WATER AND SANITATION PROGRAM: LEARNING NOTE 92034 Making toilets more affordable Key findings for the poor through • There is demand for latrines even among poor households; a sanitation microfinance loan program offered by socially-oriented microfinance institutions (MFIs) helps to Lessons learned from introducing microfinance loans increase uptake of sanitation among the for sanitation in rural Cambodia poor. A small loan size and a poor- inclusive application process are August 2014 essential to success. • MFIs can increase the number of loans INTRODUCTION well as providers of sanitation financing offered, reduce loan processing time, From 2000 to 2012, access to sani- products and services. and increase a household’s likelihood tation in Cambodia’s rural areas in- of committing to a sanitation loan creased by only 1% per year (JMP, In Cambodia, extensive previous ex- by dedicating loan officers to the 2014). By 2012, 75% of rural Cam- perience with sanitation marketing sanitation portfolio. bodians lacked access to improved approaches illustrates there is strong sanitation, and 66% practiced open household demand for sanitation and • Allowing borrowers to repay loans defecation. Though open defecation the domestic sanitation market is ca- close to where they live increases rates are highest among the poorest pable of meeting it. At the same time, the likelihood of interest in this loan rural Cambodians at 86%, they are still challenges remain in reaching low-in- product. Borrowers will hesitate if they quite high even among the richest at come households that do not have the have to travel long distances, especially 32% (CSES, 2011). Lack of access cash to meet upfront payments to pur- for small loans. to sanitation imposes significant eco- chase sanitation products. nomic and social costs on rural Cam- • A close partnership between an MFI and bodians, from higher child mortality Over a 13-month period, WSP worked a latrine business that has the motivation due to diarrhea, other fecal-borne dis- with a number of partners, including the and capability to produce and eases, to stunted growth of children. international non-profit Program for Ap- deliver on time is needed to maximize propriate Technology in Health (PATH) commitments from customers and avoid The World Bank’s Water and Sanita- and International Development Enter- losing latrine orders. tion Program (WSP) is supporting the prises (iDE), to pilot a sanitation financ- Government of Cambodia in its efforts ing program to address the challenge of • A poor inclusive sanitation loan program to increase access to sanitation among reaching low-income households with is financially viable and sustainable rural households. Achieving this goal improved sanitation solutions. This learn- given the right support, and if loans are requires effective demand generation ing note presents the lessons from this managed carefully. for sanitation, highly-engaged local pilot to promote scale-up in Cambodia governments that work closely with and to inform similar efforts in other coun- the private sector to encourage service tries. This pilot is also part of a broader delivery, and a well-functioning value sanitation marketing initiative co-funded chain that leverages the capabilities by the Bill and Melinda Gates Foundation W t r of domestic sanitation businesses as and Stone Family Foundation. 2 Making Toilets More Affordable for the Poor through Microfinance Domestic Private Sector Participation PROBLEM STATEMENT and partners (PATH and iDE) sought to partner with MFIs with While there is strong demand for improved latrines in Cambo- established scale and penetration in rural areas, a commitment dia amongst the rural poor, this demand goes unmet largely to serving poor Cambodians, and an interest in participating in because these consumers cannot afford to pay upfront for a a pilot. Through this process, VisionFund Cambodia in Kandal latrine that meets their preferences. Research showed that province (Jul 2012 – Mar 2013) and KREDIT in Prey Veng 77% of Cambodians were considering constructing a latrine, province (Nov 2012 – Jul 2013) were engaged in the pilot. yet ownership remains low, especially in rural areas.1 Ninety- five percent of households who do not own a latrine reported KREDIT was chosen in part because of its strong existing so- they were too expensive or they did not have enough money cial loan2 program targeted at the poor. As of 2011, KREDIT to purchase a latrine upfront. Having access to microloans served over 56,500 clients. It had an operating self-sufficiency could help alleviate this challenge; however, microfinance in- ratio of 123%, meaning that the organization’s operating ex- stitutions (MFIs) often perceive non-productive loans as high- penses were covered by their operational revenue. In addi- risk, particularly when the borrower is not a trusted existing tion, less