ISSUE BRIEF

STIMULATING PAY-AS-YOU-GO ENERGY ACCESS IN KENYA AND TANZANIA: THE ROLE OF DEVELOPMENT FINANCE

SANJOY SANYAL, JEFFREY PRINS, FELI VISCO, AND ARIEL PINCHOT

EXECUTIVE SUMMARY Nearly 620 million people in sub-Saharan Africa lack electricity access. Improving access to affordable and reliable energy is critical to reducing poverty and improving quality of life (IEA 2011). To improve energy access, it is important to develop financing and payment schemes that fit consumer energy budgets. “Pay-as-you-go” (PAYG) business models harness technology to provide a “one-stop-shop” solution for consumer finance and energy products. The PAYG model originated in Kenya, and addresses the key challenges of extending end-user finance and collecting payments from remote customers who often have erratic and limited cash flow. PAYG companies, at this point, typically provide basic lighting and mobile phone charging services. The technology can play an important role in expanding access to electricity services to remote and low-income populations.

WRI.ORG This issue brief draws on findings from to large companies—and five debt desk research, workshops, and inter- deals struck in 2015, the largest of CONTENTS views with PAYG companies, donors, which was a US$45 million raise by 1 Executive Summary and development finance institutions one company. Market leaders such (DFIs) active in energy access in East as M-KOPA, Mobisol, and Off-Grid 4 Introduction Africa to assess how PAYG companies Electric have begun expanding into have stepped up to serve the approxi- regional markets. 7 Section I: Exploring the mately 35 million people in Kenya and Pay-as-You-Go Model 36 million people in Tanzania who lack While encouraging progress has been 16 Section II: Stimulating access to electricity, as well as addi- made, the addressable markets in Local Debt tional millions who are underserved. Kenya and Tanzania are much larger Our paper also draws on interviews than those reached by existing compa- 27 Conclusions with stakeholders involved in Ban- nies so far, and the products they offer gladesh’s IDCOL program to provide need to be larger in capacity if they 29 Recommendations insight into how DFIs and donors are to provide more than basic light- 31 Annex I: Methodology supported the Bangladesh program, in ing and mobile charging. PAYG com- order to elicit lessons relevant to the panies will require about one billion 35 Annex II: Multi-Tier Kenyan and Tanzanian contexts. We dollars across these two countries to Framework for Energy chose Bangladesh’s IDCOL program scale for broader impact. Therefore, this Access as a reference point for two reasons: issue brief focuses on how this broader 37 Annex III: Private Sector the energy enterprises in Bangladesh impact can be created. We look at how Investor Profiles perform the same one-stop-shop role successful PAYG businesses operat- as the PAYG companies, and IDCOL ing in Kenya and Tanzania have raised 41 Annex IV: Exchange provides an example of where DFIs finance and the constraints faced by Rate Information have played a significant role in chan- the industry, and we propose recom- neling finance (US$750 million) to mendations for how donors and DFIs 42 Annex V: Details of the achieve substantial energy access goals can continue to support the develop- IDCOL Model (three million solar home systems). ment of these markets. 44 Annex VI: Solar Home System Prices Given the nascent stage of most Currently, the various types of capital energy access markets, much of the (debt, impact equity capital, grant) 45 Annex VII: Lines of Credit existing PAYG literature focuses on that PAYG companies need are avail- analyzing the innovative variations able almost exclusively from interna- 48 References of business models as well as fac- tional investors. Local financial insti- tors that could improve the enabling tutions in Kenya and Tanzania have environment. However, market been hesitant to provide financing players in both Kenya and Tanzania to PAYG customers: they perceive have evolved beyond an early-stage PAYG companies as early-stage, pilot phase. These pioneering risky businesses and are unfamiliar companies have successfully raised with the technology as well as the grant, equity, and—more recently— creditworthiness of rural consumers. debt finance to pilot, develop, and The absence of local capital sources scale their businesses. According to some extent explains the fact that to our estimates, they have reached almost all the successful PAYG com- more than half a million households panies are foreign owned and foreign through rapid sales growth. The managed. Local companies often market overall is also evolving, as lack the initial resources, as well as suggested by the participation of 52 the networks and skills, to raise both international private sector inves- early-stage capital and develop com- tors—ranging from foundations plex financial structures to raise debt

2 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

capital from international markets. financing involving private donors Local commercial banks can Local companies are also hesitant to and civil society organizations could ▪▪ begin to explore the PAYG sec- take on foreign currency risk. help private sector operators build tor, and understand company up their marketing and distribution cash flow patterns, through the Technological barriers to the PAYG infrastructure under a limited period provision of short-term trade business are falling, and the sector of donor assistance. Monitoring and finance. They can also explore is likely to see the entry of a larger verification functions could be housed mechanisms such as a debt ser- number of companies. This is not yet within public organizations such as vice coverage account to partially happening, because access to finance rural electrification agencies, in part- cover for default risks. remains a key entry barrier, particu- nerships with the respective bureau of larly for locally owned and managed standards, as per World Bank Group ▪▪ National governments can pro- companies. Finance is most critically (WBG) Lighting Africa standards. vide support through a suite of needed to build out marketing, sales, policy and regulatory measures and service infrastructure and to Drawing on the success of the IDCOL to unlock domestic commer- provide customers with financing. program and the unique needs of cial financing for distributed The relative lack of access to finance PAYG companies, we offer recom- renewable energy including, for results in fewer companies and less mendations targeted primarily to example, the development of competition in the PAYG sector. DFIs and donors regarding how they mechanisms to coordinate roles can support local financial institu- of institutions in this space and Local currency debt could eliminate tions in their efforts to expand energy encourage private sector activity currency volatility costs, reduce access in Kenya and Tanzania. by setting clear national priori- the transaction complexity associ- ties and releasing grid extension ated with raising international debt ▪▪ International DFIs and donors plans to the public. and reduce entry barriers for local can leverage their long relation- Private sector investors can help companies. In turn, new entrants ships with local financial institu- ▪▪ companies to access different and increased competition could tions in Kenya and Tanzania to types of capital and partnerships lower prices and allow offerings stimulate local finance for the in response to evolving business of products and services that offer PAYG sector. DFIs and donors needs. This may include sup- higher levels of electricity access at can provide guarantee schemes port for raising capital from local affordable prices. and lines of credit to local banks. This support would help banks commercial banks. Foundations and family offices can provide DFIs and donors have a role to develop a deeper understanding loss guarantees to local banks. play in supporting local financial and familiarity with PAYG busi- ness models, and make finance institutions to extend local currency Private sector PAYG businesses more accessible to local com- debt. In Bangladesh, international ▪▪ can adopt standardized account- panies. International DFIs and DFIs and donors channeled funds ing standards to assist in trans- donors can “crowd in” private for energy access through IDCOL, actions with local banks. a government-owned financial sector investment in PAYG by channeling their investments intermediary. IDCOL also played a The scope of this issue brief is through fund of funds run by strong role in market development. confined to analysis of financing in professional impact investors The market support roles played by support of PAYG solar home system and incentivize PAYG companies IDCOL can be adapted to the Kenyan companies. While we recognize that to invest in targeted marketing and Tanzanian contexts. The debt- PAYG products providing lower-level and distribution infrastructure financing role in Kenya and Tanzania energy services are not comprehen- through results-based financ- can be played by commercial banks sive solutions to the energy access ing. DFIs and donors can also from the very beginning. Involving challenge, we believe that our recom- provide technical assistance to commercial banks would have the mendations will also support the public organizations to support advantage of ensuring that funds broader energy access sector, includ- capacity building in monitoring are available to the sector even after ing mini- and micro-grids. donors withdraw. Results-based and verification.

ISSUE BRIEF | June 2016 | 3 INTRODUCTION The framework measures electric- wal, and Wood 2016). Other research ity access across five tiers; each indicates benefits such as preven- The Imperatives of the tier reflects a specific level of per- tion of GHG emissions (both carbon Electricity Access Challenge formance of an electricity supply dioxide and black soot) (Kaufman Nearly 1.3 billion people, or 18 per- system defined by the attributes. et al. 2000; Wang et al. 2011), cent of the world’s population, still Tier 1 and Tier 2 are the low-power increased household disposable lack access to grid electricity (IEA capacity levels (minimum 3W and income because of reduced spending 2014a). An additional one billion are 50W, respectively). At Tier 1 level, on kerosene and candles (Mills 2005; “under electrified,” a status charac- electricity access is defined as provid- Tracy and Jacobson 2012), health terized by unstable grid connection ing lighting and mobile charging for benefits such as reduced accidents with regular power outages (A.T. a minimum of four hours per day. and indoor pollution (Mills 2014; Kearney and GOGLA 2014; IEA At Tier 2 level, access additionally Samad et al. 2013) and social ben- 2013). Sub-Saharan Africa bears a includes the ability to power a fan efits such as increased evening study disproportionate share of this bur- and/or television for four hours (see hours for children (A.T. Kearney and den. Over 620 million people, nearly Annex II). GOGLA 2014; Khan and Azad 2014; two-thirds of the region’s population, Samad et al. 2013). are without electricity access (IEA The PAYG businesses that we study 2014b). Increasing access to afford- in this issue brief provide electricity able and reliable energy services is access mainly at the Tier 1 and Tier 2 fundamental to reducing poverty and levels through standalone solar home Measuring energy improving other human develop- systems (SHSs). The standalone ment indicators (IEA 2011). solar system comes with a battery, a access across charge controller, a solar panel and Electricity access has long been mea- LED (light emitting diode) bulbs, tiers allows sured by the physical connection of and a mobile charger. Larger systems a household to grid electricity or the (typically 50W and above) can poten- one to measure presence of a nearby electric pole. tially connect direct current (DC) This binary definition of electricity appliances such as a television. Even the quality and access has increasingly come into at lower tiers of electricity access, question in recent years, because it there are numerous household-level quantity of energy fails to capture the quality of electric- benefits. These benefits stem from ity services received by end users. In the fact that the SHSs replace alter- services that response, the World Bank’s Energy nate sources, which are often very Sector Management Assistance expensive. are available to Program (ESMAP) has developed a multi-tier framework for defin- Previous WRI research conducted in households. ing and measuring levels of energy collaboration with the International access. Under this approach, access Institute for Applied Systems Analy- to electricity refers to the ability to sis indicates that household kerosene obtain electricity that is character- use is significantly lower for house- ized by the following attributes: holds with SHSs, even when com- “adequate, available when needed, pared with grid customers. While 80 reliable, of good quality, affordable, percent of households with access legal, convenient, healthy and safe to grid electricity continue to use for all required applications across kerosene, only about 25 percent of households, productive enterprises homes with SHSs use kerosene. The and community institutions” (Ange- reliability of SHS electricity supply lou and Bhatia 2015). may explain this finding (Rao, Agar-

4 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

The Importance of “Pay-as- From the consumer’s point of view, communication technologies, also You-Go” (PAYG) the PAYG model offers a one-stop reduce the costs associated with Previous WRI research has under- shop, where the product and the collecting repayments (Rolffs, scored the importance of designing financing are available from the same Byrne, and Ockwell 2014). Finally, financing and payment schemes that source. The willingness of companies PAYG enables significant data fit consumer energy budgets. The to finance products gives customers’ collection. This gives enterprises the research notes that energy enter- confidence in the new technology. advantage of understanding product prises have to design innovative Indeed, energy companies have performance and consumer behavior financing and payment schemes to tried to partner with microfinance (Alstone et al. 2015). encourage consumers to purchase institutions (MFIs) but often with their products, because customers limited success. The energy service are accustomed to buying energy companies have typically been PAYG allows in small increments (Ballesteros et smaller than their counterpart MFIs, al. 2013). Consumer surveys have and partnerships have been hard to energy companies consistently indicated that the pro- manage given the differing expecta- pensity to purchase increases quite tions of the two parties. In Kenya, for to provide dramatically when consumer finance example, consumers could not access is available, both among rural house- technical maintenance services from consumers credit holds and rural small and medium the energy companies, which were enterprises (SMEs) (GreenMax limited in their geographic outreach. independently, Capital Advisors 2013; A.T. Kearney The poor after-sales service left and GOGLA 2014). The provision of many customers dissatisfied with without relying such finance has been a key chal- their products, which in turn led to a lenge for scaling up standalone solu- refusal to repay loans (Rolffs, Byrne, on partnerships tions (IRENA 2015). The same WRI and Ockwell 2014). research has indicated that there with MFIs or are several innovative approaches to The benefits of the PAYG model in providing end-user finance. Promi- providing a one-stop-shop solution other financial nent among these approaches is a to customers are several. As we have partnership between a company pro- already noted, the offer of finance by institutions— viding the renewable energy product the energy company instills trust in and a financial institution providing consumers regarding the quality of which are a loan to a consumer or a dealer. The the product. Operational efficiency PAYG model that has emerged in is improved because there is no need often difficult East Africa has several advantages for coordination between finance over this partnership approach. providers and technology providers. and ineffective With PAYG, the companies are able PAYG is a technology-driven method to provide longer-term loans than relationships. that allows consumers to pay the those usually offered by MFIs. PAYG lease amount for a given energy models also allow the provision system or pay a fee for the service of of relatively large credit amounts using the system. It uses information (to cover the cost of the renewable technology to enable remote activa- energy system) to consumers whose tion with payment receipt (Alstone credit worthiness may be unknown. et al. 2015). PAYG includes a range The credit risk is partially mitigated of business models, which differ as by the incentive system that links to how payments are accepted and to payments to service provision. whom the ownership of the system PAYG approaches, which use mobile ultimately devolves.

ISSUE BRIEF | June 2016 | 5 Objective and Scope The objective of this issue brief is to ABOUT THE SERIES assess how international develop- BOX 1 BOX ment finance institutions (DFIs) This is the third in a series of three briefs. These studies are based on interviews can support the PAYG energy access and desk research, as well as workshops held by the World Resources Institute sector in Kenya and Tanzania. Our (WRI) and the DOEN Foundation in 2012, 2013, and 2015. By leveraging the aim is to identify ways in which the experiences of socially oriented energy enterprises, civil society groups, and international development finance investors focused on energy access in developing countries, these publications community can help the PAYG use the collective knowledge of these stakeholders to help accelerate the scale- sector provide energy services with up of distributed renewable energy services in developing countries. The first the attributes of reliability, afford- brief—“Implementation Strategies for Renewable Energy Services in Low-Income, ability, and health and safety. These Rural Areas” (2013)—describes four common core business strategies employed service attributes are based on the by the enterprises and gives examples of how these strategies were implemented. multitier framework (MTF) to evalu- The second brief—“Clean Energy Access in Developing Countries: Perspectives on Policy and Regulation” (2015)—analyzes the challenges these organizations ate and monitor progress toward faced in delivering services, discusses how they have overcome these hurdles, and energy access goals developed by examines the enabling conditions that support their set-up, start-up, and scale-up. ESMAP and Sustainable Energy for This issue brief focuses on the financing strategies DFIs and donors can use to All (SE4ALL) initiative as part of support the growth of the pay-as-you-go solar home system market. the Global Tracking Framework (see Annex II for details on the multi-tier framework). The issue brief examines this challenge in the country contexts of Kenya and Tanzania. We chose Kenya and Tanzania because Kenya ommendations on which stakehold- PAYG area. The list of peer-reviewed is the country where the PAYG model ers can act to establish the enabling literature is provided in the Refer- originated and Tanzania represents conditions for appropriate support ences. To identify the main PAYG another large country where the from the international financing players, we conducted an extensive business model can have a significant community. online search of companies provid- impact. ing solar lighting and decentral- Methodology ized renewable energy solutions in The primary audience for the Kenya and Tanzania. We screened This issue brief is the product of paper is staff members within the and identified six PAYG solar home ongoing WRI initiatives in clean DFI and donor organizations who system companies operating in these energy policy and financing. It are involved in supporting energy two countries. The companies were builds on previous publications in programs in Kenya and Tanzania, or chosen based on the criterion that the series, which focused on imple- supporting the development of the they should have been able to raise mentation strategies for renewable private sector in these two countries. external equity capital (after initial energy services in low-income, rural We also aim to inform consultancies grant and angel investment) or that areas, and the policy and regulatory that provide services such as assess- they have an established PAYG offer- challenges facing the sector (see Box ments of project viability or program ing (beyond the pilot stage). This 1). This brief focuses on financing implementation support to DFI and allowed us to analyze the challenges strategies to scale PAYG models in donor organizations. in scaling up energy access using Kenya and Tanzania. We began with proven business models. Between extensive desk research focusing Following this introduction, Section April 2015 and February 2016, we mainly on end-user finance issues in I explores the PAYG model in Kenya interviewed the senior management energy access, Bangladesh’s Infra- and Tanzania. Section II examines of all six identified PAYG solar home structure Development Company how local debt could help scale up system companies by phone or in Limited (IDCOL) program, and the PAYG model to improve energy person at the company headquarters existing literature in the emerging access. We conclude with a set of rec- or the country offices in Kenya and

