional interest, which was to observe the interaction between PV dissemination and poverty reduction in the field, as we do believe that solar energy could have a significant role in rural development. Although our assumptions regarding PV and development (notably the size, targeting and use of these types of projects) were mostly invalidated by the Kenyan example, we still think that the relationship is defensible. Disseminating PV technology and promoting its use in the global marketplace are a primary goal of many projects designed by solar energy organizations. As a result, it seems that the paradigm is to try and fit poverty reduction into an already set goal to diffuse the use of PV. This may be where the problem lays. In order to be relevant to poverty reduction, PV should be fitted to development, not the other way around. Focusing on PV technology dissemination would mean that most of the efforts will go to expanding the technology throughout an area first, without worrying about who gets it or how, as long as they get it. Consequently, not much effort to target the poor and educate them on how they may use the technology to better their lives can be expected. Moreover, unless it is a large-scale project, which would benefit more people simultaneously, the poor will tend to ignore PV technology and be oblivious to its potential. By shifting this paradigm, the role of PV technology in poverty reduction could be rather substantial. Prior to designing a PV project for a particular developing country, planners and donors could

54 assess the specific poverty issues the country is facing and decide on how PV could contribute to resolving some of those issues. PV could be used in areas such as increasing access to health care for rural populations through powering off-grid health center with solar energy. Schools, vaccine refrigeration, small businesses are only a few other possible areas PV could be used. A good place to start for fitting PV projects to rural development and poverty reduction is the target country’s Poverty Reduction Strategy Paper. PRSPs are supposed to be a participatory approach, where governments, NGOs, civil society and other relevant actors come together to identify the country’s poverty issues and come up with a viable strategy to address them. Therefore, it is purportedly a country-owned document that highlights issues that are important to its constituents, and having this document in hand when looking to design a PV project for the target country would provide an insight into those areas where such project could be most relevant. Furthermore, basing the objectives of a PV dissemination program on a country’s PRSP could help in rallying support around the project, both from governments and civil society and other NGOs, and open more possibilities for larger-scale, more effective projects. In brief, in order to be relevant to poverty reduction, PV dissemination project need to become a tool rather than a goal for development projects, as we have uncovered during our site visit that PV dissemination does not necessarily mean poverty reduction or target the poor at all. Demand-side Capacity Strengthening As mentioned throughout the report, training programs that have targeted informal technicians in Kenyan PV market have by and large failed, omitting to address the utter lack of incentives for these technicians to improve skills for which they still have substantial demand. Therefore, we recommend that the target of these training programs be shifted away from the supply side toward the demand side of the technician market. Indeed, since these technicians still face a large demand for their limited skills, they would only consider improving them if they were faced with changes in demand that would make it a necessity for them. We contend that training programs aimed at consumers will raise consumer expectations. Conducting consumer awareness campaigns and social marketing programs, during which consumers would be adequately educated about PV technology, products and quality issues, would give them more power to demand higher quality services from the informal technicians. In response to the new demand for increased quality, informal technicians will have an incentive to seek additional training. But as long as consumers are not adequately knowledgeable about key quality issues, some of them as basic as which wires are the suitable match for what system, informal technicians will have no incentive to enroll in or seek training to improve their skills.