than 0.33% of its loan payments were more than business client. Yet, if a household with an income at the na- 30 days late. tional poverty level (US$900 per household per year) could save just 5% of its income, it would have enough money in VisionFund Cambodia was chosen partly because of its one year to buy a basic substructure of a latrine that is avail- previous partnership with PATH, and also because it has a able in the market at the price of around US$40-45. In other low average loan size and one of the largest outreach into words, if paying for the cost of a latrine can be spread out rural areas among MFIs in Cambodia. As of 2011, it served over a period of time, getting a latrine is becoming more af- more than 132,000 clients, had an operating self-sufficiency fordable for more households, possibly including the cash- ratio of 119% and less than 0.14% of its loan payments were strapped poor for whom large upfront payments are prob- more than 30 days late. lematic. It is important to note that the latrine product at the price given above only refers to the substructure part of the Table 1: Key MFI Metrics latrine (installation not included). Households then choose to Metric KREDIT VisionFund build a shelter at their own additional cost, pace and accord- Number of Provinces 11 19 ing to their preference, which could be made of local organic Active borrowers 56,519 132,036 materials, bricks or other materials. Total Loans Outstanding (US$) 44 million 38 million ACTION Average Loan Size (US$) 789 286 As part of WSP’s support to the Government of Cambodia Source: MFI 2011 annual reports to increase access to sanitation, especially among the The partnership was reflected in a scope of work outlining poor, consultations with MFIs, NGOs, and other potential the roles and responsibilities of each stakeholder: the MFI stakeholders were conducted to evaluate several sanitation and its credit officers, PATH and iDE, latrine businesses and financing options including savings groups, revolving funds, latrine sales agents. Because the primary goal of the pilot and latrine suppliers extending installment payments to was to learn what approaches would increase latrine up- their customers. While any of these approaches could be take among the poor, the scope of work defined how proj- viable under the appropriate market conditions, the research ect partners would provide assistance to the MFIs so they concluded that savings groups relied too heavily on donor could test different approaches and document the results. support and installment payments offered by latrine businesses For example, one MFI received a grant to help offset the were too complicated for businesses to manage. A household cost of collecting data for the pilot. Another received a loan loan product offered through an MFI was determined the guarantee to enable reduction in collateral requirements to most scalable and sustainable approach in Cambodia. WSP ensure sufficient loan demand. 1 WSP, “Sanitation Marketing Lessons from Cambodia: A Market-Based Approach to Delivering Sanitation” Oct 2012. 2 For the purposes of this paper, the term “social loan” means a loan designed to improve the living standards of the borrower. www.wsp.org Domestic Private Sector Participation Making Toilets More Affordable for the Poor through Microfinance 3 KREDIT offered both group and individual loans to a large balance at the end of the loan term that must be repaid customers in a lump sum. Balloon repayments are mostly popular with KREDIT enabled villagers to decide whether they wanted to farmers or other households with seasonal income, as they can join a group or obtain an individual loan. Some villagers chose time the lump-sum payment with their income. an individual loan because they did not want to find others to form a group or did not want to share default risk. In a fourth district, KREDIT also offered loans to individuals between US$40 and US$250. Initially, individual loans were Under the community bank model, KREDIT offered loans to offered at an interest rate of 2.65% per month and individuals groups of 4-6 households and required at least two groups were required to travel to make payments at the MFI branch in order to establish a community bank. In three Prey Veng located in the district centers. The MFI also planned to re- districts, group loans could be repaid locally. Group loan siz- quire movable collateral (e.g., motorbikes, hand tractors, wa- es ranged between US$40 and US$250 at an interest rate of ter pumps) but found this requirement difficult to implement. 