6 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

Tanzania, and we validated key data Hong Kong—based suppliers (one of ing needs of mini- and microgrids. In on funds raised and investors in the which was World Bank Group Light- the workshops, stakeholders noted ABOUT THE SERIES PAYG companies through online ing certified) to assess whether there that the recommendations have research. are any bottlenecks in supply and to general applicability for the mini- and get an overall assessment of quality microgrid sector. We recommend In the next phase, we identified the and price issues. additional research on the mini- and energy access projects and local microgrid sector to validate and financial institutional relationships Finally, we validated several of the tailor these recommendations to this of the major international DFIs and findings and generated additional important emerging sector. donors in Kenya and Tanzania. The insights from two workshops involv- brief draws on information gath- ing international DFIs, donors, local Details on interviewed personnel, ered from local bank, international financial institutions, and other workshop attendees, and interview DFI, and donor websites as well as stakeholders. The workshops were guide questions are provided in personal interviews with DFIs and convened by WRI and New Ventures. Annex I. donors in order to assess the rela- The workshops in Dar es Salaam in tionship between international DFIs April 2016 and Nairobi in July 2016 SECTION I: EXPLORING THE and donors and local banks in these focused on stimulating local capital two countries. in the PAYG energy access sector in PAY-AS-YOU-GO MODEL Tanzania and Kenya, respectively. Key Players in Kenya In order to draw lessons from the and Tanzania IDCOL model in Bangladesh, we We recognize that solar home system supplemented the literature review products that provide Tier 1 and Tier In this section, we describe the main with field interviews with the key 2 levels of access are not comprehen- PAYG players in each country, their actors involved: IDCOL management, sive solutions to the energy access indicative installed base (in terms the DFI supporters of IDCOL, and the challenge, and they should be consid- of number of estimated household IDCOL Partner Organizations. ered alongside other solutions. Mini- installations) and the main prod- and microgrids have an important uct offerings in each country. The The solar products available in East role to play. However, this issue brief market situation is dynamic and this African countries are mostly imported focuses on solar home system provid- information will change with time. from China, and we interviewed two ers and does not address the financ- New players are emerging. For instance, in Kenya, Philips has

PAYG PROVIDERS IN KENYA AND TANZANIA (DECEMBER 2015–FEBRUARY 2016) TABLE 1 TABLE Market leaders (as measured by number of Country installations) Other SHS PAYG players

Kenya M-KOPA Raj Ushanga (local partner of Azuri Technologies)

BBOXX Sun Transfer

Tanzania Off-Grid Electric Tigo Tanzania (local partner of Azuri Technologies)

Mobisol M-KOPA

Source: WRI, based on interviews (see Annex I)

ISSUE BRIEF | June 2016 | 7 recently introduced a solar lamp (225,000 in Kenya), Off-Grid Electric PAYG product and Greenlight Planet has 100,000 installations (mainly in Solar home system is reportedly planning to introduce Tanzania), and Mobisol has 35,000 a range of SHSs. In Tanzania, there installations (25,000 in Tanzania). providers with PAYG are at least two companies currently piloting their solutions, namely, The companies are growing rapidly functionality are Simusolar and Sollatek Tanzania. in terms of sales. Our interviews, Companies are also seeking to conducted during the period April emerging as key expand into other countries in the 2015 to February 2016, indicated providers of off-grid East African region. For example, that M-KOPA had 100,000 systems M-KOPA is trying to expand into at the end of 2014 and added another electricity in Kenya Tanzania, Mobisol is trying to 200,000 during 2015. Off-Grid expand into Kenya, and Off-Grid Electric monthly sales figures were in and Tanzania. There Electric is trying to expand into the range of 10,000 systems. Mobisol Rwanda. Most of the companies had sold about 10,000 systems by are at least two large that provide PAYG technologies 2014 and added 25,000 systems and products are foreign owned and across Rwanda and Tanzania in the providers in each managed. The participation of locally course of 2015. owned and managed companies in country and several the PAYG sector is limited. Tables 2 and 3 summarize the prod- uct offerings of the market leaders in other competitors. In terms of installations, the market the two countries. leaders in Kenya and Tanzania are The companies are M-KOPA and Off-Grid Electric, Table 3 indicates that all the prod- respectively. The companies treat uct offerings (except for Mobisol’s growing rapidly in installation figures as confidential, largest system, which is typically not but our estimates, based on the field targeted at households) offer Tier 1 terms of sales. interviews, are that M-KOPA now (very low) and Tier 2 (low) levels of has about 300,000 installations energy access.

PRODUCT OFFERINGS OF THE TWO MARKET LEADERS IN KENYA TABLE 2 TABLE Company Product Payment terms Total cost

M-KOPA 8W system (4 x 1W LED + torch + 50 KES (US$0.48) per day for 365 days 21,750 KES (US$208) radio) 3,500 KES (US$33.5) upfront deposit

BBOXX 15W system (4 x 1W LED + radio) 950 KES per month (US$9) for 36 35,150 KES (US$336) months 950 KES (US$9) upfront deposit

50W system (4 x 1W LED + radio + 2,150 KES (US$21) per month for 36 79,550 KES (US$761) 19” LED TV) months 2,150 KES (US$21) upfront deposit

Source: WRI, based on interviews (see Annex I) Notes: The US$ amounts are based on the currency exchange as of Jan 31, 2016, obtained from OANDA. The appliances are typically sold along with the system.

8 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

PRODUCT OFFERINGS OF THE TWO MARKET LEADERS IN TANZANIA TABLE 3 TABLE Company Product Payment amount/frequency/ terms Total cost over three years

Off-Grid Electric 2*1W lights package 15,000 TZS (US$8.25) per month of 555,000 TZS (US$305.25) (fee-for-service or ▪▪subscription subscription model) ▪▪15,000 TZS (US$8.25) upfront 3*1W lights + radio package 22,000 TZS (US$12.1) per month of 814,000 TZS (US$447.7) ▪▪subscription ▪▪22,000 TZS (US$12.1) upfront 3*1W lights + radio + TV 30,000 TZS (US$16.5) per month of 1,110,000 TZS (US$610.5) package (the TV is not sold ▪▪subscription with the unit) ▪▪30,000 TZS (US$16.5) upfront Mobisol 80W system (3*2W LED + 49,100 TZS (US$27) per month for 36 1,895,600 TZS (US$1,043) (rent-to-own model) radio + torch + 19” LED TV) ▪▪months ▪▪128,000 TZS (US$70) upfront 120W system (3*2W LED + 62,900 TZS (US$35) per month for 36 2,460,400 TZS (US$1,353) torch + 24” LED TV) ▪▪months ▪▪196,000 TZS (US$108) upfront 200W system 82,900 TZS (US$46) per month for 36 3,261,400 TZS (US$1,794) (village shop system) ▪▪months ▪▪277,000 TZS (US$152) upfront Source: WRI, based on interviews (see Annex I) Notes: The US$ amounts are based on the currency exchange as of Jan 31, 2016, obtained from OANDA. The appliances are sold with the system, unless noted otherwise.

While the main product offered by all companies is essentially a SHS, at The market leaders—M-KOPA in Kenya and the time we conducted the interviews there was one key difference between Off-Grid Electric in Tanzania—offer smaller Off-Grid Electric and the others. All the other major companies offer a products. The main competitors—BBOXX in “rent-to-own” plan, where owner- ship of the system is transferred to Kenya and Mobisol in Tanzania—offer larger the end customer after complete repayment of the loan. Off-Grid systems. However, most offerings are still at Electric, on the other hand, offers a the Tier 1 and Tier 2 level of energy access. “fee-for-service” model by providing an energy service against regular The customers’ payments are designated in payments but never transferring ownership of the system to the end local currency on a daily or a monthly basis. customer. Companies can, however, switch between business models. As

ISSUE BRIEF | June 2016 | 9 KEY FUNDS RAISED BY PAYG COMPANIES OPERATING IN KENYA

TABLE 4 TABLE AND TANZANIA (AS OF FEBRUARY 2016)

Company Details of funds raised

Azuri Technologies US$1.7 million: Barclays Social Innovation facility (2013)a ▪▪ b ▪▪US$0.75 million: AECF REACT Round 2 (2012) ▪▪Amount undisclosed: IP Group Plc, a VC fund (2012) BBOXX Capital US$15 million: previous backers and ENGIE and Ceniarth, Oikocredit (2015)c ▪▪US$3 million: DOEN and Bamboo Finance (2015)d ▪▪US$1.9 million: Synergy Growth and Khosla Impact (2013)d ▪▪ e ▪▪US$0.30 million: AECF REACT Round 2 (2013) Mobisol €10.7 million: DEG (2015)f ▪▪£288,000: GSMA grant (2013) ▪▪US$1.1 million: AECF REACT Round 2 (2013)g ▪▪Amount undisclosed: DEG (2012)h ▪▪ i ▪▪€200,000: EEP (2012) M-KOPA US$19 million: led by Generation Investment Management (2015)j ▪▪US$12.45 million: led by LGTVP (2015)d ▪▪US$20 million: syndicated by Commercial Bank of Africa (2014)d ▪▪£350,000: GSMA grant (2013) ▪▪ k ▪▪US$~1.9 million: led by LGTVP (2011) Off-Grid Electric US$45 million: Packard Foundation, Ceniarth, Calvert Foundation (2015) ▪▪US$25 million: led by DBL Partners (2015)d ▪▪US$7 million: IFC, Cordiant Capital (2015)d ▪▪US$16 million: led by SolarCity, Zouk Capital, and Vulcan Capital (2014)d ▪▪ d ▪▪US$7 million: Vulcan Capital, SolarCity, Omidyar Network (2014) Devergy (a mini grid company in Tanzania) £350,000: GSMA grant (2015) ▪▪US$809,000: led by Acumen (2015)l ▪▪ m ▪▪€115,000: DOEN project subsidy (2012) Powerhive (a mini grid company based in Kenya) US$20 million: led by Prelude Ventures LLC (2016)d ▪▪ d ▪▪US$12 million: Enel Green Power (2015) Source: WRI, based on interviews validated by online research (see Annex I) Notes: a Barclays Bank PLC, “Financing Innovation,” n.d. Retrieved on January 3, 2016, from: https://www.barclayscorporate.com/why-barclays/supporting-uk-business/azuri.html#. b Ashden Award Winner, “Case Study Summary Azuri Technologies, Africa,” 2013. Retrieved January 3, 2016, from: https://www.ashden.org/files/Azuriwinner.pdf. c PV Magazine, “BBOXX Secures US$15m Funding for Africa Solar Ventures,” 2015. Retrieved January 3, 2016, from: http://www.pv-magazine.com/news/details/beitrag/bboxx- secures-15m-funding-for-africa-solar-ventures_100022514/#axzz3x1GppyKz. d Crunchbase. e AECF REACT Round 2, “BBOXX Solar Franchising Model in East Africa,” 2013. Retrieved on January 3, 2016, from: http://www.aecfafrica.org/sites/default/files/project_files/BBOXX%20LTD.pdf. f DEG Investment in Mobisol GmbH, 2015: Retrieved on January 3, 2016, from: https://www.deginvest.de/DEG-Documents-in-English/About-DEG/Responsibility/Investment-related- information/201507_Mobilsol_EN.pdf. g AECF REACT Round 2, “Mobisol—Affordable and Sustainable Solar PV,” 2013. Retrieved on January 3, 2016, from: http://www.aecfafrica.org/sites/default/files/project_files/Mobisol-GmbH.pdf h Africa-EU Renewable Energy Cooperation Programme (RECP), “DEG-Direct Investments,” 2012. Retrieved on January 3, 2016, from http://www.africa-eu-renewables.org/_funds/deg/. i Mobisol Presentation, EEP Project of the Year Finalist,” 2015 Retrieved on January 3, 2016 from http://eepafrica.org/wp-content/uploads/151202-Mobisol-presentation-EEP-PotY-pitch-v1-0_ External.pdf. j PV Magazine, “M-KOPA Solar Secures US$19m Equity Investment,” 2011. Retrieved on January 3, 2016, from http://www.pv-magazine.com/news/details/beitrag/m-kopa-solar- secures-19-million-equity-investment_100022256/#axzz41e3ShTFS. k M-KOPA, “M-KOPA Concludes Series A Investment,” 2011. Retrieved on January 3, 2016, from: http://www.m-kopa.com/updates/m-kopa-concludes-series-a-investment/ LGTVP (US$500k debt, 37.5k equity), http://www.lgtvp.com/HOPS/Organisation/M-Kopa.aspx, Acumen (US$1.1m) http://acumen.org/investment/m-kopa/ and d.o.b Foundation (€240k/US$260k) http://www.dobequity.nl/east-africa-fund/m-kopa-kenya/. l Africa Capital Digest, “Acumen Leads Group Backing Solar Firm Devergy,” 2015. Retrieved on January 3, 2016, from: http://africacapitaldigest.com/acumen-leads-group-backing-solar-firm- devergy/, Acumen (US$700k) http://acumen.org/investment/devergy/, OPES Impact Fund (€100k/US$109k) http://www.opesfund.eu/file/investimenti_doc/investimenti_9_2.pdf. m Stitching DOEN, 2012: Annual Report, 2012. Retrieved on January 3, 2016, from: http://services.gdl-webservices.nl/DOEN/jaarverslagen/2012UK/magazine.html#/spreadview/26/.

10 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

of September 2016, we understand gies. Based on the initial success Packard Foundation, Ceniarth, that Off-Grid Electric had switched of these pilots (largely funded by ▪▪ Calvert Foundation debt facil- to a rent-to-own model while grants), companies have been able ity of US$45 million to Off-Grid BBOXX was considering a fee-for- to raise significant equity rounds. Electric. USAID provided a grant service model. Increasing confidence in the model of US$5 million of the US$45 is further evidenced by the fact that million (2015). Success to Date in Raising debt investments also have been International Capital recently forthcoming. The key debt Of the seven debt deals in the investments are summarized below. sector, five were struck in 2015. PAYG companies have raised sig- This excludes the Africa Enterprise nificant capital from international ▪▪ Barclays facility to Azuri of Challenge Fund (AECF) investments investors in the last few years in the US$1.7 million (2013). disbursed in the form of repayable form of grants (for developing and Commercial Bank of Africa’s syn- grants (which are treated as zero- testing products and markets) and, ▪▪ interest rate loans in the books more recently, in the form of equity dicated debt facility to M-KOPA of US$20 million (2014). of the company). Financiers are capital (for growth and scaling up). responding to the need for debt In 2015, on the strength of several DEG’s loan to Mobisol of undis- capital for scale-up. However, these previous equity rounds, companies ▪▪ closed amount (this debt invest- international investors generally have also been able to raise debt for ment is part of the total €10.7 provide debt capital in “hard” funding their further scale-up and million in 2015). currencies. All the debt deals growth. In Table 4, we summarize (except the debt deal syndicated by the key fund raises of PAYG compa- Debt component of US$6 million the Commercial Bank of Africa for nies operating in Kenya and Tan- ▪▪ in the LGTVP-led round invest- M-KOPA) are in foreign currency. The zania. In order to demonstrate the ment to M-KOPA (2015). debt deal syndicated by Commercial strong investor interest in this sector, Bank of Africa was facilitated by a Oikocredit securitization deal of we have also included the fund raises guarantee from the Gates Foundation ▪▪ US$0.5 million with BBOXX (2015). of two mini grid companies operat- (Bill and Melinda Gates Foundation ing in these two countries. There is IFC (US$4.5 million), Cordiant 2016). In fact, none of the interviewed significant interest among donors ▪▪ Capital (US$2.5 million) debt to companies had secured letters of in supporting the development of Off-Grid Electric (2015). credit or any other form of short-term products and market-testing strate- trade finance from local banks.