55 Appendix A Site Visit Research Questions

Objective of the Kenya Site Assessment: There are several reports and manuals related to Solar Home System project design in rural areas of developing countries. A common thread among these manuals is the necessity of incorporating flexibility into any project design. Their arguments are well prepared and their reports are based on extensive experience. However, similar to other “Blue Print” types of project design manuals and reports, such ideals exist only in theory. The researcher is left with the task of determining how the blueprint translated to the reality of the project site. This question will frame much of our research. The field site-assessment to Kenya intends to uncover the reality of SHS dissemination projects on the ground. There are key questions to ask to not only end-users of SHS, but also to nonusers, project implementation agencies, NGOs, venders, university professionals, and government officials. These site assessments will provide the real-world context by which our research may be validated or dismissed.
Box 4-1. The Kenya Situation In Kenya less than 0.5% of rural households have access to grid electricity while population growth exceeds the rate of rural connections despite major investments in the rural electrification program. Kenya has been an active market for solar home systems for almost a decade, during which more than 150,000 units have been installed on a commercial basis and an estimated 15 -25,000 new systems being installed yearly. Whereas the figures are based on module sales, it is important to note that not every system is ‘complete’ (i.e., uses a charge controller). The private sector has responded quickly to this growing consumer need, offering an increased and diversified supply of solar energy components and technologies on the market. Currently there are more than ten solar module distributors and at least five companies manufacturing balance of system components (BOS). Scores of agents market install and service solar electric equipment in rural areas. Presently the estimated size of the Kenya solar PV market is valued at about US$5 million and has been growing at more than 25 % per year for the last 7 years. Today, there are more rural Kenyan households with solar electric systems than the government’s Rural Electrification Programme customers, despite the fact that PV equipment is taxed and REP equipment is subsidized. This demand has created and sustained a growing market for solar products and services, mainly from households with ability to meet initial costs of purchase and installation. Most of these relatively well to do households have acquired systems on a cash basis, indicating that affordability, among other considerations, is an important factor in the solar energy technologies uptake among rural households.
(Source: ESMAP, Kenya Implementation Manual: Financing Mechanisms for Solar Electric Equipment, July 2000)

Research Survey Groups and Questions: End Users: Why end-users? As the recipients and users of rural solar electrification, end-users are arguably the most aware of the pros and cons of a given PV program. Furthermore, as the purchasers of

56 solar systems, they are keenly aware of the viability of various financing schemes and the utility they are willing to assign to these technologies. End-users have little say in the quality of products yet suffer the most from poor quality technologies. From a capacity building perspective, end-users require education on technologies and maintenance. The site visit surveys are further intended to inform our research regarding the capacity of rural solar electrification to reduce poverty, to target the poorest if indeed it is intended to, and to increase the social capital of rural communities. Process: The interview may be conducted as a focus group or individuals depending upon availability. We will ask our contact at the site to aid us in forming a group for our visit and will speak to as many individuals as possible if this is not possible. The later is the most likely scenario. Kenyans traditionally enjoy an afternoon meal together and this may prove a beneficial time for such surveys (assuming we are welcome during this time). We intend to present ourselves as consultants for SEIA, including who SEIA is and its desire to extend overseas, and would appreciate feedback form their experience with PV projects. We do not wish to appear as though we are judging or measuring them and would welcome input regarding how we may further accomplish this. Questions: What do you do for a living? This question will be asked to determine the type of employment including security of income (seasonal, cyclical, full-time). We are interested in understanding issues of income as it relates to financing programs and expendable capital. This question is intended as an icebreaker but also provides us with necessary data regarding literacy, income, and the local labor market. How did you buy your system? This question relates to our financing research and provides an insight into local preferences, satisfaction with existing financing schemes, information on financing options, and availability of financing. In short, how much information do they have regarding options? Also, this provides insight as to the correlation between financing options and post-sale quality maintenance (research suggests cash sales promote lower quality in maintenance services). How did you learn about this PV project and how to operate your system? This question is directly related to capacity building and social mobilization. It also indicates the stakeholders involved in the process (were education efforts sponsored by commercial, public or NGO related bodies, as this relates to the financing and capacity building processes). What do you see as the benefits/limitations of your PV system? This question will assess customer satisfaction, and help us find out how the end-user’s life has improved (new income generating activities, higher productivity, improved health, improved standard of living, etc.). It is intended to uncover system quality concerns and affordability and financing issues.




Procedure: Jungo will work with the interpreter while Josh records responses on paper. This information will be analyzed in a cross-comparison qualitative case-by-case process. Due to data limitations, we will not attempt a quantitative analysis of our findings.