2.9% per month.3 The group of households shared default risk Thus, KREDIT ultimately decided to offer loans with no collat- as a collateral substitute and the community (group of groups) eral to test if this would stimulate loan demand. To offset the shared the risk of losing access to future loans in the event of increased risk, the MFI increased the interest rate on individual default. Many of the borrowers under this program were exist- loans to 3%, and ultimately changed to allow repayment at ing customers and several new customers took out follow-on the village level. KREDIT required a declining balance repay- loans, so risk of losing access to future loans was a serious ment method for all individual loans. Under a declining bal- consequence of default. KREDIT required a balloon repayment ance method, principal repayments are spread out over the method for all group loans. Under a balloon repayment, the duration of the loan and interest is only charged on the actual entire principal is not amortized over the life of the loan, leaving principal, rather than the initial amount borrowed. Table 2: Summary of MFI sanitation loan offerings characteristics Characteristic KREDIT individual loans KREDIT group loans VisionFund Cambodia group loans From 2 to numerous households, all Basic Structure 1 household, 1 district 3-6 households, 3 districts districts Each group shares default risk and Collateral No collateral requirement group of groups shares risk of access Each group shares default risk to future loan opportunities Initially MFI branch office; changed to Repayment village based on demand through a Village Village Location village “teller” collecting payments US$40-US$250 (Loans disbursed in US$40-US$250 (Loans disbursed in US$40-US$350 (loans disbursed in Size KHR) KHR) KHR) Duration 6-12 months (borrower choice) 6-12 months (borrower choice) 4-12 months (borrower choice) Repayment Declining balance or balloon method Declining balance method Balloon method Method (customer choice) Interest Rate 2.65-3% per month 2.9% per month 2.6-2.8% per month 3 All loans were disbursed in KHR during the pilot, but have been converted to US$ for this Learning Note at a rate of 4,000 KHR to US$1. www.wsp.org 4 Making Toilets More Affordable for the Poor through Microfinance Domestic Private Sector Participation VisionFund Cambodia offered group loans through a loan applications were approved, the latrine businesses, in community bank model the ideal scenario where the business was ready to deliver, In VisionFund Cambodia’s community bank repayment method, distributed the latrines to households within a few days. Fol- the MFI offered group loans with a 2.6-2.8% interest rate per lowing confirmation that the latrines had been received, the month. VisionFund Cambodia charged 2.6% for loans funded MFIs disbursed loan payments to the latrine businesses once through Kiva4 and 2.8% for loans funded by other sources. per week. In order to make this process work smoothly, PATH Groups of households shared the risk of default. Customers field staff provided considerable coordination support. could choose loan terms of 4-12 months as well as whether to use a declining balance or balloon payment method. Ninety per- It should be noted that some households might not imme- cent of customers chose to use the declining balance repayment diately install the complete latrine after the purchase of the method, underscoring the popularity of that payment method. basic substructure. Monitoring data suggest that within 6-12 months around three quarter of all households have installed Figure 1: Sanitation Loan Implementation Process their toilet. In the pilot, there was no specific data collected on LatrineLatrine businessproduce businesses produce latrines andand latrines build up stock build up stock installation rates and shelter types for households that took up a sanitation loan. Sales agents of the latrine businesses sell latrines to households through a group sales method LESSONS LEARNED MFI field loan officers attend group sales meeting to offer credit to households for latrine purchases Lesson 1: There is demand for latrines even among poor households. A sanitation loan program helps to Sales orders and immediate post-meeting loan applications are completed at the end of group sales increase uptake among the poor. Because many households, especially those with lower income MFIs perform normal loan review and approval process levels, cite an inability to pay the up-front costs of a latrine as a major barrier to accessing improved sanitation, purchasing a Once approved, latrine businesses deliver latrines to households latrine on credit may help increase sanitation uptake rates. MFIs disburse loans to latrine businesses upon confirmation of For VisionFund Cambodia, the sanitation loans reached latrine delivery to households three times more poor households than their normal loans.5 For KREDIT, the proportion of the sanitation loan taken up MFIs review monthly loan performance data and follow-up on late payments by poor households has been proportionate to poor popula- tion in the province.6 In other words, the loan pilot was poor- Latrines sold and distributed directly to groups and