There is now a strong community of international investors in the PAYG sector. In our analysis, 17 foundations, 21 impact funds, four venture capital funds, two corporate venture capital funds, and eight large companies have invested in PAYG companies in East Africa. Their details are available in Annex III along with information on nine prominent angel investors. The strong interest among international investors is in stark contrast to the absence of local investor activity.

ISSUE BRIEF | June 2016 | 11 TYPICAL STRUCTURE OF CAPITAL MOVEMENT IN PAYG COMPANIES SURVEYED FIGURE 1

EQUITY, DEBT, OR GRANT (IN “HARD” CURRENCY)

A tax-efficient structure for PAYG parent company in Western investment purposes Europe or United States

Parent company and offshore vehicle invests in in-country subsidiary via equity or debt

PAYG subsidiary in Kenya or Tanzania

Potential local investment

Source: WRI, based on interviews (see Annex I)

Given that the leading investors are ratios. PAYG companies are using 3. The PAYG company would be mostly international, PAYG compa- securitization (the pooling together responsible for bringing in other nies are developing complex financial of cash flows originating from con- active customers and for non structures to raise required capital. tractual agreements and selling the payment. The PAYG company We have already noted that most of cash flows to investors as tradable would need to cover higher-than- the PAYG companies operating in financial securities) as an off-bal- expected defaults if necessary. Kenya and Tanzania have foreign ance-sheet financing instrument. A ownership and management. The securitization deal has already been This kind of arrangement is depicted parent companies are all incorporated executed by Oikocredit with BBOXX. in Figure 2. overseas. Figure 1 depicts a typical A typical securitization deal would structure and the way various types of work in the following manner: In the case of the Oikocredit deal capital move. with BBOXX, the first issue of securi- 1. The PAYG company “sells” active ties has bundled 2,500 active con- customer contracts to a special Given the need for capital in the tracts and raised 52 million Kenyan purpose vehicle (SPV). The sale shillings (US$500,000). The notes sector, one possible option is “off- value of the customer contracts balance-sheet” financing (where have an interest rate of 21 percent would reflect a discount based on and an average maturity of 2.5 years the debt is not on the books of the the expectation of defaults. company) and does not affect the (Hirtenstein 2012). debt equity or other leveraging 2. The SPV would issue “notes” to investors.

12 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

Tanzanian shilling, on five occasions, depreciated 5 percent or more STRUCTURE OF A TYPICAL SECURITIZATION DEAL against the U.S. dollar within one year. The largest depreciation in one FIGURE 2 year was about 17 percent in 2015. PAYG COMPANY Contract Companies can either absorb the Sells active increased debt-servicing costs or Sale price customer CUSTOMER and profit pass them on in the form of higher contracts rates for end customers, who pay in local currency. Indeed, in January SPV 2016, we observed a price increase for consumers. To understand the extent of the increased costs, we Sales price of Notes compare the product offerings of active customer contracts the market leaders in Kenya and in Tanzania in Table 5.

INVESTOR The currency risks point to the importance of borrowing in local cur- Source: WRI, based on discussion in Nairobi convening, July 2016 rency. Foreign lenders rarely provide local currency loans. Interest rates for local currency loans are high in Kenya and Tanzania, but the saved hedging costs can outweigh the dif- ference in coupon rates. Risk-mitiga- tion mechanisms exist almost exclu- sively for high transaction volumes Challenges with Foreign These risks are tangible. The Kenyan (Muench and Issler 2015). These Currency Debt shilling and the Tanzanian shilling mechanisms are often not practical have been volatile against the dollar. for PAYG companies, given their While the tactics devised by PAYG In Annex IV, we track the movement relatively small size, and actively companies to secure capital may of the Kenyan and the Tanzanian deter the entry of other private sector have been effective thus far, they currencies against the U.S. dollar players, who may not have the capac- are not without risks. There are two since 2004. The Kenyan shilling fell ity to borrow from international main challenges associated with about 8 percent and the Tanzanian markets. However, there are efforts foreign currency debt: shilling fell by about 17 percent from underway to make hedging facilities The structures are complex with April 2014 to April 2015. From April available to energy access companies. ▪▪ high transaction costs (involving 2015 to January 2016, the currencies In particular, the German environ- legal and tax advisory services). depreciated a further 8 percent and ment ministry has announced that 10 percent, respectively. In the entire it will make available €30 million The PAYG companies are period from April 2004 to January through KfW development bank ▪▪ exposed to currency risks if 2016, the Kenyan shilling, on four to the hedging specialist TCX, to the local currency depreciates occasions, depreciated 5 percent or be used for projects that promote because their assets (customer more against the U.S. dollar within renewable energy and energy effi- payments) are in local curren- one year. The largest depreciation ciency investments in sub-Saharan cies. in one year was about 21 percent Africa (The Lab 2016). The currency in 2009. In the same period, the hedging products are offered through

ISSUE BRIEF | June 2016 | 13 COMPARISON OF PRICE INCREASES IN THE MARKET-LEADING PRODUCT

TABLE 5 TABLE OFFERINGS IN KENYA AND TANZANIA

Total increase of customer payments 2015 package price 2016 package price in January 2016 over previous year M-KOPA, KENYA

40 KES per day for 365 days 50 KES per day for 365 days 24% Upfront: 2,999 KES Upfront: 3,500 KES

OFF-GRID ELECTRIC, TANZANIA

12,000 TZS per month of subscription 15,000 TZS per month of subscription 25% 12,000 TZS upfront 15,000 TZS upfront

20,000 TZS per month of subscription 22,000 TZS per month of subscription 10% 20,000 TZS upfront 22,000 TZS upfront

Source: WRI, based on interviews (see Annex I)

MFX Solutions, a dedicated currency The energy addressable market of raising exclusively international hedging facility for the microfinance, defined above is in terms of popu- capital. We have made some broad SME, and impact investment sectors. lation. PAYG companies provide assumptions regarding product costs MFX products can be accessed either solar home systems, at a rate of one to provide a rough estimate of the by the investor or the company and, system per household. Assuming a effective interest costs to customers, unlike more commercial hedging typical household size of five, the which are presented in Table 7. The transactions, can provide contracts addressable market is 5.7 million high effective interest rate is partly for relatively small-value transac- households in Kenya and 5.3 million explained by the need for the compa- tions (MFX Solutions 2013). households in Tanzania. Adoption nies to recover initial investment in of solar home systems in Kenya and product development and distribu- The Competitive Dynamics Tanzania is growing rapidly but tion infrastructure. the addressable market is far from In Kenya, 35 million people live penetrated (the current penetration A large market, enthusiastic cus- without electricity. In Tanzania, figures are in the thousands whereas tomer acceptance (rapid sales the number is 36 million people the market is in the millions). growth indicates that customers (IEA 2014a). PAYG represents seem willing to pay the effective an opportunity to provide energy From the customer’s point of view, high interest rates because they see access to a significant part of these the financial offerings of the PAYG the value of the product over other populations. The GSMA defines the companies are expensive. In part, options), and a level of profitability addressable market as people living these high costs reflect the risk that adequately covers the risk of without access to modern electricity of lending to hitherto unbanked lending to unbanked customers are and water but living within range customers. They also reflect the lack factors likely to lure new entrants. In of Global System for Mobile (GSM) of general manufacturing infra- our analysis, entry barriers are likely communication networks (within structure in these two countries, as to fall because of three developments range of mobile coverage), as shown well as the high transaction costs in the PAYG ecosystem, illustrated in in Table 6. Figure 3.

14 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

ADDRESSABLE MARKETS IN KENYA AND TANZANIA TABLE 6 TABLE

Mobile connection Country GSM population penetration Electrification rate Energy-addressable coverage (2013) (4th quarter, 2012) (2012) market (2013)

Kenya 68% 68% 23% 28,398,473

Tanzania 76% 55% 15% 26,663,260

Source: Nique 2013

ESTIMATES OF EFFECTIVE INTEREST RATE PAID BY PAYG CUSTOMERS FOR

TABLE 7 TABLE TWO TYPES OF SYSTEM

System size Small (8W) Medium (15W)

Estimated system cost including import costs, 16,000 KES (US$153) 19,850 KES (US$190) dealer margin, and inclusion of GSM chip

Customer payments for PAYG 50 KES (US$0.50) per day for 365 days 950 KES per month (US$9) for 36 months 3,500 KES (US$34) upfront 950 KES (US$9) upfront

Internal rate of return (IRR) over lease period 125% 52%

Source: WRI Notes: Assumptions based on data from product pricing data, cost prices based on enquiries from Chinese suppliers (see Annex I), internal rate of return calculated using XIRR function in Excel.

ISSUE BRIEF | June 2016 | 15 THREE REASONS WHY ENTRY BARRIERS SHOULD FALL IN THE PAYG SECTOR FIGURE 3

Companies can implement PAYG without investing in technology development. TECHNOLOGY ▪▪ Service providers: Angaza Designs (San Franciso, Nairobi), Lumeter Networks (California). OUTSOURCING ▪▪

MOBILE INTERFACE ▪▪ Companies can integrate with mobile and mobile money networks without building the necessary interfaces themselves. OUTSOURCING ▪▪ Service providers: KopoKopo (Seattle, Nairobi), Maxcom (Dar es Salaam).

CREDIT SCORE ▪▪ Companies can outsource customer credit-risk scoring to third parties. OUTSOURCING ▪▪ Service providers: InVenture (Santa Monica, with office in Kenya), First Access (New York).

Source: WRI, based on literature review and interviews (see Annex I)

There are already examples of capital is available almost exclu- SECTION II: STIMULATING partnerships between key ecosystem sively from international investors players and PAYG operators. Angaza (whether in the form of debt, impact LOCAL DEBT has already worked with d.light and equity capital, or grant) makes it The Importance of Local Debt with Philips to offer their solar lamp particularly difficult for local com- Previous WRI research has indicated through a PAYG service. KopoKopo panies. Local businesses often lack the importance of publicly supported has worked with BBOXX in Kenya. the skills, networks, and/or initial efforts to raise awareness and build resources needed to raise capital in capacity for distributed renewable In order to provide energy access to international markets, particularly if energy lending among local financial this large population in an afford- complex financial structures have to institutions. The research has led to able manner, it is necessary to create be developed. If finance were locally the recommendation that interna- market conditions that allow the available for investment in market- tional initiatives, as well as national entry of a large number of players. ing and distribution and to provide government action, could help the Growing competition should increase asset finance, then the market would sector develop by introducing a suite the product offerings available to attract a larger number of players, of policy and regulatory measures unserved and underserved popula- including local companies. A com- aimed at unlocking domestic com- tions and reduce overall costs. The petitive market place, with effective mercial financing for distributed technical barriers of entry to the monitoring and verification systems, renewable energy (Doukas and PAYG market are already falling. would enable the provision of energy Ballesteros 2015). There are two business barriers to access solutions to the underserved entry: the need for PAYG companies populations of Kenya and Tanzania International DFIs and donors have to have an extensive marketing and with the service attributes of reliabil- a long track record in helping to distribution network and the ability ity, affordability, legality, and health build the capacity of local financial to finance customer deployments and safety. institutions for specific sectors. in large numbers. The fact that

16 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

CAPITAL NEEDED FOR THE PAYG SECTOR IN KENYA AND TANZANIA TABLE 8 TABLE Kenya Tanzania

Addressable market for PAYG 5.7 million households 5.3 million households

Assuming a market penetration of 35% 2 million households 1.9 million households

ASSUMING THAT:

10% of households have Tier 1 (very low) energy access with a SHS costing US$150 US$30 million US$28.5 million

15% of households have Tier 2 (low) energy access with a SHS costing US$600 US$180 million US$171 million

10% of households have Tier 3 (medium) energy access with a SHS costing US$1,500 US$300 million US$285 million

TOTAL US$510 MILLION US$485 MILLION

Source: WRI, based on estimated system prices and stated assumptions.

Indeed, they have a long relationship Company (FMO), KfW Banken- To obtain the system costs, we made with banks in Kenya and Tanza- gruppe, Deutsche Investitions- und some broad assumptions about the nia. In Annex VII, we provide the Entwicklungsgesellschaft mbH landed cost of systems in Kenya and details of lines of credit provided by (DEG), and Oesterreichische Ent- Tanzania (at the point of customer DFIs to banks in Kenya and Tan- wicklungsbank AG (OeEB). The China interface). The US$150 system cost zania. Fifteen banks in Kenya have Development Bank and the African is based on an 8W–10W system. The existing active lines of credit with Development Bank also have partner- US$600 cost is based on a 50W sys- international DFIs. Most of these ships with existing banks in Kenya tem and the US$1,500 cost is based lines of credit are for general SME and Tanzania. on projected costs of a 200W system development, but recently Agence (such a system is currently not avail- Française de Développement (AFD) The institutional relationships able in these two countries). has launched two green lines of between international DFIs and credit: one with Co-operative Bank local banks are especially important The indicative figures in Table and the other with Chase Bank. In when we consider the amount of 8 are meant to serve only as an Tanzania, 10 banks have lines of financing needed to provide energy illustration that a substantial credit with international DFIs. All access in Kenya and Tanzania. We amount of capital is required the key international development have already seen that the GSMA for the PAYG sector to scale to finance institutions have a presence estimates the PAYG addressable real impact. The rough and ready in East Africa: AFD and Proparco, market at 5.7 million households in model indicates that about one European Investment Bank (EIB), Kenya and 5.3 million households in billion dollars is necessary for the the World Bank, the International Tanzania. How much capital would PAYG sector to have a substantial Finance Corporation (IFC), the the PAYG sector need in order to impact, especially if a reasonable Netherlands Development Finance make substantial contributions to proportion of its target households electricity access? are to have anything more than the most basic level of energy access.

