79207 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The World Bank does not guarantee the accuracy of the data included in this work. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. The material in this publication is copyrighted. FINANCIAL SECTOR ASSESSMENT PROGRAM MALAWI TECHNICAL NOTE ACCESS TO FINANCIAL SERVICES SEPTEMBER 2008 INTERNATIONAL MONETARY FUND THE WORLD BANK MONETARY AND CAPITAL MARKETS FINANCIAL AND PRIVATE SECTOR DEVELOPMENT DEPARTMENT VICE PRESIDENCY AFRICA REGION VICE PRESIDENCY i Contents Abbreviations .................................................................................................................ii I. Overall Assessment ................................................................................................ 1 A. Demand for financial services ........................................................................... 1 B. Microfinance service providers.......................................................................... 4 II. Constraints to Increasing Access in Malawi .......................................................... 7 A. Barriers to access: physical access, eligibility and affordability ....................... 7 B. Key binding constraints to providing financial services .................................... 8 III. Mobile Banking and Access to Finance................................................................. 9 A. Technology and access to finance...................................................................... 9 B. M-Banking initiatives in Malawi ..................................................................... 10 C. Legal and regulatory framework for m-banking .............................................. 13 D. Role of government in developing m-banking ................................................ 18 IV. Payments Infrastructure and Access to Finance .................................................. 20 A. Malswitch......................................................................................................... 20 B. Streamlining the payments system................................................................... 23 V. The Role of Government in Increasing Access ................................................... 25 A. Current government financial sector participation .......................................... 25 B. Future Government financial sector participation ........................................... 26 VI. Conclusions and Recommendations .................................................................... 28 A. Understanding the demand for access to financial services............................. 28 B. Facilitating the development of m-banking ..................................................... 29 C. Enhancing the efficiency of the payment system ............................................ 31 D. Rationalizing development finance institutions and programs ........................ 32 E. Redefining the role of government .................................................................. 32 VII. Bibiliography ....................................................................................................... 33 VIII. Individuals Consulted .......................................................................................... 35 ii ABBREVIATIONS AIDS Acquired Immune Deficiency Syndrome AML Anti-Money Laundering APIP Agricultural Productivity Investment Program CDD Customer Due Diligence CGAP Consultative Group against Poverty DEMAT Development of Malawi Enterprise Trust DFID Department for International Development ECCH Electronic Cheque Clearing House FAO Food and Agricultural Organisation FIU Financial Intelligence Unit FMB First Merchant Bank FSAP Financial Sector Assessment Program G2P Government-to-person GoM Government of Malawi HIV Human Immuno Virus KYC Know Your Customer MFI Microfinance Institution MK Malawi Kwacha MPC Malawi Post Corporation MRDF Malawi Rural Development Fund MRFC Malawi Rural Finance Company MSB Malawi Savings Bank MUSCCO Malawi Union of Savings and Credit Cooperatives OIBM Opportunity International Bank of Malawi POS Point of Sale RBM Reserve Bank of Malawi RTGS Real Time Gross Settlement SACA Small Holder Agriculture Credit Administration SACCOS Savings and Credit Cooperatives SEDOM Small Enterprise Development Organization of Malawi SME Small and Medium Enterprise TNM Telekom Networks Malawi USAID United States Agency for International Development 1 I. OVERALL ASSESSMENT 1. This Technical Note the Malawi 2007 FSAP reviews the current challenges of increasing access to financial services in Malawi.1 After a cursory assessment of the state of the financial sector in Chapter 1, it summarizes the key challenges of rural financial sector development (Chapter 2); then discusses the opportunities that branchless banking technology offers (Chapter 3), the options for strengthening the financial infrastructure (Chapter 4) and the role of government in increasing access (Chapter 5). It concludes with recommendations on how authorities may seek to realize these opportunities (Chapter 6). 2. Below is a summary of the key recommendations contained in this report. Box 1. Selected Issues in Access to Finance Key Findings  A dearth of household level demand and supply information  Albeit with recent modest improvements, the continuing adverse financial condition of state-owned financial institutions and programs, notably MRFC  Tremendous opportunities for technology based solutions for increasing access to financial services  Financial institutions still competing on retail payments system infrastructure Recommendations  Conduct comprehensive market survey of demand for financial services  Move government out of active provision of retail financial services.  Create regulatory framework for outsourcing to agents of cash handling, account opening, and risk-based KYC  Move the wholesale RTGS system to the RBM  Sell retail payment system to the banking sector at a realistic price  Write down and sell Malswitch to the banking sector to allow more space for a cheaper, interoperable system A. Demand for financial services 3. Up to 90 percent of Malawi’s adult population is unbanked. Analysis reveals two substantial segments of the adult population, totaling up to 4.2 million or 60 percent of the adult population, which could be prime targets for one or more transformational approaches to branchless banking. Figures 1 and 2 describe this market segmentation in greater detail. Further study of the market is warranted to refine this analysis for public and private sector decision makers. In estimating the 1 This Note was prepared by Samuel Munzele Maimbo, World Bank and Mark Pickens, CGAP. 2 possible market for financial services, the 6 million adults in Malawi can be segmented as follows.2  275,558 (4.6 percent) are banked and have accounts with traditional commercial banks. While there are 551,116 deposit accounts in the country, 3 anecdotal evidence from Malawian bankers suggests bank customers have a minimum of two accounts each. This mirrors the situation with banked, middle- and upper-income individuals in other African economies.  