inclusive and better able to serve the poorer segments of the individuals communities than their traditional loans. As most of the loans Latrines (excluding the shelter) were sourced from indepen- issued were group loans not requiring collateral, this mecha- dent latrine businesses, which hired commission-based sales nism proved to be generally poor-inclusive. agents responsible for selling latrines to groups of households within a given area. MFI field loan officers also attended these The findings of the pilot are consistent with research show- meetings to offer households the option to purchase latrines ing that latrine uptake rates increased fourfold among poor on credit. Sales orders and loan applications were completed households comparing cash on delivery and six-month at the end of these sales meetings (see Figure 1). Once the spread payments.7 4 Kiva is a non-profit organization that through leveraging the internet provides no-interest funds to its worldwide network of microfinance institutions. For more information see www.kiva.org. 5 This is based on PPI USAID poverty tool which is used by VisionFund Cambodia to assess its borrowers. From all VF sanitation loans, 53% were disbursed to households living on PPI USAID Poverty Line, and 21% to those living under the PPI USAID Extreme Poverty Line. See also http://www.progressoutofpoverty.org. 6 For KREDIT, 32% of all sanitation loans were disbursed to poor households, as compared to an overall average of 27% poor households in the province. Poor households are defined as ID-poor category I and category II as per the official poverty identification system of the Cambodian government. See also http://www.mop.gov.kh. 7 IDinsight, “Microfinance Loans to Increase Sanitary Latrine Sales,” Policy Brief, June 2013. www.wsp.org Domestic Private Sector Participation Making Toilets More Affordable for the Poor through Microfinance 5 Figure 2: Cash vs. Credit Latrine Orders in Prey Veng and Kandal Lesson 3: Reducing loan processing times can Provinces increase sanitation uptake and may require removing Cash vs. Credit Latrine Orders 1475 regulatory barriers for loan approvals. Cash Giving loan officers the authority to approve loans imme- of Orders 1015 996 982 Credit Number 853 diately is another way to reduce loan processing times. 637 645 490 351 However, the Credit Bureau of Cambodia (CBC) requires a credit check for any loan. This step adds a delay to loan Mon 1 Mon 2 Mon 3 Mon 4 Mon 5 Mon 6 Mon 7 Mon 8 Mon 9 processing times, as credit checks can only be completed in an MFI’s district branch offices. It also increases an MFI’s During the first half of the pilot, latrine cash orders exceeded operational costs as bank staff must travel between villages latrine credit orders. However, as households became more and offices to process credit check information. Further- comfortable with the idea of paying for a latrine with a loan, more, the CBC charges MFIs a US$0.18 fee for each credit credit orders increased significantly (Figure 2). check for a loan under US$500. For VisionFund Cambodia, the CBC credit check requirement added another layer of Lesson 2: Dedicated loan officers can streamline and complexity as its loan approval procedure already required expedite the loan process. district branch managers to travel to the field to verify each Households faced long delays while waiting for loan approv- household’s loan application.8 als, especially during the initial months of the pilot. Interviews with latrine producers and sales agents found that dedicat- Obtaining a waiver for credit checks on small loans under ed loan officers reduced the time households spent waiting, a reasonable threshold (e.g., US$50, US$100) can facilitate sometimes by as much as three weeks, thereby increasing faster loan processing, reduce the delay between the house- loan volume and interest revenue for MFIs. Sales agents and hold’s decision to purchase a latrine and its delivery, and ulti- latrine businesses expressed that they rely on strong relation- mately increase sanitation uptake rates. ships with the loan officer for their own sales and revenue. As such, dedicated loan officers can make a sanitation loan pro- Development partners and governments can support MFIs gram more attractive for latrine businesses as well, improving in developing strategies to overcome regulatory costs or bur- the effectiveness of the program so long as latrine businesses dens like the CBC credit check requirement. are capable and ready to deliver latrines. Lesson 4: Households may be willing to pay a slightly Interviews with sales agents indicate that customers are most higher interest rate in exchange for a closer and more motivated to buy a latrine after a group sales meeting, espe- convenient payment location. cially when they learn they can purchase the latrine on credit. In the initial KREDIT individual loan design, borrowers