ISSUE BRIEF | June 2016 | 17 Relying on their own means, indi- Bank (TIB). The loans to the participat- Providing a level playing field for vidual early-stage companies would ing financial institutions will be at a ▪▪ the household solar sector. find it extremely challenging to raise low interest rate to create an incen- this amount of capital and would tive in the form of potential higher ▪▪ Protecting customers by holding probably be able to provide only the margins. TIB will also share in the solar system providers accountable most basic energy access service to credit risk of the underlying renewable and enforcing quality standards. a smaller proportion of the popula- energy company loans. It is expected Increasing customer awareness. tion. It is important to leverage that both working capital and term ▪▪ existing financial institutions to loans would be made available to the Facilitating the import of solar facilitate a capital flow of this rural renewable energy companies. ▪▪ equipment. magnitude. This requires the sup- An additional technical assistance port of both the national government component is included both for the Mobilizing access to finance. and the international community. renewable energy companies, to help ▪▪ them apply for loans and to set up PAYG The program envisages support Current DFI Initiatives systems, and for participating financial being available to countries following institutions, to build their capacity in the signing of the compact agree- Table 9 summarizes the range of initia- appraising these loan applications. The ment (DfID 2016). tives that international DFIs already credit line and the technical assistance employ in the area of energy access in should particularly help local Tanzanian The World Bank Group, in partner- Kenya and Tanzania. The specific initia- solar companies adopt PAYG methods. ship with the Global Off-Grid Light- tives indicated under each category are ing Association (GOGLA), is collect- examples and do not provide an exhaus- We believe that the type of finan- ing metrics from a sample of PAYG tive list. The table is intended to convey cial support proposed by the solar companies to standardize the range of initiatives and indicate where World Bank in Tanzania needs metrics applicable to the industry as the majority of the funding is going. to be expanded. Support given to a whole. The aim is to enable com- national governments to increase gen- mercial investors to understand the In terms of fund commitment, most eration capacity and grid infrastructure risks and opportunities of the sector. international development financial should be complemented by measures support goes to large-scale genera- to strengthen the capacity of local finan- Lessons from Bangladesh in tion capacity and grid expansion. To cial institutions that can provide the Stimulating Local Capital expand energy access, the connection necessary resources for private PAYG fee for rural households is reduced Previous WRI research has com- companies to provide energy access. for some period of time after the grid mented on the lessons that can be expands into new areas. learned from Bangladesh, which The Energy Africa program of the UK’s has made extensive progress in the Department for International Develop- The limitation, in our opinion, is that installation of SHSs. The country ment (DfID) also aims to address support to local financial institu- has created a centralized framework many of the issues we highlight in tions to develop lending programs that allows international finance to this issue brief. Launched in October for energy enterprises has not been flow through a national institution to 2015 and aimed at 14 priority countries explored actively in Kenya and Tan- energy access enterprises and serves as (including Kenya and Tanzania), the zania, as can be seen from the above an important gatekeeper of technical campaign aims to sign a country-spe- categories of activities. One excep- standards and other issues relevant to cific compact agreement that outlines tion is the US$10 million Short-term providers of clean energy access solu- policy actions required to foster Renewable Energy Companies Credit tions (Doukas and Ballesteros 2015). development of the household solar Line Program, part of the overall In terms of the number of installed sector. The policy areas identified under US$200 million World Bank sup- systems, the Bangladesh model has Energy Africa include the following: port to Tanzania. The program is been very successful. The program designed to make funds available to Removing policy uncertainty by was established in 2003 and, as of participating financial institutions ▪▪ including off-grid electrification April 2014, three million SHSs had through Tanzania’s TIB Development as part of the national electrifica- been installed, benefiting 13 million tion strategy.

18 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

INTERNATIONAL FINANCIAL INSTITUTIONAL SUPPORT TO FINANCE ENERGY

TABLE 9 TABLE ACCESS (FOCUS: KENYA AND TANZANIA)

SUPPORT TO CREATE ADDITIONAL GENERATION CAPACITY AND GRID EXPANSION ▪▪AFD (€50 million) support to Lake Turkana Wind Power (private sector IPP). ▪▪AFD (€60 million) support to KETRACO (Kenya Electricity Transmission Company) for transmission line infrastructure. ▪▪AFD (€67 million) support to TANESCO (Tanzania Electric Supply Company). ▪▪KfW (US$88.6 million), AFD (€170 million) support to KenGen (Kenya Electricity Generating Company) for geothermal power stations. ▪▪World Bank (US$330 million) support to Ministry of Energy and the Kenya Power and Lighting Company. ▪▪World Bank (US$250 million) support to Government of Kenya for the Kenya Power and Lighting Company. ▪▪World Bank (US$200 million) support to the Government of Tanzania for the Rural Energy Agency. SUPPORT FOR MINI GRIDS AFD (€33 million) for Kenya Power to install renewable energy in 23 mini grids. ▪▪KfW (€15 million grant) and GIZ (€7.5 million) support to the Kenyan government for mini grids in northern Kenya. ▪▪IFC support to Tanzania for mini grid technical standards and advisory work. ▪▪DfID financing for green mini grids (including grants to developers—the grants can also be used as interest rate subsidies and results- ▪▪based financing).

DIRECT SUPPORT TO PROJECTS AND COMPANIES Proposed European Commission (€75 million) support for convertible grants to close “financing gap.” The support is available to projects ▪▪in multiple global areas. ▪▪DEG (€10.7 million) direct grant, loan, and equity support to Mobisol GmbH ▪▪FMO (US$2 million) equity to Orb Energy. DIRECT SUPPORT TO MFIS PROVIDING LOANS TO ENERGY ACCESS OPIC (U.S. Overseas Private Investment Corporation) and SDC (Swiss Development and Cooperation Agency) support to the Participatory ▪▪Microfinance Group for Africa (PAMIGA) in May 2015.

INVESTMENT IN FUND OF FUNDS (THESE FUNDS TARGET SHS AND MINI GRID COMPANIES) ▪▪IFC (US$10 million) investment in the responsAbility Energy Access Fund. ▪▪DfID (£30 million) proposed commitment to the “Flexible Fund”. ▪▪DfID, along with donors from other countries (US$244 million), to the Africa Enterprise Challenge Fund (AECF). SUPPORT TO COMPETITIONS (THAT SUPPORT SHS AND MINI GRID PLAYERS) ▪▪DfID (£30 million) support to the GSMA for funding mobile enabled energy, water, and sanitation innovation in Africa and Asia. SUPPORT TO SOLAR HOME SYSTEM MARKET DfID support through EnDev (Energising Development) for results-based financing (operated in Tanzania by by the Dutch NGO, SNV, for pico solar). ▪▪DfID Energy Africa to generate the policy and market shifts necessary to overcome the barriers and rapidly accelerate the growth in the ▪▪African solar household industry. ▪▪The World Bank Group Harmonized Metrics for the Distributed Solar Industry.

Source: WRI, based on information from institutional websites and interviews (see Annex I)

ISSUE BRIEF | June 2016 | 19 people—equivalent to 9 percent of the total population (IDCOL 2014a). The Partner Organizations in Bangladesh

We will look at the Bangladesh model provide the product, the required after- in more detail to draw lessons for Kenya and Tanzania. Bangladesh set sales service, and the microcredit. In doing up IDCOL as a government-owned, non banking finance company. so, they build trust among consumers in IDCOL channels international capital to the renewable energy sector, and the same way that PAYG companies do in the SHS program has been its larg- East Africa. The PAYG companies have the est program to date. The Bangladesh government felt it necessary to set up additional advantage of reduced costs and a specialized organization because, in 2003, its existing financial insti- operations risks, achieved using technology. tutions did not have the required relationships with international DFIs and donors. However, IDCOL not only channeled funding, it also played a major role in development of the INTERNATIONAL DFI SUPPORT COMMITTED TO whole sector. No such organization, BANGLADESH SHS PROGRAM (ALL FIGURES IN with responsibility for both channel- 10 TABLE ing funds and developing the sector, US$ MILLION) exists in Kenya or Tanzania today. Despite these differences, the Bangla- Institution Credit Grant desh experience is worth investigat- ing, given the magnitude of its success World Bank 449 – and the striking similarity between the operations of the PAYG compa- JICA 103 – nies and those of IDCOL’s Partner Organizations (POs). ADB 88 –

The POs provide a one-stop shop, as KfW – 18.5 do the PAYG companies. They sell and install the solar home systems GTZ – 16.5 in rural households and collect the monthly payments for the system. SIDA – 7 These home visits give the POs the opportunity to provide any needed USAID – 2.4 technical service. DfID – 28 IDCOL channels funds from inter- GPOBA – 7 national DFIs and donors to these POs. How much money does it take? Others – 30.6 The aggregate nominal support from 2003 to 2015 of the various TOTAL 640 110 international financial institutions, categorized into debt and grant, in Source: WRI based on information from interviews of IDCOL management (see Annex I) the Bangladesh SHS program is sum- marized in Table 10.

20 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

The program was designed to attract A total of US$750 million has been market participants and become a self-sustaining model. The program committed, of which about 85 percent started with five Partner Organiza- tions and now has more than forty. is in the form of low-cost debt and the While it began with a capital subsidy to households and a grant to the remaining 15 percent in the form of grants. partners (based on the number of solar home systems installed), this Initially designed by the World Bank, the mode of operation was intended to program has been supported by all major be phased out over time. Indeed, in the last two years, the program has development partners. continued to sell and install SHSs without additional grant disburse- ment (Figure 4).

NUMBER OF SOLAR HOME SYSTEMS INSTALLED, CUMULATIVE CREDIT, AND GRANTS DISBURSED, 2013–2015 (US$) FIGURE 4

4,000,000 600

3,500,000 525 Credit and Grants Disbursed (US$ million)

3,000,000 450

2,500,000 375

2,000,000 300

1,500,000 225 Number of SHSs Installed 1,000,000 150

500,000 75

0 0 2013 2014 2015

Incremental SHS installed (nos.) Cumulative credit disbursed (US$million) Cumulative SHS installed (nos.) Cumulative grants disbursed (US$million)

Source: WRI, based on IDCOL annual report.

ISSUE BRIEF | June 2016 | 21 More information on IDCOL is The other critical lesson from the (Doukas and Ballesteros 2015; IFC presented in Annex V. The important IDCOL model is the way IDCOL 2012). The security in the IDCOL thing to note is that, in addition to provided financing to energy access model is achieved as follows: hypoth- channeling financing, IDCOL played enterprises. WRI (and other) ecation of receivables, hypotheca- additional roles to ensure energy researchers have often commented tion of solar assets in the field, and access with service attributes of on the requirements of collateral personal guarantee of the founders of affordability, reliability, and health that severely constrain the ability of the energy enterprise or a corporate and safety (Table 11). energy access enterprises to bor- guarantee. There is no requirement row from local financial institutions for additional collateral.

ROLES PLAYED BY IDCOL (IN ADDITION TO FINANCING) TABLE 11 TABLE IDCOL role in ensuring access Details

Results-based financing IDCOL provided results-based financing (known as the institutional development grant) per ▪▪system installed. In the last few years the grants have been phased out, but the results-based financing helped the ▪▪Bangladesh energy enterprises invest in infrastructure in rural areas.

Capital buy-down grants In 2002, when the costs of solar systems were high, IDCOL provided households a capital sub- ▪▪sidy to buy solar home systems. This grant has also been phased out.

Technical standards IDCOL set up an independent Technical Standards Committee to ensure product quality and ▪▪reliability.

Vendor certification IDCOL certified vendors by working with independent testing centers to ensure product quality ▪▪and reliability.

Installation areas IDCOL established that installations should be carried out only in off-grid areas to ensure that ▪▪off-grid populations have higher priority.

Environmental standards IDCOL enjoined its partner energy enterprises to provide battery recycling with the certified bat- ▪▪tery vendors to ensure health and safety.

Monitoring and verification IDCOL has 12 quality control offices, 130 quality inspectors, and 11 field auditors. IDCOL also ▪▪has a customer call center to receive complaints. The monitoring and verification function was necessary to ensure that the technical standards ▪▪were implemented.

Source: WRI, based on literature review and interviews.

22 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

Each IDCOL borrower (under the SHS program) maintains two accounts with IDCOL: DEFINITIONS BOX 2 BOX ▪▪ The proceeds account: Revenue Hypothecation is the practice whereby the debtor pledges an asset to secure a collections from households debt. Hypothecation is used for creating charge against the security of movable flow into this account as do the assets and the possession of the security remains with the borrower. disbursements from IDCOL. The energy enterprises fund all Collateral is property or some other asset (typically financial assets such as operating expenses from this cash deposits, stocks, or bonds) that a borrower offers as a way for a lender to account. The account is audited secure the loan. by IDCOL and there is a lien in The debt service coverage account works as an additional security measure favor of IDCOL in case of default. for lenders. It is a deposit that is equal to a given number of months of projected The debt service coverage debt service obligations (including both interest and principal). ▪▪ account: This account has to be maintained at twice the quarterly repayment to IDCOL.

By using these two cash flow accounts, IDCOL has been able to obtain a clearer view of underlying IDCOL was the sole external finan- ecosystem in the two countries, while cash flows and dispense with the cier in the SHS market in Bangla- building their capacities to discharge requirements of additional collateral. desh. Indeed, commercial banks these functions. in Bangladesh have been hesitant Adapting IDCOL for Kenya to finance the SHS providers. By A possible division of roles and and Tanzania contrast, in Kenya and Tanzania, as responsibilities is illustrated in The implementation of the IDCOL we have already observed, there is Table 12. model led to the distribution significant private sector interest. of three million SHSs to rural On the other hand, there is no one With the above distribution of roles households in Bangladesh. The organization that performs the apex and responsibilities in mind, we “market creation” has led to the financing and the technical roles will examine a possible structure to expansion of local manufacturing played by IDCOL. stimulate local debt capital, as shown capacities. For example, in 2002, in Figure 5 below. Before we do so, Bangladesh had no indigenous In our view, the best way to stimulate however, we will discuss the role battery manufacturer and now local debt in Kenya and Tanzania is that can be played by a results-based there are several (Rahimafrooz, to involve the commercial banks at financing program. For the first five Rimso Battery, Panna Battery, and the very beginning. This would make years, IDCOL provided a grant to Hamko Group are the larger battery the financing of the SHS market the PO for every SHS installed. The manufacturers). In Annex VI, we sustainable and scalable because, amount of the grant was reduced present a comparison of prices for after the donors and DFIs withdraw, every year and was withdrawn from systems currently sold in Bangladesh commercial banks can lend to the the sixth year (see Annex V). Addi- and in Kenya and Tanzania. Despite market from their own funds, once tional useful lessons are now emerg- the success of the SHS market in they understand the risk-return ing from the EnDev (Energising Bangladesh, there is limited interest opportunities in the sector. The Development) results-based financ- on the part of commercial banks to various other roles of IDCOL should ing (RBF) program run in Tanzania finance SHS providers (Doukas and be housed within several existing by SNV, the leading Dutch NGO. Ballesteros 2015). players in the energy and financial

ISSUE BRIEF | June 2016 | 23 POSSIBLE DIVISION OF FINANCING ROLES AND RESPONSIBILITIES IN KENYA AND TANZANIA TABLE 12 TABLE

Role/responsibility Current state Next steps

Provide credit to the PAYG Major international DFIs have presence in both Leverage the existing relationships with Kenyan companies countries; 15 banks in Kenya and 10 banks in and Tanzanian banks and provide loss guarantee Tanzania have active lines of credit with interna- to stimulate lending, supplemented by lines of tional DFIs. credit for the off-grid sector.

Provide results-based financing SNV, the Dutch NGO, runs a results-based financ- Use results-based financing incentives, funded to the PAYG companies ing program in one area of Tanzania with funding by public and private donors, to help early-stage support from donors including GIZ and DfID. PAYG companies set up marketing and distribu- tion infrastructure in other areas of Kenya and Tanzania.

Implement technical and product The Lighting Africa program of the World Bank Develop the comprehensive testing and field-level standards Group (WBG) has developed quality standards monitoring capacity of public organizations in and certified vendors. Kenya and Tanzania.

Implement environmental No country specific standards exist. Work with the WBG’s Lighting Africa and Light- standards ing Global programs to develop standards and procedures and work with the Rural Electrification Authority (in Kenya) and the Rural Energy Agency (in Tanzania) to implement the standards across the country. Source: WRI analysis based on literature review and interviews.

In the Tanzania program, an RBF projections. The success of a results- In this scheme of things, interna- fund of €1 million was hosted by the based financing program depends tional DFIs and donors can play two Tanzania Investment Bank (Kleijn, on an in-depth assessment of the key roles: Sebastian, and Veen 2016). The RBF market. Companies should have the fund was used to pay out incentives freedom to follow their business ▪▪ Leveraging existing partnerships that were earned by energy enter- models (therefore, such a facility with local banks, to develop prises for every verified sale. Results- should not be restricted to PAYG either specific lines of credit or based financing can help incentivize models). For donors, the advantage guarantee schemes. early-stage PAYG companies to is that the grant is linked to the Helping public organizations target specific areas. The companies end-user beneficiary. The combined ▪▪ in Kenya and Tanzania with are more willing to invest in market- experience in Bangladesh and Tanza- efficient implementation of ing and distribution infrastructure nia seems to indicate that companies monitoring and verification (which, as we have noted earlier, is will not vacate a target area when measures. a key remaining entry barrier, along the grants are removed because they with finance), based on their sales have already invested in the requisite sales infrastructure.