57 Non-Users Why non-users? Non-users will be surveyed to determine what prevents them form participating in PV programs. For those who lack the resources to participate, we may determine if market failures have caused their exclusion. For those who choose not to participate, we seek to determine their reasoning and their perceptions of their existing energy sources (kerosene, batteries, biomass, etc.). We are also interested in determining what social and cultural norms (power relations, class or caste systems, opportunity discrimination) exist at the site and we believe non-users may be more willing to discuss these conditions. We recognize that actors may be unwilling to divulge such sensitive information and will seek for such information in the survey responses. Process: For determining non-users, we will network through users to identify potential survey participants. We will inquire users if neighbors or friends are participants and will request politely to be introduced if the individual in question is a non-user. We intend to present ourselves as consultants for SEIA, including who SEIA is and its desire to extend overseas, and would appreciate feedback form their experience with PV projects. Questions What is your perception of [the solar technology being used]? This question is intended to determine if the-nonuser has chosen to not participate, lacks the resources to participate, is socially excluded, or is unaware of the program. These answers may point to issues of financing, capacity building, quality assurance, and trust. We have chosen not to ask, “Why don’t you participate” as we do not wish to deny the non-user perspective. How much do you spend in time and money on energy sources? This question intends to avoid requesting non-users how much they may be willing to pay due to our assumption that responses may reflect deflated figures. Instead we seek to determine if their current energy expenditures are on par with solar-related costs. We also seek to determine preferred non-solar energy sources. If they have been denied credit due to lack of collateral, this may identify them as the poorest members of the community. This may better inform our questions regarding the suitability of rural solar electrification as a poverty reduction strategy.


Procedure: Jungo will work with the interpreter while Josh records responses on paper. Commercial Vendors Energy Alternatives Africa Ltd. (EAAL) Austral Ken Ltd. (AKL) Why vendors? Accessible vendors providing sales and maintenance services have been recognized by the IEA as necessary components of solar project sustainability. The vast majority of rural solar initiatives rely upon vendors for technical supply and maintenance. In addition, supply side market failures (limited market information, failure to remain technologically abreast, geographical limitations) will have the greatest implications for vendors. Vendors support capacity building through implementing accreditation and standardization

58 programs. In terms of financing, vendors may provide credit options for customers while operating as consumers within financial markets. Process: We will present ourselves to these vendors as consultants working for SEIA and will explain SEIA’s intention to support developing country solar initiatives. Questions Do you offer a warranty? According to the IEA, warranty programs for components and installation are fundamental aspects of a high quality service system. With this question, we seek to determine service consistency as a measure of quality assurance. In addition, we will inquire into the process of recording warranties and responding to claims, as these are additional quality assurance concerns. Under what certification programs do your hardware and technicians operate? Certification programs are a necessary aspect of both capacity building and quality assurance. In particular, we will search for standardization within the industry. We are also interested in how employee certification correlates to increased wages and/or status and responsibility as recognition of achievement has been demonstrated to inhibit ”brain drain.” How do you maintain accessibility for responding to user needs? We seek to determine how accessible vendors are to customers. In particular, what forms of communication are made available to customers and are such forms only available to wealthier parties. This question related to both issues of quality assurance and poverty reduction



Procedure: We have contacted the above-mentioned organizations and have appointments with them during our field trip. We anticipate not having language barriers with vendor representatives and intend to have in-office meeting with both parties. We have chosen to speak with a regional vendor and a small-scale vendor, as we believe the differences in perspective may yield insight into the business environment. We will create written records of these conversations and we will request if we may tape record the conversation. University Faculty Nairobi University Kenyatta University Why university faculty? We wish to speak to university faculty working in Photovoltaic for their perspective on rural solar electrification efforts in Kenya. We believe that their broad theoretical perspective combined with the reality of their developing country experience will provide valid contextual information. Process: We will present ourselves as MA IDS students and explain the purposes of our trip. Jungo has studied under our survey subjects. Questions: We will be speaking with university faculty during the later section of our field trip and intend to discuss our in-country experiences and broader research questions. We also intend to inquire as to any contacts they may have within those government agencies involved with