were Because rural poor households have many competing de- required to make repayments at district branches. However, mands on their time and resources, this motivation can dis- very few people were willing or able to travel to the district sipate after a delay, reportedly often leading to a cancellation capital (sometimes 25 km away) to do so. Thus, in April 2013, of a purchase. Thus, dedicated loan officers who process the KREDIT individual loan model was modified to a com- loans immediately after a sales meeting can have a significant munity bank model, where field “tellers” collected loan repay- impact on loan volume and interest revenue by minimizing ments in each village. As a result, individual loan demand in- delays. However, it is difficult for MFIs to support dedicated creased, even though interest rates increased as well from credit officers for a small loan program like this without ad- 2.65% to 3% per month. In fact, demand increased to such ditional support from development partners. Therefore, the an extent that credit officers could no longer process loan ap- additional benefit of a dedicated credit officer needs to be plications in a timely manner. weighed against the costs of human resources for MFIs. 8 KREDIT’s operational procedure did not require an approval visit by the branch manager. KREDIT loans would be approved by the loan officer in the field, with the caveat that customers pass the CBC credit check that would be done afterwards in the district branch. Hence, for KREDIT the CBC credit check was less of an obstacle and not subject to a waiver in the pilot. www.wsp.org 6 Making Toilets More Affordable for the Poor through Microfinance Domestic Private Sector Participation Lesson 5: A close partnership between an MFI and a Neither MFI experienced loan defaults nor delinquent pay- latrine business that has the motivation and capability ments over 30 days. Default rates under this pilot were lower to produce and deliver on time is needed to maximize than those for KREDIT and VisionFund Cambodia’s other commitments from customers and avoid losing orders. loan portfolios. This could be because of the relatively low Selecting business partners with the desire and capability to risk profile of small sanitation loans, and maybe also due deliver latrine products effectively is an important part of setting to the methods used by MFIs to manage the loans. Both up a successful sanitation loan program. Ineffective business MFIs followed up promptly with households who were late practices can be a risk to a loan program. For example, during on their payments. This may have reduced the rate of port- the pilot, some sales agents offered latrines on credit to house- folio at risk. holds that had not been approved as creditworthy by credit officers, leading to processing delays, lost or canceled orders, By the end of the pilot, both MFIs achieved loan self-sufficien- and general household frustration with the process. Additionally, cy ratios greater than 100%, indicating that costs of offering because delivery is largely a fixed cost, a latrine business may sanitation loans can be covered by the loan interest revenue. wait to deliver latrines until volume has increased, maximizing Loan performance data indicate MFIs go through a learning the use of a single delivery trip. However, late delivery of latrines curve in which loan self-sufficiency rates improve over time by a latrine business may cause households that have decided (Figure 3). Pilot data seem to suggest that MFIs with previ- to purchase a latrine with cash to abandon the purchase deci- ous experience implementing and scaling up social loans may sion. This is especially true during the initial months of a sanita- reach loan-self-sufficiency faster than those without. tion loan program, when latrine order and delivery volume is low. Similarly, average acquisition cost per loan, or the direct In the pilot, the selection of the best latrine businesses was costs of sales meetings and the loan application and ap- guided by i) the frequency of interaction between business- proval process, decreased and stabilized after peaking in es and sales agents, ii) the level of investment in production the third month of the pilot. This indicates MFIs learned equipment, and iii) the training/coaching businesses would how to acquire loans more efficiently during the pilot (see have received. Figure 4). In the future, MFIs may consider developing detailed scopes of Generally, larger loan sizes were associated with greater rev- work with their business partners, drawing on learnings from enue. MFI 1 provided a larger average loan size than MFI 2 the pilot to establish clear expectations of roles and responsibili- (US$70 vs. US$55, respectively). The average revenue per loan ties. For example, MFIs may be able to align their incentives with for MFI 1 was US$5.53 compared with US$4.89 in MFI 