24 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

renewable energy products and the repayment capacity of rural A POSSIBLE STRUCTURE TO STIMULATE LOCAL customers. The PAYG companies DEBT CAPITAL IN KENYA AND TANZANIA are young and often cannot meet FIGURE 5 the track record and collateral requirements set by the banks. Banks INTERNATIONAL DFIS AND DONORS also perceive a significant risk that consumers will default on payment to the PAYG provider, if the grid is extended to the household during the primary lease period. A multipronged approach will be required to help banks overcome these challenges: LOCAL BANKS PUBLIC ORGANIZATIONS Lines of credit/ Help banks understand the risks Standards and monitoring and verification ▪▪ guarantees and opportunities in the sec- tor and the business models. In particular, help banks under- stand the cash flow patterns of PAYG companies so that they can define more accurate meth- Customers PAYG companies ods, based on the data, of valuing the solar assets for collateral and loan-servicing purposes.

Source: WRI analysis based on literature review and interviews. Help banks become more com- ▪▪ fortable with the PAYG compa- nies by providing trade finance that facilitates the procurement of equipment (such support can be provided with the extension We discuss each of these roles below. the opportunity in energy lending, of existing trade finance support and they could explore offering schemes of the African Develop- Helping Local Banks Develop loans to their retail customers to ment Bank) (AfDB n.d.). Lines of Credit buy renewable energy systems. We have already seen that previous loan Help banks take the initial In the workshops that we conducted ▪▪ programs of this kind, involving steps through a risk-guarantee in Kenya (July 2016) and Tanzania partnerships between financial scheme. (February 2016), we identified a few institutions and energy companies, key challenges that banks face in Help banks expand schemes for have not been particularly successful lending to the clean energy access ▪▪ asset finance through lines of and that the PAYG model filled the sector. Some banks do not see the credit provided by international void. However, it is possible that the sector either as commercially viable DFIs. growth of the PAYG companies will or as a fit with their priorities. make retail financial institutions Help make grid expansion plans However, even banks that do have take a renewed look at this market. ▪▪ publicly available so that the an interest in the energy sector may However, even the banks that place PAYG companies can provide not be enthusiastic about financing more emphasis on corporate banking their systems in areas where PAYG companies. For banks with (and therefore could look at the the grid is not scheduled to be a large retail presence, the success PAYG companies) do not have the extended during the repayment of the PAYG model could reveal technical capacity to understand period for the system.

ISSUE BRIEF | June 2016 | 25 Guarantee schemes already exist in Income recognition (how much Helping public organizations Kenya and Tanzania, for example, ▪▪ income is recognized and when?) implement monitoring and the Development Credit Authority verification capacity program of USAID (USAID 2015) ▪▪ Provision for bad debts (when and SIDA’s guarantee instrument and how much provision is Monitoring and verification are (SIDA 2012), but for a guarantee needed for delayed payments?) critical to ensure that energy access scheme to be truly effective for a services have the necessary attributes Capitalization of expenses (how specific sector, the exact terms of a of affordability, reliability, legality, ▪▪ much of marketing and R&D guarantee scheme should be part and health and safety. Monitoring expenses can be capitalized?) of an overall financing strategy and and verification is also a critical risk- mitigation measure for the industry. be carefully negotiated between the Stock and capital asset Customers are less likely to default if international donor and the local ▪▪ accounting (how are assets their products function well and they financial institution. The guarantee at customer sites and future are receiving the required levels of should be an effective risk-mitigation receivables accounted?) measure but at the same time ensure service. Independent monitoring and that the recipient institution is using Grants accounting verification schemes would also help the opportunity to build its appraisal ▪▪ other donors assess the effectiveness and collection skills in the sector. of their support. The key guarantee terms include the amount of credit that can be deployed against a guarantee amount on a year-on-year portfolio basis, the The international DFIs have existing maximum individual loan size, the share of the losses that will be borne, relationships with banks in Kenya and the events that need to occur both for the origination of the loan and Tanzania and should leverage and the revocation of the guarantee and the costs of the guarantee. these relationships to develop lines

If banks are to lend to PAYG com- of credit or guarantee programs. panies, they need confidence in the cash flows of these companies and The banks could utilize the risk faith in their accounting standards. The World Bank Group’s project on guarantee and deploy the initial Harmonized Metrics for the Distrib- uted Solar Industry, in partnership capital and develop relationships with industry associations such as GOGLA, should actively work with with PAYG companies. As their the relevant national accounting standard bodies—the Institute of understanding of the sector grows, Certified Public Accountants in Kenya, and the National Board of they would feel confident to expand Accountants and Auditors in Tanza- nia—to define clear guidelines for the the scope of the funding. following issues:

26 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

The WBG’s Lighting Africa program The key consideration when assign- CONCLUSIONS has already made progress in defin- ing responsibilities to different ing standards and certifying products entities is efficiency. To this end, we In this issue brief, we have seen that and vendors. Their effective imple- recommend that the local banks take PAYG technology has significant mentation requires the creation of a responsibility for implementation advantages in terms of helping country-wide monitoring infrastruc- of energy access financing, while consumers at the “bottom of the ture. In the case of battery recycling, public organizations ensure that pyramid,” who are accustomed to standards and procedures should be systems to maintain specified quality buying energy in small increments, developed that are specific to Kenya standards are installed. The public switch over to renewable energy and Tanzania. organizations should have incentives systems. PAYG companies in Kenya for timely and transparent verifica- and Tanzania are currently offering The task of ensuring that the WBG’s tion. At the same time, verification products that meet the Tier 1 (very Lighting Africa standards are imple- processes should be applicable to all low) and Tier 2 (low) energy access mented on the ground and that players in the market. Otherwise, levels of the multi-tier framework standards for battery recycling are customers may run the risk of being for defining and measuring levels of developed and implemented could be offered products of unverified quality energy access. Even at these levels, the responsibility of public organiza- from vendors who have not had to numerous benefits are realized, tions in each country. The key public incur the overhead costs of under- largely through substitution of kero- organizations are the Bureau of Stan- going certification processes. One sene and other fuels. The product dards in Kenya and Tanzania, the possible solution is to have the REAs offerings are currently expensive in Rural Electrification Authority (REA) be in charge of the overall rural elec- terms of the effective interest rate in Kenya, and the Rural Energy trification project with the financial paid by the customer. Nonetheless, Agency (REA) in Tanzania. The REA facility housed in specified commer- target customers seem to accept in Kenya was established, under the cial banks. Whatever arrangement the products, and PAYG company provisions of the Energy Act No. 12 each individual country chooses, the sales are reportedly growing fast. of 2006, as a body corporate under key consideration is to develop an PAYG companies have also attracted the Ministry of Energy. The REA in inter-organizational coordination international private sector investor Tanzania is an autonomous body mechanism, and efficient and trans- interest and have been successful in under the Ministry of Energy and parent systems and processes. Above raising grants (for product develop- Minerals. Both these organizations all, efficient coordination requires ment and launches), equity capital have the task of promoting rural political will and an understanding (for business development), and debt electrification in their countries. that providing energy access with the (for scale-up as business models have They have experience of renewable attributes of affordability, reliability, been proven). However, this interna- energy and both have experience of legality and health and safety is a tional capital has not been matched working with the World Bank. The collective, national goal. by interest from domestic financial World Bank could devise a role for institutions. The dependence of the itself in helping these public bodies sector on international debt capital to implement the quality standards. providers, in particular, is risky because of foreign currency volatility. Companies can either absorb foreign currency fluctuations or pass them on to customers, who pay in local currencies. In turn, costs of raising international capital are high because transaction structures tend to be complex and this, in turn, can lead to offerings being more expensive than they otherwise would be.

ISSUE BRIEF | June 2016 | 27 To provide energy access to the large The IDCOL program of Bangladesh Results-based financing programs, populations in Kenya and Tanzania, has some very useful lessons for an important service performed by in an affordable manner, it will be Kenya and Tanzania. The POs of IDCOL, can be run by NGOs (SNV, necessary to create market condi- Bangladesh played the same role as the Dutch NGO, runs one such tions that allow the entry of a large the PAYG companies do in Kenya program in Tanzania). Results-based number of players. Greater competi- and Tanzania, providing a one- financing programs help companies tion should increase the number of stop shop for customers. Unlike in invest in the last-mile marketing and product offerings available to under- Bangladesh, however, where inter- distribution infrastructure that is served populations, make available national DFI and donor money was necessary to market and service rural and decrease the costs of auxiliary channeled through a non banking energy (including PAYG) systems. services, and reduce the overall risks financial institution that performed a To ensure that service attributes are (and therefore expected market wide variety of roles, we believe that, met, public organizations in Kenya returns) of market players. We argue in Kenya and Tanzania, commercial and Tanzania should take on the that many of the technical entry banks should be involved early on other roles performed by IDCOL. In barriers to the PAYG market are in channeling capital. Involving the particular, they can set up country- already falling and that the key entry commercial banks should aid in the wide monitoring and verification sys- barrier is the availability of finance sustainability and scale-up of the tems to ensure that products in the that would enable PAYG providers to sector because the financing would field adhere to WBG Lighting Global finance customer deployments. We not depend only on donor funds. and Lighting Africa standards. In argue that, if this source of finance The banks can adopt the methods some areas, such as standards for were locally available, it would pro- that IDCOL used to monitor cash recycling, standards would need to tect companies and customers from flow patterns of energy enterprises be developed. the fluctuations in foreign currency to define more accurate methods, markets. More important, the market based on the data, of valuing the As we have already seen, 17 founda- would attract a larger number of solar assets for collateral and loan- tions, 21 impact funds, four venture players, including local companies. servicing purposes. IDCOL played capital funds, two corporate ven- Local entrepreneurs find it particu- roles other than financing and these ture capital funds, and eight large larly challenging to navigate inter- additional roles were critical to companies have invested in PAYG national capital markets, especially ensuring access with the necessary companies in East Africa. DFIs when the financing structures are service attributes of affordability, should consider investing in a “fund complex. reliability, legality, and health and of funds” run by professional impact safety. In Kenya and Tanzania, fund managers, rather than taking The international DFIs and donors responsibility for these roles could the riskier option of investing in already have long-tanding relation- be handled by organizations in the individual companies. ships with banks in Kenya and public sector. Public sector organiza- Tanzania. They should leverage these tions—such as those responsible for relationships to help stimulate local rural electrification—can develop, lending in the PAYG sector. Local adapt, and monitor standards that banks often have limited under- protect not only consumers but also standing of the sector and an exag- investors and lenders. In addition, gerated sense of the risks of lending if grid expansion plans were to be to companies in the sector. These made publicly available, companies challenges can be addressed with could focus on areas where the credit-guarantee schemes, mecha- chances of the grid arriving during nisms to understand the underlying the repayment period of the system cash flow patterns of PAYG com- are small. panies, and technical assistance to companies preparing loan applica- tions and to banks seeking to under- stand loan applications.

28 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

RECOMMENDATIONS It is important that incentives International DFIs and donors and risk-mitigation measures are should use technical assistance to The solution to the challenge of simultaneously created as a package build the capacity of the rural electri- financing the scale-up of PAYG to stimulate local lending. Credit fication and standards organizations energy access lies not so much inthe lines should be appropriately priced in Kenya and Tanzania to define and development of new initiatives but to create the right incentive. The adapt standards for reliability, and in the use and redirection of existing exact terms of the credit guarantee health and safety, and implement approaches for PAYG, particularly scheme should be carefully discussed monitoring and verification schemes the use of credit guarantees, lines between the international and local on the ground to ensure adherence of credit, technical assistance, institutions to ensure that this is a to these standards. They should and investment in a “fund of meaningful risk-mitigation measure also support the rural electrification funds.” It does, however, require while still requiring local banks to authorities and encourage them to a coordinated approach, not just develop their appraisals and collec- make grid expansion plans publicly between international DFIs and tion mechanisms. Technical Assis- available. This would enable PAYG donor agencies but also between tance should be made available both companies to plan their expansion national government agencies to companies developing loan appli- and manage the risk of default in involved in rural electrification and cations and to banks that appraise the event that customers stop using private sector investors who are them so that both parties understand the PAYG systems when alternate increasing their investments in the the use of credit guarantee and systems become available. energy access area. We offer key risk management tools. Technical stakeholders the recommendations Assistance can include use of the the Finally, international donors should below: WBG’s Harmonized Metrics (devel- consider “crowding in” other private International DFIs and Donors opment of which is in progress) and sector foundations and family offices appropriate accounting standards to support results-based financing Our recommendation is that interna- to document industry-level PAYG programs. The results-based financ- tional development finance institu- repayment rates. In addition to ing programs should allow donors tions strengthen the capacity of the local banks that could provide debt, to ensure that their support is linked local financial system to provide the international DFIs and donors to end-user beneficiaries and allow debt capital to PAYG companies. The should support equity investment by companies the freedom to invest in international support for increased investing in impact funds run by pro- marketing and distribution infra- generation capacity and grid infra- fessional fund managers. The DFIs structure that would yield longer- structure should be complemented and donors should use their support term benefits. (not necessarily within the same to provide the correct market incen- project) by measures to strengthen tives to “crowd in” private sector Local Commercial Banks the capacity of local financial institu- investment. As we have already seen, We recommend that local commer- tions that can provide the necessary 17 foundations, 21 impact funds, four cial banks—especially those with a resources for PAYG private compa- venture capital funds, two corporate corporate and SME lending focus— nies to provide energy. An example is venture capital funds, and eight begin to explore the opportunities in the proposed World Bank support to large companies have invested in distributed renewable energy lend- Tanzania, where the overall package PAYG companies in East Africa. DFIs ing. Local commercial banks could includes a risk-sharing arrangement should supplement the efforts of start engaging with PAYG companies with the TIB Development Bank. professional impact fund managers (and other distributed renewable by addressing key gaps. Investing in energy companies) by providing funds would also be less risky than short-term trade finance. They supporting individual companies or should work toward understanding projects. the cash flow patterns of PAYG com-

ISSUE BRIEF | June 2016 | 29 panies and explore the use of mecha- Private Sector Investors Private Sector PAYG Businesses nisms such as a debt service coverage We recommend that private sec- Companies should develop and adopt account to partially cover default tor investors support companies accounting standards for income risks. Local banks should undertake in structuring their business and recognition, inventory valuation, a review of the sector to assess the financial models in ways that allow accounting treatment for grants, and exact components of credit, guar- companies to access different types bad debt recognition. This should antee, and technical assistance they of capital and partnerships in be done collaboratively with the would need. response to evolving company needs. World Bank Group’s Harmonized National Governments This includes the option of raising Metrics for the Distributed Solar capital from local commercial banks. Industry and local accounting We reemphasize previous WRI rec- Raising capital from local banks bodies (Institute of Certified Public ommendations that national govern- requires understanding the lending Accountants, Kenya; National ments should help the sector develop criteria followed by local banks and Board of Accountants and Auditors, by introducing a suite of policy examining whether private sector Tanzania). These accounting and regulatory measures aimed at investor expectations could actually standards would be of help when unlocking domestic commercial hinder the investee company in its working with commercial banks. financing for distributed renewable attempts to borrow from commer- energy (Doukas and Ballesteros cial banks. For instance, the private None of these recommendations 2015). In particular, for PAYG, the sector investor may be asking the should be considered in isolation. government can take the following company to invest in marketing for We strongly recommend that the key steps: future growth. This might cause stakeholders—international financial the company to incur losses, which institutions and donors engaging ▪▪ Send strong signals to private would make it difficult to access in the energy access space, private sector operators and the com- bank borrowing. In such an instance, sector investors and those designing mercial sector by making energy equity investors can set aside some clean energy policy and regulatory access a national priority in of their investment money as a time interventions, national and subna- which both private and public deposit with the bank, against which tional governments, and rural elec- actors must participate. their investee company can borrow. trification organizations—effectively Set up mechanisms for the coor- In the short term, this does not lead coordinate and engage with each ▪▪ dination of different institutions to any net cash flow for the company, other. Coordination and engagement in a transparent and efficient but it does provide an opportunity to will be essential in designing effec- manner to meet the energy develop a relationship with a com- tive support and interventions for access goals, then allow the mercial bank. distributed renewable energy and mechanisms to be run by inde- scaling up energy access where it is pendent professionals. Foundations and family offices can needed most. In particular, there is provide a guarantee to the local bank a need for a working group with a Absorb—fully or in part—the (on behalf of individual companies) functioning secretariat to convene ▪▪ foreign currency fluctuation risks or contribute to the loss guarantee the various stakeholders on a regular that would enable international of a fund that might be set up by basis and facilitate an action-driven DFI capital to flow into the com- a DFI. Foundations and family agenda. For efficiency, this work- mercial banking sector. offices should also consider provid- ing group could be aligned with an ing technical assistance (through existing structure or institution, such ▪▪ Make grid expansion plans professional capacity development as a donor coordination group or a publicly available so that private organizations) to help local compa- country-specific SE4ALL hub. The sector businesses can manage nies develop the business processes impact of our specific recommenda- their investment decisions. needed to scale their businesses. tions for each stakeholder group will be greatly amplified by effective coordination.