59 solar power. In addition we will inquire as to any government documents to which they may be able to direct us. Procedure: We will create written records of these conversations and we will request if we may tape record the conversation. Government Officials Why government officials? Government agencies are responsible for coordinating policy responses and regulation within all sectors relating to solar energy production. We intend to inquire into these policy and institutional frameworks. In particular, we are aware of how these frameworks operate on paper and seek to determine how implementation differs from such blueprints. Speaking with government officials may also help in determining levels of political will as our research has demonstrated the necessity of such commitment? Solar power policies are the responsibility of several agencies and we do not expect our network of contacts to guarantee the availability of a government official. We have been attempting to organize such meetings since early February. Process: We will present ourselves as MA IDS students working with SEIA. In reality, we do not expect our intended survey participants Questions: What is the role of the Kenyan government and your agency in the solar energy sector? This question seeks to determine the levels of interest, commitment and understanding on the part of government agencies. These are necessary aspects of capacity building at the public level. In addition we seek to determine what perceptions of future solar power extension exist within the agency. Do you work with other agencies? This question seeks to determine the levels of interagency co-operation as it relates to solar power policies.


Procedure: We will create written records of these conversations and we will request if we may tape record the conversation. Implementation Agencies Muranati Tea Growers SACCO Corporate Bank of Kenya Ltd. Kenya Rural Enterprise Program Why implementation agencies? In Kenya, NGOs, CBOs, and financial institutions primarily undertake implementation. NGOs and CBOs offer micro-credit options while financial institutions provide more robust funding options. In assessing a project, the perspective of the implementing managers may provide information regarding the challenges of applying blueprint project deigns to the real world. In addition their perceptions of the project may be contrasted to that of end-users (as both groups may regard solar energy differently). From a financing

60 perspective, project managers are responsible for choosing project specific financing mechanisms and may hold valid insights as to the viability of various options. From a quality assurance and capacity building perspective, project managers must organize suppliers and depend upon the quality and performance of technical and human inputs. These are the institutions involved in the projects we intend to visit. Process: We have organized interviews with implementing organizations and have requested to speak with project managers if available. We will present ourselves as independent consultants for SEIA (we feel the term “consultant” suggests greater responsibility then “student”). Questions These questions will be directed towards project managers What where your greatest implementation challenges? We recognize that challenges will prove site specific; however, we intend to determine how project managers have overcome challenges in applying blueprint project designs in the real world. This will provide information as to the general design phase, may point out existing market failures, and may confirm that financing, capacity building, and quality assurance continue to offer unique development challenges. We will also inquire as to the project design features that may overcome such the challenges they have experienced. What are the characteristics of your targeted population? This question is intended to determine income levels, accessibility, participants level of solar energy familiarity (and if the organization provide training), and will aid our research regarding solar energy and poverty reduction. What are your short term and long-term goals for the project? In addition we will inquire as their definitions of short and long term. This question is intended to determine the sustainability of the project. Are they supporting solar sector growth and are local communities becoming more self-sufficient? When does the organization consider a project complete and how is success measured? These questions relate to our financing research. What was your design process? To be compared with our project design research. In addition we will consider how our tentative lists of best practices and lessons learned may fit into this project design. What are the implications of the project design on our research? What was the role of micro-credit in this project? This question seeks to determine how the availability of micro-credit expanded the target population. In addition, how prevalent were loans in the financing of the project? Does the project manager view micro-credit as a viable financing option? We will also inquire as to the nature of the micro-credit program (lending and payment requirements, default rates, eligibility, etc.).





These questions will be directed to financial institution representatives: What is the nature of your loan program? This question will be asked to both microcredit and conventional financial organizations to determine their respective loan and payment requirements, consumer eligibility, default rates, etc.