2. those of latrine businesses by providing working capital loans to business owners who seek them.9 MFIs may also benefit from Plateaus in loan self-sufficiency ratios, like that shown during technical assistance from an external support organization that months 4 and 5 in Figure 3 are explained by increases in sales has experience in a sanitation loan program to understand the meetings (at greater cost) before greater latrine sales and rev- capabilities and constraints of potential latrine business part- enue was realized. Both MFIs disbursed a greater number of ners. These might include ability to access capital necessary to loans each month until the final month of the pilot, which saw produce latrines and the ability to deliver latrines on time. a taping off (see Figure 5). Lesson 6: A poor-inclusive sanitation loan program Table 3: MFI Total Revenue and Average Loan Size has a relatively low risk profile and can be financially Average loan size Total interest Model Total # of loans viable and sustainable given the right support. It can (principal) revenue help socially-oriented MFIs widen their customer MFI 1 1,053 US$70.6010 US$5,822 bases and achieve their missions. MFI 2 941 US$55.03 US$4,606 9 Loans to latrine businesses were not processed under this pilot as they either did not need them or were not motivated enough to seek loans. 10 For reasons of confidentiality, the names of the MFIs are not disclosed when discussing business performance. www.wsp.org Domestic Private Sector Participation Making Toilets More Affordable for the Poor through Microfinance 7 Figure 3: Average Weighted Loan Self Sufficiency Ratio for two WHAT ELSE WE NEED TO KNOW MFIs In addition to the lessons learned from this pilot, MFIs in part- Average Weight Loan-Self-Sufficiency Ratio (LSSR) nership with other actors may wish to test further innovations 115% in offering a sanitation loan program to better understand the 102% 88% 90% impact on sanitation uptake on rural households. 57% 61% LSSR 42% 47% 19% Are MFIs willing and interested in sustaining and scal- Mon 1 Mon 2 Mon 3 Mon 4 Mon 5 Mon 6 Mon 7 Mon 8 Mon 9 ing-up a sanitation loan program without continuous external support? Figure 4: Average Acquisition Cost per Loan for two MFIs Behind the success of the pilot is the continuous support Average Acquisition Cost Per Loan provided to the MFIs offering the sanitation loan product $6.7 to rural consumers. This support, either in terms of risk $5.1 sharing or a small grant, has been essential for the MFI Cost (USD) $4.7 $3.6 $3.6 $3.4 $3.9 to cope with emerging risks and to help them go through $2.9 $3.1 the learning curve while introducing the latrine loan in their lending portfolio. More learning is needed on the type and Mon 1 Mon 2 Mon 3 Mon 4 Mon 5 Mon 6 Mon 7 Mon 8 Mon 9 intensity of support that MFIs consequently need to sus- Figure 5: Total New Loans Issued by two MFIs tain and scale up to ultimately achieve a situation where the product is fully integrated in their operations without New Loans Issued external support. 312 328 339 277 295 of Loans Number Should sanitation loan products also cover the cost of 146 105 80 112 latrine shelters? Many Cambodian households do not view a latrine as com- Mon 1 Mon 2 Mon 3 Mon 4 Mon 5 Mon 6 Mon 7 Mon 8 Mon 9 plete without a concrete water basin and a surrounding shel- ter. However, a latrine with this form of infrastructure costs between US$200 and US$300, over four times the size of the Both MFIs allowed existing and new clients to take out loans average sanitation loan provided under the pilot. under the sanitation loan pilot, with each MFI waiving prohi- bitions against clients maintaining more than one loan at a A larger loan that could cover the cost of concrete water basins time as long as one of the loans was for sanitation. The MFIs and latrine shelters would likely increase improved sanitation reasoned that sanitation loans were small, so they did not uptake while contributing to higher MFI sales and interest rev- pose the same over-indebtedness risk to borrowers. Existing enue. This hypothesis could be tested while introducing lower- clients made up only around 25% of each MFI’s total loan cost complete toilet options and the option of loan repayment, portfolio. particularly amongst the poorest households. Allowing new clients increases risk somewhat, but enables To what extent were the poorest households excluded MFIs to broaden their customer base. VisionFund Cambodia from and affected by the sanitation loan program? and KREDIT were both able to convert about 15% of new cli- Though the pilot sanitation loan programs increased sanitation ents to larger, income-producing loans by the end of the pilot. access among the poor, qualitative research suggests some of Though 15% retention rates are not considered very high by the poorest households were excluded from group loans be- MFI standards, both