30 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

ANNEX I: METHODOLOGY

PAY-AS-YOU-GO SOLAR HOME SYSTEM COMPANIES INTERVIEWED TABLE A1 TABLE Company Person interviewed

Azuri Technologies Simon Bransfield, founder and CEO

BBOXX Limited Mansoor Hamayun, founder and CEO Christopher Baker-Brian, founder and CTO Joanis Holzigel, former managing director, BBOXX Capital Kenya Anshul Patel, VP projects, BBOXX Kenya

M-KOPA Jesse More, founder and managing director

Mobisol GmbH Thomas Duveau, head of business development Klaus Maier, corporate development manager Robert Zeidler, product manager, East Africa Saad Latif, development manager, Mobisol Tanzania

Offgrid Electric Graham Smith, vice president of business development Justin Heath, vice president sales Ari Zotloff, vice president new country expansion

Sun Transfer Dr. Gathu Kirubi

Interview Guide: Companies The purpose of the interview is to understand how the organization is structured, the key sources and types of financing, challenges faced in raising finance, and the product and service offering.

1. Where is your company head quartered? In which countries do you have operations, and how are you structured in the countries in which you operate?

2. What is the scale of your operations? Particularly in Kenya and Tanzania, what is the indicative volume of sales that you are achieving?

3. What have been the key fund raises so far? Who have been the main investors? In which currencies has the money been raised, and how has the money been deployed in the countries of operation?

4. What is the main product and service offering? Do you have only a pay-as-you-go offering? What are the consumer price points in local currency?

5. What is the supply chain? Do the main vendors of key components provide any credit?

ISSUE BRIEF | June 2016 | 31 DFIS AND DONORS INTERVIEWED TABLE A2 TABLE Organization Person interviewed Location AFD Diane Jegam Nairobi African Development Bank Walter Odero Nairobi DfID Sebastian Meanney London Steven Hunt London Sabita Thapa Nairobi Leanne Jones (Tanzania) Dar es Salaam European Commission Balthasar Klimbie, rural electrification expert The Hague IFC Andrew Abduel Mnzava Dar es Salaam FMO Marc Buiting Amsterdam GIZ Jasmin Fraatz Nairobi KfW Dr. Jens Drillisch Frankfurt Olive Muthoni Nairobi Viviana Klein Dar es Salaam Norway Katrine Vestbostad Dar es Salaam Monica Blaalid USAID Pamela Baldinger Washington, DC World Bank Monali Ranade Washington, DC Richard Hosier Washington, DC John Coleman MacLean (consultant) Seattle

Interview Guide: DFIs and Donors 1. The purpose of the interview is to understand your current programs in the energy access area in the countries of Kenya and Tanzania, your interest in decentralized renewable energy and the type of analysis that may be helpful in supporting your projects in this area.

2. What are your current projects in the area of energy access in Kenya and Tanzania?

3. Would you view solar home systems and mini-micro grids as viable options for energy access?

4. Which government electrification authority do you engage with?

5. What kind of analysis would help you support energy access through decentralized renewable energy?

32 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

BANGLADESH STAKEHOLDERS INTERVIEWED TABLE A3 TABLE Organization Organization category Person interviewed IDCOL Manager of the energy access program Nazmul Haque, director investment, head of advisory Farzana Rahman, VP and unit head, renewable energy Sadia Haque, relationship manager World Bank DFI supporter of IDCOL Zubair Sadeque, senior energy specialist JICA DFI supporter of IDCOL Zaki Ziaul Islam, deputy program manager DfID Donor supporter of IDCOL Roqibul Islam, private sector development, DfID GIZ Donor supporter of IDCOL Sajib Seb, technical advisor KfW Donor supporter of IDCOL Tazmilur Rahman, senior sector specialist Bright Green Energy Foundation IDCOL Partner Organization Dipal Baura, founder Rahima Frooz IDCOL Partner Organization Nitai Saha, general manager Syed Ishtiaque, head of sales

Interview Guide: Bangladesh Stakeholders 1. The purpose of the interview is to understand the factors that contributed to the success of the IDCOL program in Bangladesh, its current challenges, and lessons (if any) regarding how DFIs and donors can support energy access programs in East Africa.

2. How much DFI/donor money was needed for the IDCOL Solar Home System and how much of this was grant?

3. What were the yearly performance figures? What are the types of challenges being faced currently?

4. Who were the main Partner Organizations? How were they supported? What prompted their rapid expansion?

5. What were the processes for loan/grant approval and disbursement, security, technical validation, verification and monitoring?

6. What are the unique features of Bangladesh that made the program successful? Can this be replicated in East Africa? If not, are there lessons that can be derived?

REPRESENTATIVE CHINESE SUPPLIERS INTERVIEWED

TABLE A4 TABLE Organization Organization website Person interviewed

Bennu Solar www.bennu-solar.com Yotam Ariel, CEO

Omnivoltaic Power Co. www.omnivoltaic.com Xiaojun Kang, managing director

ISSUE BRIEF | June 2016 | 33 ATTENDEES AT THE NAIROBI ATTENDEES AT THE DAR ES WORKSHOP ON “STIMULATING SALAAM WORKSHOP ON TABLE A5 TABLE LOCAL CAPITAL FOR PAY- A6 TABLE “STIMULATING LOCAL CAPITAL AS-YOU-GO ENERGY ACCESS FOR PAY-AS-YOU-GO ENERGY COMPANIES,” JULY 2016 ACCESS COMPANIES,” FEBRUARY (REPRESENTATIVE SAMPLE) 2016 (REPRESENTATIVE SAMPLE)

Organization Name Organization Name DFIS AND DONORS DFIS AND DONORS IFC Arthur Itotia Njagi AFD Dennis Munuve DfID Sabita Thapa IFC Andrew Abduel Mnzava KfW Olive Muthoni Development Partners Earneus Kaijage EIB Nicholas Nzioka DfID Daniel Emmanuel European Union Sanne Willems KfW Viviana Klein USAID Benson Kimithi Norway Monica Blaalid LOCAL COMMERCIAL BANKS European Union Mikael Melin Citigroup Karanja Gichiri Sweden Jorgen Eriksson CfC Stanbic Bank Jeff Alondo World Bank Richard Hosier, Zubair Sadeque, John Coleman Equity Bank Eric Naivasha MacLean (consultant) Faulu MFB Bank Charles Ndungu LOCAL COMMERCIAL BANKS KCB Bank Zacharia K. Cheruiyot Bank of Africa, Tanzania Wasia Mushi, Deo NIC Bank Daniel Waweru Chalamila AMFI Paul Kihiu CBA Godfrey Munisi PRIVATE SECTOR INVESTORS CRDB Focus Mrosso AECF Ann Kitonga, Jeet Malde Equity Bank Joseph Ila, David Mukaru AlphaMundi Group Tim Radjy, David I&M Bank T. Srikanth Mutheee, William Githui Michael Mungure NMB Bank DOB Equity Mercy Mutua Stanbic Bank Angela Mandi Energy Access Ventures Emmanuel Beau GOVERNMENT Lendable Daniel Goldfarb Rural Electrification Boniface Gissima Lundin Foundation Charlotte Ward Agency Nyamo-Hanga Oikocredit Mark Roesink PRIVATE SECTOR INVESTORS Responsibility Alexander Bashian Sun Funder Lais Lona OTHER STAKEHOLDERS OTHER STAKEHOLDERS GVEP Leah Kaguara, Juliette Open Capital Advisors David Loew Page, Linda Onyango IMED Donath Olomi GSMA Ilana Cohen GVEP Shashank Verma MFX Solutions Luz Leyva SNV Martijin Veen Open Capital Advisors Andreas Zeller Practical Action Noah Mayieka Power for All Kate Montgomery TAREA Eng. Mathew Matimbwi TAREBI Fredrick Tunutu

34 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

ANNEX II: MULTI-TIER Under the Sustainable Energy for quality and extent of energy services All (SE4ALL) initiative, the World across several attributes. The MTF FRAMEWORK FOR Bank and the International Energy measures energy access in “tiers” ENERGY ACCESS Agency developed a Global Track- from 0 (no access) to 5 (high access). ing Framework to chart a course Multi-tier Framework to achieve universal energy access, ESMAP offers a more comprehensive double the use of renewable energy, definition of energy access that is The Energy Sector Management and improve energy efficiency. “meaningful for households, pro- Assistance Program (ESMAP) is a Building on SE4ALL’s Global Track- ductive enterprises and community trust fund with a global mandate ing Framework, and in consultation facilities…and [has] a number of to provide knowledge and techni- with development partners, ESMAP attributes: it must be adequate in cal assistance to low- and middle- developed the Multi-tier Framework quantity, available when needed, of income countries for environmen- (MTF) to evaluate and monitor prog- good quality, reliable, convenient, tally sustainable energy solutions. ress toward energy access goals. The affordable, legal, healthy, and safe” ESMAP is funded by 13 bilateral MTF introduces an energy access (Angelou and Bhatia 2015). donors and administered by the definition that incorporates the World Bank.

ISSUE BRIEF | June 2016 | 35 MULTI TIER MATRIX FOR ACCESS TO HOUSEHOLD ELECTRICITY SUPPLY, SERVICES AND CONSUMPTION TABLE A7 TABLE

TIER 0 TIER 1 TIER 2 TIER 3 TIER 4 TIER 5 1. Capacity Power* Very Low Power Low Power Medium High Power Very High Power Min 3 W Min 50 W Power Min 800 W Min 2 kWh Min 200 W AND Min 12 Wh Min 200 Wh Min 1.0 Min 3.4 Min 8.2 kWh

ATTRIBUTES Daily kWh kWh Capacity OR Lighting of Electrical lighting, Services 1,000 lmhrs per air circulation, day and phone television, and phone charging charging are possible 2. Duration Hours Min 4 hrs Min 4 hrs Min 8 hrs Min 16 hrs Min 23 hrs per day Hours per Min 1 hrs Min 2 hrs Min 3 hrs Min 4 hrs Min 4 hrs evening 3. Reliability Max 14 Max 3 disruptions per week disruptions of total duration <2 hours per week 4. Quality Voltage problems do not affect the use of desired appliances. 5. Affordability Cost of a standard consumption package of 365 kWh per annum is less than 5% of household income. 6. Legality Bill is paid to the utility, prepaid card seller, or authorized representative.

7. Health and Safety Absence of past accidents and perception of high risk in the future.

* The minimum power capacity ratings in watts are indicative, primarily by Tier 1 and 2, as the efficiency of end-user appliances is critical to determining the real level of capacity, and thus the type of electricity services that can be performed.

TIER 0 TIER 1 TIER 2 TIER 3 TIER 4 TIER 5 Tier criteria Not applicable Task lighting General lighting Tier 2 AND Tier 3 AND Tier 4 AND Phone charging Television Any medium-power Any high-power Any very high- Fan (if needed) appliances appliances power appliances

TIER 0 TIER 1 TIER 2 TIER 3 TIER 4 TIER 5 Annual consumption levels, in kilowatt-hours (kWh) <4.5 ≥4.5 ≥73 ≥365 ≥1,250 ≥3,000 Daily consumption levels, in watt-hours (Wh) <12 ≥12 ≥200 ≥1,000 ≥3,425 ≥8,219

Source: Angelou and Bhatia 2015.

36 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

ANNEX III: PRIVATE SECTOR INVESTOR PROFILES

BRIEF PROFILES OF PRIVATE SECTOR INVESTORS WHO HAVE INVESTED EITHER EQUITY OR GRANTS IN PAYG COMPANIES IN EAST AFRICA TABLE A8 TABLE

Office PAYG investees Investor Website location Profile (East Africa) FOUNDATIONS/FAMILY OFFICES 1 Alstom Foundation www.alstom.com/foundation Global Supports NGO projects proposed EGG Energy by employees that improve living conditions around Alstom sites 2 Calvert Foundation www.calvertfoundation.org Bethesda, MA, Affordable housing, education, Off-Grid Electric United States small business, microfinance 3 Ceniarth www.ceniarthllc.com London, New Energy access, agriculture, BBOXX, Off-Grid York, San innovation Electric Francisco 4 DOEN Foundation www.doen.nl/web/home-1.htm Amsterdam Green and socially inclusive BBOXX, Founda- businesses tion Rural Energy Services, M-KOPA 5 Godley Family Foundation www.godleyfamilyfoundation. Ponte Vedra Health, environment, and under- Angaza Designs org Beach, FL, United served communities States 6 Gates Foundation www.gatesfoundation.org Seattle Health, development, policy, and M-KOPA advocacy 7 Generation Investment www.generationim.com London, New York Sustainability M-KOPA Management

8 Hivos Foundation www.hivos.org The Hague Sustainable food, renewable Foundation Rural energy, transparency, and Energy Services accountability 9 Jasmine Social Investments www.jasmine.org.nz New Zealand Social entrepreneurs Off-Grid Electric 10 Lundin Foundation www.lundinfoundation.org Vancouver, SMEs M-KOPA Canada 11 Marshall Foundation www.marshallfoundation.com Tucson, AZ, Education, healthcare, youth- Off-Grid Electric United States oriented services 12 Mulago Foundation www.mulagofoundation.org San Franciscos Livelihoods, energy, health, Off-Grid Electric conservation, education 13 Packard Foundation www.packard.org Los Altos, CA, Conservation, science, popula- Off-Grid Electric United States tion, heath 14 Pi Investments www.pi-investments.com San Francisco Sustainable and just economy Powerhive Bay Area 15 Segal Family Foundation www.segalfamilyfoundation. Watchung, NJ, Health and youth Off-Grid Electric org United States 16 Shell Foundation www.shellfoundation.org Global Energy access, livelihoods, M-KOPA sustainable transport, agriculture 17 The World We Want www.theworldwewantfounda- Stockholm Energy, conservation, agricul- Off-Grid Electric Foundation tion.org ture, education

ISSUE BRIEF | June 2016 | 37 BRIEF PROFILES OF PRIVATE SECTOR INVESTORS WHO HAVE INVESTED EITHER EQUITY OR GRANTS IN PAYG COMPANIES IN EAST AFRICA (CONTINUED) TABLE A8 TABLE

Office PAYG investees Investor Website location Profile (East Africa) IMPACT FUNDS 1 Acumen Fund www.acumen.org Global Energy, agriculture, healthcare, M-KOPA, housing, water, and education SolarNow, Devergy 2 Better Ventures www.better.vc Oakland, CA, Mobile, health, sustainability SunFunder United States 3 Boma Investments www.bomainvests.com Santa Barbara, CA, Social and environmental Lumeter United States 4 Bamboo Finance www.bamboofinance.com Global Housing, healthcare, education, BBOXX, Greenlight energy, livelihoods, water, and Planet sanitation 5 Blue Haven Initiative www.bluehaveninitiative.com Geneva Affordable and green housing, M-KOPA education, energy and environ- ment, financial inclusion, health 6 Dob Equity www.dobequity.nl LA Veessen, The Socially and financially sustain- M-KOPA Netherlands; able African companies Nairobi 7 Energy Access Ventures www.eavafrica.com Paris, Nairobi Energy access Off-Grid Electric 8 Global Partnerships www.globalpartnerships.org Global Health, green tech- Greenlight Planet nology, rural livelihoods, microentrepreneurship