61 Is your organization engaged in any non-financial activities related to project management/implementation? This question intends to confirm our research findings that financial organizations in Kenya are involved in various non-financial support activities related to project implementation. We seek to determine what those activities tend to be, and how they are conducted. We also want to find out what difficulties these organizations face in supporting the dissemination of PV rural electrification, especially for the poor.

Procedure: We will create written records of these conversations and we will request if we may tape record the conversation. Methodology The purpose of site assessments is to see the actual sites and meet with people in face to learn on-the-ground information that cannot be extracted from literature. During the stay in Kenya, the team will visit various SHS stakeholders and (1) collect written manual, reports, and evaluation paper of SHS that cannot be collected in the U.S., and (2) interview the stakeholder to learn about the on-the-ground truth of SHS dissemination in rural community in Kenya. The key questions will be asked in the form of semi-formal, or informal questionnaires and interviews. Intended numbers of interviewees are rather small in order to extract deeper information from few people. The interview will not use paper questionnaires but small memo for interviewees. The target groups of interviews are SHS end-users, non-users, local CBO, NGO, implementation agencies, venders, financial institutions, university professors, international aid-agency workers, and Kenya government officials. For the interview, the team has employed personal translator of English/Swahili speaker. If the interviewees do not speak or feel comfortable with English interview, the team will use the translator to communicate each other. However, there are possibility that interviewee do not speak neither English nor Swahili. In this case, the team will try to find the person who can translate at the site. If the team cannot find a translator, team will give up on interview. 4.5 Interview Schedule
Day 15 Activity Arrive in Kenya Meeting with Hosting company PO Box 4900, 00200 City Square, Nairobi, Kenya Confirmation of meeting appointment, a Cell: 254720844930 or 254-733driver/translator, payment for their services 707150 E-mail:[email protected] Moses Mwaya Agumba PO BOX 76406, Nairobi 254-2-714529 Dissemination of PV and other solar related product and services to rural areas. Contact Address Description

Onsare James


Interview Association

Solar Energy Network (SOLARNET) Solar Electric Kenya

Interview NGO

Mark Hankins, EAA, PO Box Dissemination of PV and other solar related product 76406, Nairobi Tel: and services to rural areas. +254-2-565616

K-Rep is a leading micro finance organization. KRep has been working with the World Bank and other international organization for dissemination of SHS in Kenya.


Interview MicroFinance

K-Rep Development Agency


Interview Vender

Energy Alternatives Africa Ltd.

Products include PV, Wind energy systems, and other renewable energy products and services. PO Box 76406, Nairobi, Kenya Provide consulting, design, installation, engineering, FAX: 254-2-720909 project development services, education and training services research, site survey and assessment services


Part of IFC-GEF The Photovoltaic Market Muramati Tea Growers General Transformation Initiative Muramati Tea Manager Target group: 15,000 firmers Site Assessment 1 Growers SACCO PO Box 1145, Murang’a Project: Solar Home System (PV) Tell: 254 156 22991 US$0.6Mil Part of IFC-GEF The Photovoltaic Market Address: The mall of Westlands, Transformation Initiative Site Assessment 2 Solagen (SACCO) Nairobi Kenya Target group: 15,000 firmers, US$2Mil Tel: 441160/1 Solar Home System (PV), Solar garden lights, solar water heating systems, etc Mt.Kenya Site Assessment 3 Community with K-Rep Site Assessment 4 Interview Nairobi University Interview Kenyatta University Extra day Organization day Leaving Kenya Dr.Jama Dr.Buelongo Nairobi Nairobi Ask insights from academic professional in SHS Ask insights from academic professional in SHS K-Rep has evaluated SHS project in Mt.Kenya Community in 2000 for the World Bank site assessment.



21 22 23 24 25 26

Sponsor Documents

Or use your account on


Forgot your password?

Or register your new account on


Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in