MFIs were able to realize new revenue as cause these households were perceived as unable to repay a result of retaining clients. During the pilot period, the MFIs the loan. More information is needed to determine the extent participating in the sanitation pilot were able to disburse ap- to which poor households were excluded from groups and the proximately US$55,000 in 195 follow-on loans to new clients. extent to which taking up a sanitation loan affects the trade- www.wsp.org 8 Making Toilets More Affordable for the Poor through Microfinance Domestic Private Sector Participation off in household spending patterns, as CONCLUSION Acknowledgments the poor may not spend on other basic Though many rural Cambodian house- The Learning Note was prepared by Adam needs while paying off the loan. Such un- holds lack improved sanitation, there is Newman (Deloitte Consulting), Susanna Smets derstanding will help identify other mea- considerable demand for latrines, es- and Phyrum Kov (WSP). The Note benefits from sures that can help them gain access to pecially when offered on credit. Small the comments of Almud Weitz, Jemima Sy, John sanitation more effectively (e.g., savings, loans issued by MFIs to either individu- Ikeda, Emily Rand, Eddy Perez (WSP), Yi Wei and subsidies, loan guarantees). A complete als or groups of households can be a Khemra Ros (iDE), Vichet Ouch (KREDIT) and toilet with shelter is a direction that may viable way to increase uptake of im- Sok Kea Cheang (VisionFund Cambodia). Special increase uptake of sanitation by house- proved sanitation in rural areas, includ- thanks also go to PATH and iDE for capturing the holds, and increase viability of sanitation ing among the poor. If such a program lessons learned from sanitation financing program loan programs. However, it will remain is planned, tested, and scaled up care- implementation; and to VisionFund Cambodia and important to communicate behavior fully, MFIs with good lender practices KREDIT for participating in the pilot program and change messages to poor households could reasonably expect low default contributing information to this learning note. to reinforce the social norm for stopping rates and high loan self-sufficiency, open defecation and position a toilet with while supporting their social mission. a natural shelter as providing the benefits About the project (such as privacy, convenience and no Sanitation loan programs are likely to be In 2012-13, the World Bank’s Water and more shame) that households seek. viable in other countries, though each Sanitation Program (WSP) worked with the area’s local context should be taken Program for Appropriate Technology in Health Figure 6: While organic shelters are less into consideration when structuring any (PATH), International Development Enterprises desired than concrete alternatives, they replication of loan programs. For exam- (iDE), VisionFund Cambodia and KREDIT to test are more affordable ple, in certain contexts, a strong local a sanitation financing program in Cambodia. The government could be a key partner and goal was to see how microcredit could be used support sanitation uptake by offering to help rural households obtain durable sanitation loan guarantees or subsidies to enable technologies and thus provide tangible health and poor households to pay for sanitation. quality-of-life benefits. Development partners can play a role in supporting sanitation loan programs PATH acted as the project facilitator. During the by sharing lessons learned in the form pilot phase, funds were provided to cover the of pilots and innovations, and resources additional operational costs each MFI incurred in such as the WASH microfinance tool- running the pilot program, but not the actual loans kits created by Water.org and Micro- themselves. Save.11 Development partners can also support MFIs in testing other innovative new lending approaches (e.g., though Contact us mobile banking) to understand the im- For more information please email pact that has on sanitation uptake. wspeap@worldbank.org or visit www.wsp.org 11 See for details http://www.microsave.net/resource/water_sanitation_and_hygiene_microfinance_toolkits. W t r WSP is a multi-donor partnership created in 1978 and administered by the World Bank to support poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. WSP’s donors include Australia, Austria, Denmark, Finland, France, the Bill & Melinda Gates Foundation, Luxembourg, Netherlands, Norway, Sweden, Switzerland, United Kingdom, United States, and the World Bank. The findings, interpretations, and conclusions expressed herein are entirely those of the author and should not be attributed to the World Bank or its affiliated organizations, or to members of the Board of Executive Directors of the World Bank or the governments they represent. © 2014 International Bank for Reconstruction and Development/The World Bank