9 Grey Ghost Ventures www.greyghostventures.com Global Information and communica- M-KOPA tions technology (ICT), clean technology, and other adapted technology for underserved communities 10 Khosla Impact www.khoslaimpact.com Global Products and services for base BBOXX, SunFunder of the economic pyramid (BoP) 11 Invested Development www.investeddevelopment. Boston, Nairobi Mobile, ICT, alternative energy, EGG Energy com and agriculture 12 LGT Venture Partners www.lgtvp.com Global Health, education, energy, M-KOPA agriculture, ICT, livelihood, and housing 13 Novastar Ventures www.novastarventures.com Nairobi East Africa–based businesses SolarNow 14 Omidyar Networks www.omidyar.com Global ICT, education, financial inclu- Off-Grid Electric sion, policy and advocacy, and property rights 15 Oikocredit www.oikocredit.coop CC Amersfoort, Microfinance, fairtrade, agricul- BBOXX The Netherlands ture and renewable energy 16 OPES Impact Fund www.opesfund.eu Milan Energy and water access, Devergy education, waste management, agriculture, health

38 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

BRIEF PROFILES OF PRIVATE SECTOR INVESTORS WHO HAVE INVESTED EITHER EQUITY OR GRANTS IN PAYG COMPANIES IN EAST AFRICA (CONTINUED) TABLE A8 TABLE

Office PAYG investees Investor Website location Profile (East Africa) IMPACT FUNDS (CONTINUED) 17 Radicle Capital www.radiclecapital.com Louisville, KY, Agriculture, eco-friendly product Sunfunder United States suppliers, social and environ- mental entrepreneurial start-ups 18 Serious Change LP www.angel.co/serious-change New York Sustainability, biodiversity, and Off-Grid Electric social justice 19 Synergy Growth www.synergy-energy.co London Energy BBOXX 20 TreeHouse Investments www.treehouseinvestments. Dorado, Puerto Climate change and poverty M-KOPA, com Rico Sunfunder 21 The Social www.tsef.com New York Angaza Designs Entrepreneurs Fund VENTURE CAPITAL/PRIVATE EQUITY 1 Vulcan Capital www.capital.vulcan.com Seattle Financial services, Internet and Off-Grid Electric technology, life sciences and media, communication 2 Fidelity Growth Partners www.eightroads.com Global Software/Saas, mobile, Greenlight Planet consumer, Internet, fintech 3 Prelude Ventures LLC www.preludeventures.com San Francisco Climate innovation Powerhive 4 Tao Capital Partners www.taocap.com San Francisco Alternative transportation and Powerhive energy, among other sectors CORPORATE VC 1 Caterpillar Venture Capital www.caterpillar.com/en/ Menlo Park, CA, Distributed power, analytics, Powerhive company/innovation/cater- United States robotics, additive manufacturing, pillar-ventures.html business models 2 Total Energy Ventures www.total.in/en/total-energy- Paris Energy Powerhive ventures-innovating-start-ups Large Company 1 Barclays Social Innovation www.resources.barclays.com/ Global Healthcare (Africa), educa- Azuri Technologies Facility citizenshipreport/articles/ tion (United States), financial social-innovation-facility.html inclusion 2 E.ON www.eon.com/en.html Essen, Germany Energy Rafiki Power 3 ENEL Green Power www.enelgreenpower.com/ Rome Renewable energy Powerhive ena/en-gb 4 Engie (previous name GDF) www.engie.com/en Courbevoie, Energy BBOXX, Fenix France International, EGG Energy 5 First Solar www.firstsolar.com Tempe, AZ, United Solar photovoltaic (PV) solutions Powerhive States 6 MTN www.mtn.com Johannesburg Mobile Network Operator Fenix International

ISSUE BRIEF | June 2016 | 39 BRIEF PROFILES OF PRIVATE SECTOR INVESTORS WHO HAVE INVESTED EITHER EQUITY OR GRANTS IN PAYG COMPANIES IN EAST AFRICA (CONTINUED) TABLE A8 TABLE

Office PAYG investees Investor Website location Profile (East Africa) CORPORATE VC (CONTINUED) 7 SolarCity Corporation www.solarcity.com San Mateo, CA, Solar PV solutions Off-Grid Electric United States 8 Schneider Electric www.schneider-electric.co.in/ Rueil-Malmaison, Electrical equipment Fenix International, en France SunFunder, Energy Acess Ventures INDIVIDUAL ANGELS 1 Don Weil www.angel.co/donweil San Francisco Angel investor Angaza Designs 2 Frank McCrea www.linkedin.com/in/ Toronto Founder and CEO of Procom Off-Grid Electric frankmccrea 3 Jean and Steve Case www.casefoundation.org Washington, DC Founder and former CEO of AOL M-KOPA 4 Julie Chin www.angel.co/julie-chin San Francisco, Ex Google Angaza Designs Bay Area 5 Jim and Karen Linder www.linseedcapital.com Omaha Faculty at the University of Angaza Designs Nebraska 6 Ryan Allis www.linkedin.com/in/ryanallis San Francisco Cofounder and former CEO of Off-Grid Electric iContact 7 Richard Branson https://en.wikipedia.org/wiki/ British Virgin Founder Virgin Group M-KOPA Richard_Branson Islands 8 Tom Dinwoodie https://en.wikipedia.org/wiki/ San Francisco Cofounder and former CEO of Fenix International Tom_Dinwoodie iContact, PowerLight Corporation 9 Warner Philips www.angel.co/warner-philips San Francisco Cleantech Fenix International

40 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

ANNEX IV: EXCHANGE RATE INFORMATION

MOVEMENT OF THE KENYAN SHILLING AND THE TANZANIAN SHILLING AGAINST THE U.S. DOLLAR, 2004–2016 TABLE A9 TABLE

U.S. Kenya KE % ∆ Tanzania TZS % ∆ BDT % ∆ Date dollar Euro shilling against shilling against Bangladeshi against (US$) (€) (KES) US$ (TZS) US$ taka (BDT) US$

30 January 2016 1 0.923 101.7 -8.8% 2,158 -10.2% 77.42 -1.6%

30 April 2015 1 0.907 92.71 -7.8% 1,937 -17.0% 76.17 0.3%

30 April 2014 1 0.722 85.44 -3.7% 1,607 -1.2% 76.42 0.2%

30 April 2013 1 0.765 82.29 -0.4% 1,588 -1.7% 76.55 5.0%

30 April 2012 1 0.754 81.92 0.3% 1,562 -4.8% 80.35 -10.8%

30 April 2011 1 0.674 82.18 -9.6% 1,488 -8.7% 71.67 -5.2%

30 April 2010 1 0.755 74.27 1.4% 1,359 -3.2% 67.92 -0.8%

30 April 2009 1 0.757 75.34 -20.8% 1,316 -10.6% 67.37 0.2%

30 April 2008 1 0.641 59.69 9.3% 1,176 5.1% 67.51 -0.6%

30 April 2007 1 0.732 65.23 8.5% 1,236 -4.6% 67.08 -1.2%

30 April 2006 1 0.791 70.79 7.8% 1,180 -10.8% 66.28 -4.3%

30 April 2005 1 0.777 76.30 2.1% 1,052 3.7% 63.40 -9.6%

30 April 2004 1 0.835 77.88 – 1,091 – 57.32 –

Data sourced from: http://www.oanda.com/currency/converter/.

ISSUE BRIEF | June 2016 | 41 ANNEX V: DETAILS OF as credit and channeled US$75 Two Types of Grants Provided to the million (BDT 527.1 crore) as grants Partner Organizations THE IDCOL MODEL to POs (IDCOL 2014c). IDCOL has provided a capital buy- The IDCOL program was structured down grant to consumers and an around an innovative lending model The IDCOL model can be institutional development grant to with multiple layers. The World diagrammatically represented as the Partner Organizations (struc- Bank lent to the Government of shown in Figure A1. tured as a specific amount per system Bangladesh at 1–2 percent interest sold). The grants have reduced for 40 years; the Government of Credit Provided to Partner significantly over time (Haque 2014) Bangladesh then lent to IDCOL at Organizations as shown in the Table A11 below. 3–6 percent interest for 15 years. IDCOL provided refinancing of the IDCOL on-lent to POs at 6–9 percent loans provided by Partner Organiza- The capital buy-down grant has interest for 5–10 years. The tenure tions to the end users. As shown in reduced over the period mainly and finally POs lent to households Table A10 below, the loan interest rates because system costs have come at 12 percent interest for three have tended to increase, the repayment down. The US$20 capital buy-down years or less. This spread generated periods have tended to shorten, and the grant available today is for systems income for each entity and created refinanced amount tended to decline in under 30 Wp (these smaller systems an incentive for interest alignment recent years (Rai et al. 2015). are expected to be purchased by from households to the government poorer customers). The phase-out (Bardouille, Aidun, and Muench 2014). This rising interest rate and short- of grants and subsidies over time is ening tenure have been part of the taken as evidence that the program As of 2014, IDCOL had disbursed effort to make the program more represents a self-sustaining, com- US$481 million (BDT 3,619.6 crore) market-oriented. mercial solution to the energy access

DETAILS OF THE IDCOL MODEL

FIGURE A1 Provides approval Technical Standards SUPPLIERS Seeks approval Committee

Supply Pay for equipment equipment

PO Selection Applies Provides approval NGO/PO IDCOL Committee Select POs Provides approval

Pays down- Sells SHS & Grants & soft- payment & provides service term credit installment

Provides solutions HOUSEHOLD DONORS Operations Seeks operation-related solutions Committee

Source: IDCOL 2014d

42 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

challenge (Bardouille, Aidun, and depreciated price, in the event that not sell an expired battery to any Muench 2014). Indeed, there are households become grid-connected second party and the battery shall be suppliers today outside the IDCOL within a specified time period after returned to any of IDCOL’s POs or system who do not provide financing purchase (Sadeque et al. 2014). the supplier of the battery. to customers. Environmental Standards Monitoring and Verification Technical Standards and Vendor IDCOL developed the “Policy Guide- Initially, IDCOL verified every Certification lines on Disposal of Warranty Expired installation, but from 2005 onward The IDCOL Technical Standards Batteries” in June 2005. Based on the it has been verifying on a sample Committee, an independent commit- guidelines, IDCOL facilitated an agree- basis. IDCOL has 12 quality control tee of experts, provides guidelines on ment between battery manufacturers offices, 130 quality inspectors, and the standards for products eligible for and POs. According to the agreement, 11 field auditors. It also manages a inclusion in the SHS program. The POs are responsible for notifying call center for customer complaints Technical Specifications Guidelines customers three months before the (Sadeque et al. 2014). provide detailed specifications on the warranty expiration date and advis- product and associated components, a ing customers to replace the battery. Program Management PO representatives are responsible minimum set of requirements for the IDCOL’s selection committee screens for collecting batteries from custom- installation services and operating envi- potential POs according to program ers and safely transporting them ronment, and a long list of equipment eligibility criteria. The company’s PO to regional locations of the battery suppliers and local agents that meet selection guidelines specify that POs manufacturer. The manufacturer is the set standards (IDCOL 2014b). should be able to demonstrate institu- responsible for collecting batteries tional capacity: audit and accounting from regional centers and transport- Installation Areas management, adequate staffing, a ing them to a site where the batteries IDCOL refinanced SHS installa- certain number of years of operation, will be recycled or disposed of in an tions only if they were refinanced in and experience providing credit in off- environmentally friendly manner. The “off-grid” areas. IDCOL also insisted grid areas (Rai et al. 2015). IDCOL’s buy-back agreement is signed between that POs extend a buy-back guaran- Operations Committee consists of the PO and the household. Accord- tee to give households the option to both PO and IDCOL representatives, ing to this clause, the household shall sell their SHS back to IDCOL for a who manage and oversee implemen- tation of the program.

DETAILS OF LOANS PROVIDED BY POS TO END USERS UNDER THE IDCOL MODEL

TABLE A10 TABLE Period 2003–8 2009 2010 2011 2012–15 Loan tenors (years) 10 6–10 6–8 6–8 5–7 Interest rate (per annum) 6–8% 6–8% 6–8% 6–8% 6–9% Refinanced amount 80% 80% 80% 80% 70–80%

GRANTS PROVIDED BY IDCOL TO THE POS (US$)

TABLE A11 TABLE Period 2003 2004–5 2006–7 2008–9 2010–11 2012 2013–14 Capital buy-down grant 70 55 40 40 25 25 20 Institutional development grant 20 15 10 5 3

ISSUE BRIEF | June 2016 | 43 ANNEX VI: SOLAR HOME SYSTEM PRICES

COMPARISON OF PRICES OF SOLAR HOME SYSTEMS IN BANGLADESH AND EAST AFRICA TABLE A12 TABLE Category: Small system 1 Bangladesh East Africa (Kenya) Product Cash price Lease-to-own terms Total payments Product Cash price Lease-to-own terms Total payments over lease period over lease period BDT BDT BDT (US$) KES KES KES (US$)

10 W, 7,200 Per month: 511 for 12 months 8,007 (US$99) 8W, 4x1W 18,600 Per day: 50 for 365 days 21,750 (US$208) 2x3W Upfront: 1,875 LED + radio Upfront: 3,500 LED

Category: Small system 2 Bangladesh East Africa (Kenya) Product Cash price Lease-to-own terms Total payments Product Cash price Lease-to-own terms Total payments over lease period over lease period BDT BDT BDT/US$ KES KES KES

20 W, 10,250 Per month: 440 for 24 months 12,760 (US$158) 15W, 4x1W 73,550 Per month: 950 for 36 35,150 (US$336) 3x3W Upfront: 2,200 LED + radio months LED Upfront: 950

Category: Mid-level system Bangladesh East Africa (Kenya)

Product Cash price Lease-to-own terms Total payments Product Cash price Lease-to-own terms Total payments over lease period over lease period BDT BDT BDT/USD KES KES KES

50W, 4x3W 19,500 Per month: 940 for 24 months 27,260 (US$338) 50W, 4x1W Per month: 2,150 for 36 79,550 LED, 15” Upfront: 4,700 LED + radio months (US$761) TV + 18” TV) Upfront: 2,150

Category: High-capacity system Bangladesh East Africa (Tanzania)

Product Cash price Lease-to-own terms Total payments Product Cash price Lease-to-own terms Total payments over lease period over lease period BDT BDT BDT/USD TZS TZS TZS

80W, Per month: 1,340 for 24 months 54,940 (US$681) 80W, 3x2W 1,421,700 Per month: 49,100 for 36 1,895,600 7x3W LED, Upfront: 6,700 LED + torch months (US$1,043) 15”TV + + radio + Upfront: 128,000 1 DC fan 19” TV

Notes: Cash price is the price the customer would pay if she or he were to pay it Currency conversion rates (as of Jan 31, 2016): upfront. Not all companies in Kenya and Tanzania offer this option. 1 BDT = 0.01239 US$ | 1 KES = 0.00957 US$ | 1 TZS= 0.00055 US$ Source: OANDA

44 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

ANNEX VII: LINES OF CREDIT

KENYAN BANK PARTNERSHIPS WITH INTERNATIONAL DFIS IN THE AREAS OF MSMES OR GREEN FINANCING TABLE A13 TABLE

Bank DFI partners Equity Bank EIB, FMO, KfW, IFC, China Development Bank, AfDB, Norfund (shareholder through Norfininvest AS)a CFC Stanbic AFD Co-operative Bank of Kenya AFD, DEG, EIB, IFC, FMO Kenya Commercial Bank Group IFC I&M Bank FMO, Proparco (shareholder), DEG (shareholder), IFCb Chase Bank AFD, EIB, FMO, OeEB (Austrian), Proparco IFC, DEG (shareholder) Commercial Bank of Africa AFD Bank of Africa Kenya FMO (shareholder), Proparco (shareholder in Bank of Africa Group), IFC, BIO Belgium (shareholder in Bank of Africa Group) Diamond Trust Bank DEG, Proparco, IFC (shareholder in Diamond Trust Bank Group) Family Bank EIB NIC Bank Kenya IFC, Proparco ABC Bank EIB Consolidated Bank of Kenya EIB K Rep Bank EIB Prime Bank EIB, Proparco

Notes: Only active lines of credit are mentioned. a http://equitybankgroup.com/blog/2015/04/norfund-and-norfinance-complete-purchase-of-12.223-stake-in-equity-group-holdings-limited b I&M Bank Kenya and Tanzania are wholly owned subsidiaries of I&M Holdings Ltd., of which Proparco and DEG are shareholders.

ISSUE BRIEF | June 2016 | 45 TANZANIAN BANK PARTNERSHIPS WITH INTERNATIONAL DFIS

TABLE A14 TABLE Bank DFI partner CRDB EIB, DANIDA (shareholder), IFC, DEG Exim Bank Proparco, FMO, DEG, Norfund National Microfinance Bank EIB, FMO Access Bank IFC (shareholder), KfW (shareholder), AfDB (shareholder) Akiba Commercial Bank FMO (shareholder) Banc ABC DEG, Proparco, IFC, OPIC Bank of Africa Tanzania Proparco (shareholder in Bank of Africa Group), IFC, BIO Belgium (shareholder in Bank of Africa Group), AFD I&M Bank FMO, DEG (shareholder), Proparco (shareholder) NIC Bank Tanzania Proparco TIB Development Bank (former IDA World Bank Tanzania Development Bank)

EAST AFRICAN APEX BANK PARTNERSHIPS WITH INTERNATIONAL DFIS

TABLE A15 TABLE Bank DFI partner PTA Bank EIB, FMO, KfW, JBIC, OPEC, OPIC, AfDB, China Development Bank East African Development Bank Nordic Development Fund (NDF), EIB, DEG (shareholders) OFID, AfDB

46 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

EAST AFRICAN APEX BANK PARTNERSHIPS WITH INTERNATIONAL DFIS

ISSUE BRIEF | June 2016 | 47 REFERENCES AfDB (African Development Bank Group). n.d. DfID (Department for International Development). IDCOL. 2014b. “Technical Specifications for Trade Finance Program. Retrieved on September 2016. “Energy Africa Summary Document.” June. Solar Home Systems (SHS).” Dhaka: IDCOL 6, 2016, from: http://www.afdb.org/en/topics- Retrieved in July 2016 from Nairobi Workshop. Solar Program, Technical Standards Committee. and-sectors/initiatives-partnerships/trade- finance-program/. Doukas, A., and A. Ballesteros. 2015. “Clean IDCOL. 2014c. IDCOL Annual Report 2013– Energy Access in Developing Countries: 14: Toward a Greener Future. PowerPoint Alstone, P., D. Gershenson, D. N. Turman- Perspectives on Policy and Regulation.” Presentation by IDCOL. April 25, 2014, Pattaya, Bryant, D. Kammen, and A. Jacobson. 2015. Washington, DC: World Resources Institute. Thailand. Off-Grid Power and Connectivity Pay-as-You-Go Available online at: http://www.wri.org/sites/ Financing Digital Supply Chains for Pico Solar. default/files/clean-energy-access-developing- IDCOL. 2014d. IDCOL website downloads. Lighting Global Market Research Report. countries-issue-brief.pdf. Retrieved on March 16, 2016, from: http://idcol. org/home/downloads. Angelou, N., and M. Bhatia. 2015. “Beyond GreenMax Capital Advisors. 2013. Tanzania Connections: Energy Access Redefined.” Market Intelligence Final Report. New York. IEA (International Energy Agency). 2011. World Washington, DC: ESMAP (Energy Sector Energy Outlook 2011: Energy for All. Paris: Management Assistance Program). Available Haque, Nazmul. 2014. “Results-Based Financing: International Energy Agency. at: http://www.worldbank.org/content/dam/ Models, Experience and Lessons Learned.” Worldbank/Topics/Energy%20and%20Extract/ IDCOL Solar Home System Program— IEA. 2013. World Energy Outlook 2013. Paris: Beyond_Connections_Energy_Access_ Bangladesh. Dhaka: IDCOL. Retrieved from International Energy Agency. Redefined_Exec_ESMAP_2015.pdf. Africa Carbon Forum: http://africacarbonforum. com/2014/docs/Presentations/Day%201/W2/ IEA. 2014a. World Energy Outlook 2014, A.T. Kearney, and GOGLA (Global Off-Grid W2_Nazmul%20Haque_Bangladesh.pdf. Electricity Database. Retrieved from: http:// Lighting Association). 2014. Investment and www.worldenergyoutlook.org/resources/ Finance Study for Off-Grid Lighting. Retrieved Hirtenstein, Ann. 2012. “African Sunshine energydevelopment/energyaccessdatabase/. from GOGLA website: http://global-off- Can Now Be Bought and Sold on the Bond grid-lighting-association.org/wp-content/ Market.” Bloomberg News. Retrieved on January IEA. 2014b. Africa Energy Outlook. A Focus uploads/2013/09/A-T-Kearney-GOGLA.pdf. 27, 2016, from: http://www.bloomberg.com/ on Energy Prospects in Sub-Saharan Africa. news/articles/2016-01-12/african-sunshine- A World Energy Outlook Special Report. Bardouille, P., C. Aidun, and D. Muench. 2014. can-now-be-bought-and-sold-on-the- Paris: International Energy Agency. Retrieved “Annex I: The Nexus between the Successful bondmarket?utm_content=buffer5d8bc&utm_ from: https://www.iea. org/publications/ SHS Program in Bangladesh and the Emerging medium=social&utm_source=facebook. freepublications/publication/WEO2014_ DESCO Sector in Sub-Saharan Africa and India.” com&utm_campaign=buffer. AfricaEnergyOutlook.pdf. New York: Persistent Energy Partners. Hogarth, R., and I. Granoff. 2015. “Speaking IFC (International Finance Corporation). 2012. Ballesteros, A., E. Norford, S. E. Alzner, T. Truth to Power: The Relationship between Energy From Gap to Opportunity: Business Models Nagle, and L. Yonavjak. 2013. “Implementation and Poverty and What It Means for Scaling for Scaling Up Energy Access. Retrieved Strategies for Renewable Energy Services in Ambition.” London: Overseas Development from: http://www.ifc.org/wps/wcm/connect/ Low-Income, Rural Areas.” Washington, DC: Institute and Oxfam America. Retrieved from OA topics_ext_content/ifc_external_corporate_site/ World Resources Institute. Available online website: http://policy-practice.oxfamamerica.org/ ifc+sustainability/learning+and+adapting/ at: http://www.wri.org/sites/default/files/pdf/ static/media/files/FINAL_speakingpowertotruth_ knowledge+products/publications/publications_ implementation_strategies_renewable_energy_ SH.pdf. report_gap-opportunity. services_low_income_rural_areas.pdf. IDCOL (Infrastructure Development Company Kaufman, R., R. Duke, R. Hansen, J. Rogers, Bill and Melinda Gates Foundation. 2016. Limited). 2014a. IDCOL website. Retrieved on R. Schartz, and M. Trexler. 2000. Rural “How We Work: M-KOPA Kenya Ltd.” Retrieved June 11, 2015, from: http://idcol.org/. Electrification with Solar Energy as a Climate on September 30, 2016, from: http://www. Protection Strategy. Renewable Energy Policy gatesfoundation.org/How-We-Work/Quick-Links/ Project. Program-Related-Investments/M-KOPA-Kenya- Ltd.

48 | Stimulating Pay-as-You-Go Energy Access in Kenya and Tanzania: The Role of Development Finance

Khan, S., and A. Azad. 2014. “Social Impact Nique, M. 2013. “Sizing the Opportunity SIDA. 2012. “Sida’s Guarantee Instrument.” of Solar Home Systems in Rural Bangladesh: of Mobile to Support Energy and Water July 2012, updated June 2016. Retrieved on A Case Study of Rural Zone.” IAFOR Journal Access.” GSMA and U.K. Aid. http://www. September 15, 2016, from: http://www.sida.se/ of Sustainability, Energy and the Environment: gsma.com/mobilefordevelopment/wp-content/ English/partners/our-partners/Private-sector/ 1–17. Aichi, Japan: International Academic uploads/2013/12/Sizing-the-Opportunity-of- Innovative-Finance-/. Forum. Mobile_Nov-2013.pdf Siegel, J. R., and A. Rahman. 2011. “The Kleijn, M., J. Sebastian, and M. Veen. OPIC (Overseas Private Investment Corporation). Diffusion of Off-Grid Solar Photovoltaic 2016. “Pico-Solar for All.” The Hague: SNV 2015. “OPIC, Calvert Foundation Finalize Technology in Rural Bangladesh.” Energy, (Netherlands Development Organization). http:// Support to PAMIGA Finance, Spreading Climate, and Innovation Discussion Paper. www.snv.org/public/cms/sites/default/files/ Renewable Resources Access in Rural Africa.” Medford, MA: Tufts University, Fletcher School explore/download/case_study_-_pico-solar_ Washington, DC: OPIC. of Law and Diplomacy. for_all.pdf. Rai, Neha, et al. 2015. “Financing Inclusive Tracy, J., and A. Jacobson. 2012. “The True Cost The Lab (Global Innovation Lab for Climate Low-Carbon Resilient Development: Role of of Kerosene in Rural Africa.” Washington, DC: Finance). 2016. “Germany to Provide €30 Million Central Bank of Bangladesh and Infrastructure Lighting Africa, an Initiative of the World Bank. to TCX for Currency Hedging Instruments.” Development Company Limited.” London USAID (U.S. Agency for International Retrieved on September 6, 2016, from: http:// International Institute for Environment and Development). n.d. Development Credit climatefinancelab.org/press-release/germany- Development. Authority. Retrieved on September 6, 2016, from: to-provide-e30mn-to-the-long-term-fx-risk- https://www.usaid.gov/what-we-do/economic- management-instrument-for-sub-saharan-africa- Rao, Narasimha, Anjana Agarwal, and Davida growth-and-trade/development-credit-authority- climate-investments/. Wood. 2016. “Impacts of Small-Scale Electricity putting-local-wealth-work. Systems: A Study of Rural Communities in India Lighting Africa. 2015. “Lighting Africa: About Us, and Nepal.” Washington, D.C.: World Resources USAID (United States Agency for International What We Do.” Washington, DC: Lighting Africa, Institute. Development). 2015. “Development Credit an Initiative of the World Bank. Retrieved on July Authority: Impact Brief 2015.” https://www.usaid. 14, 2015, from: https://www.lightingafrica.org. Rolffs, P., R. Byrne, and D. Ockwell. 2014. gov/sites/default/files/documents/1865/dca_ “Financing Sustainable Energy for All: Pay-as- impactbrief15_v15_link_spreads_160824b.pdf MFX Solutions. 2013. “Our Products.” Retrieved You-Go vs. Traditional Solar Finance Approaches on August 1, 2016, from: http://mfxsolutions. in Kenya.” STEPS Working Paper 59. Brighton, Wang, L., S. Bandyopadhyay, M. Cosgrove- com/our-products/. UK: University of Sussex, STEPS Centre (Social, Davies, and H. Samad. 2011. “Quantifying Technological, and Environmental Pathways to Carbon and Distributional Benefits of Solar Mills, E. 2005. “The Specter of Fuel-Based Sustainability). Home System Programs in Bangladesh.” World Lighting.” Science 308(5726): 1263–64. Bank Policy Research Working Paper Series. Sadeque, Z., D. Rysankova, R. Elahi, and R. Soni. Mills, E. 2014. “Light for Life: Identifying and 2014. “Scaling Up Access to Electricity: The Reducing the Health and Safety Impacts of Case of Bangladesh.” Washington, DC: World Fuel-Based Lighting.” Nairobi: United Nations Bank. Live Wire. 88702. Environment Programme. Samad, H. A., S. R. Khandker, M. Asaduzzaman, Muench, D. and S. Issler. 2015. “Currency and M. Yunus. 2013. “The Benefits of Risk and Mitigation Strategies for the Off-Grid Solar Home Systems: An Analysis from Energy Sector.” Persistent Energy Capital and Bangladesh.” World Bank Policy Research responsAbility. http://media.wix.com/ugd/4c83e Working Paper (6724). Washington, DC: World 3_242d019d9d904df3b6e7caf69447acf0.pdf Bank.

ISSUE BRIEF | June 2016 | 49 ACKNOWLEDGMENTS ABOUT THE AUTHORS The authors would like to thank the following Sanjoy Sanyal is a Senior Associate at the individuals for their valuable insights and World Resources Institute. During the research critical reviews of this work: Dana Rysankova, and writing of this brief, he was the country Leanne Jones, Marina Pannekeet, Robert director of New Ventures India. Voskuilen, Marc Buiting, John Coleman Contact: [email protected] MacLean, and Christine Eibs Singer. The authors would also like to thank the following colleagues at WRI for their contributions: Jeffrey Prins previously worked as a fellow Athena Ronquillo-Ballesteros, Davida researching finance and energy access at World Wood, and Laura Valeri Malaguzzi for their Resources Institute. guidance; Allison Lee for research and editing contributions; Giulia Christianson, Lily Odarno, Contact: [email protected] Bharath Jairaj, and Michael Westphal for their peer reviews; Emily Matthews for copyediting Feli Visco is a research consultant at New and proofreading; and Hyacinth Billings and Ventures India. Julie Moretti for graphic design, editing, and production support. Contact: [email protected]

Funding for this project was provided by the DOEN Foundation. While our reviewers were Ariel Pinchot is a research analyst in the very generous with their time and advice, this Sustainable Finance Center at World Resources working paper represents the views of the Institute. authors alone. Contact: [email protected]

50 | ABOUT WRI WRI is a global research organization that works closely with leaders to turn big ideas into action to sustain a healthy environment—the foundation of economic opportunity and human well-being.

Our Challenge Natural resources are at the foundation of economic opportunity and human well-being. But today, we are depleting Earth’s resources at rates that are not sustainable, endangering economies and people’s lives. People depend on clean water, fertile land, healthy forests, and a stable climate. Livable cities and clean energy are essential for a sustainable planet. We must address these urgent, global challenges this decade.

Our Vision We envision an equitable and prosperous planet driven by the wise management of natural resources. We aspire to create a world where the actions of government, business, and communities combine to eliminate poverty and sustain the natural environment for all people.

Our Approach COUNT IT We start with data. We conduct independent research and draw on the latest technology to develop new insights and recommendations. Our rigorous analysis identifies risks, unveils opportunities, and informs smart strategies. We focus our efforts on influential and emerging economies where the future of sustainability will be determined.

CHANGE IT We use our research to influence government policies, business strategies, and civil society action. We test projects with communities, companies, and government agencies to build a strong evidence base. Then, we work with partners to deliver change on the ground that alleviates poverty and strengthens society. We hold ourselves accountable to ensure our outcomes will be bold and enduring.

SCALE IT We don’t think small. Once tested, we work with partners to adopt and expand our efforts regionally and globally. We engage with decision-makers to carry out our ideas and elevate our impact. We measure success through government and business actions that improve people’s lives and sustain a healthy environment.

Each World Resources Institute issue brief represents a timely, scholarly treatment of a subject of public concern. WRI takes responsibility for choosing the study topics and guaranteeing its authors and researchers freedom of inquiry. It also solicits and responds to the guidance of advisory panels and expert reviewers. Unless otherwise stated, however, all the interpretation and findings set forth in WRI publications are those of the authors.

Copyright 2016 World Resources Institute. This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of the license, visit http://creativecommons.org/licenses/by/4.0/. 10 G STREET NE SUITE 800 WASHINGTON, DC 20002, USA +1 (202) 729-7600 WWW.WRI.ORG ISBN 978-1-56973-911-2