313,254 (5.3 percent) receive microfinance services through an institution oriented towards low-income clients, such as the Malawi Rural Finance Corporation (MFRC). Malawi Savings Bank (MSB) and Opportunity International Bank of Malawi (OIBM) have 184,254 deposit accounts. These are joined by 29,000 clients from MFIs such as PRIDE Malawi and Finca Malawi, and an estimated 100,000 members of savings and credit cooperatives (SACCOs).4 Figure 1. Market Segments in Malawi’s Financial Sector 100% 4.6% Clients of traditional banks 90% 80% 6 mil 5.3% MF & SACCO clients 70% adults 25% “Missing middle" 60% 50% 40% 40% Below poverty line 30% 20% 25% Poorest (food insecure) 10% 0% Population: 12 million 6 million Adults  1.5 million (25 percent) individuals could be labeled the “missing middle� as they do not belong to the cohort of Malawians with some level of service (of whatever quality) from traditional commercial banks or low-income oriented banks, MFIs or SACCOs, but who have greater disposable income than the 65 percent of the population whose incomes fall below the poverty line. The missing middle is large (1.5 million), has greater disposable income than the majority of Malawian adults, and yet is unserved.  3.9 million (65 percent) of the population falls below the poverty line and are unbanked. 65 percent of population lives below national poverty line of USD 0.44 2 The figures in this section are estimated based on market interviews. The Mission recommends that a household based market survey of financial services in the near term will improve the accuracy of these estimates. 3 Data as of December 2006. Provided by RBM to FSAP team during July 2007 mission. 4 Sources: data from RBM, interviews with PRIDE and Finca, Malawi Microfinance Network. MAMN Strategic Business Plan 2007-2009. USAID, March 2007. 3 per day.5 Depending on the quality of the annual harvest, upwards of 25 percent of the population is unable to meet daily food requirements.6 These could be divided into two segments of the economically active poor (2.4 million) and the poorest (1.5 million) who struggle to meet their nutritional requirements. 4. As Figure 2 illustrates, already-banked customers of commercial banks may pay to have access to electronic banking for the convenience it provides. However, branchless banking may also appeal to two new cohorts of clients among the unbanked. One cohort is comprised of the approximately 1.8 million current microfinance customers and unbanked individuals above the poverty line. These individuals have greater disposable income than the 65 percent of Malawians under the poverty line but, so far, are either completely unbanked or, in the case of MFI and SACCO clients, are probably under-served, lacking efficient payment and remittance services which are not provided by MFIs and SACCOs. This cohort seeks access to better products, at an affordable price, particularly for those who cannot easily access a bank or MFI branch due to distance. Figure 2. Focus on 3.6 million People in Two Cohorts Focus for Branchless Banking 275k Traditional bank clients Value (and may pay for) convenience 313k MF & SACCO clients Seek better products, price, convenience 1.51 mil “Missing middle" Above poverty line, but face barriers such as price, distance which BB can address 2.4 mil Economically active Build around cash-based activities (ag. poor subsidies, crop sales, cash transfers) 1.5 mil Poorest Food insecure, out of cash economy MFI, SACCO & “missing middle� = 1.8 mil 6 million Adults Economically active Poor = 2.4 mil 5. A second cohort exists among those beneath the poverty line. This group is large, 2.4 million individuals, but due to their economic circumstances may be more difficult to serve on a sustainable basis. They are predominantly rural (as is 85 percent of the population) and employed in agriculture (as are 80 percent of Malawians).7 However, there are entry points: this group includes most of the 2 million households which receive one or more government agricultural subsidies. They are economically active, and likely save in cash or in-kind. 5 National Statistical Office, quoted in Luboyeski, et al. Microfinance Sector Assessment in the Republic of Malawi USAID, 2004. 6 Food and Agricultural Organization (FAO), quoted in Luboyeski, et al. 2004. 7 Luboyeski, et al. 2004. 4 B. Microfinance service providers 6. Since 2002, Malawi has had a well articulated cabinet approved National Microfinance Policy and Action plan which aims to increase access to financial services nation-wide. In it are guiding principles for supporting the development of the sector including: (i) crating an enabling legal and regulatory framework; (ii) supporting conducive economic policies; (iii) developing the capacity of service providers and their clients; (iv) supporting the development of industry infrastructure including the Malawi Microfinance Network; (v) promoting best practices at the level of institutions and donors; and (vi) increasing coordination among sectors on issues affecting financial services delivery. 7. However, implementation of that policy has so far produced mixed results and access to financial services remains limited. The microfinance industry remains nascent – almost all microfinance institutions (MFIs) are constrained by limited financial resources and have achieved limited outreach and sustainability. As of September 2007, the industry had an estimated 388,843 loans outstanding (US$29,916,506) and 626,560 savers (US$36,430,335) representing notable growth from June 2006, when the industry had at least 191,848 loans outstanding (MK2.99 billion) and 243,521 depositors (MK 3.2 billion)8. (see table below). Table 1. Microfinance Market Share (as at 30 September 2007) Savers Loans MF Legal Status Number Value (MK) % Number Value (MK) % Providers (Value) (Value) MRFC GoM Corporation 211,335 350,921,537 6.8% 24,238 705,560,077 16.7% MARDEF GoM Project 0 0 0% 105,015 694,162,819 16.5% SEDOM GoM Trust 1,641 1,038,715 0% 10,738 155,000,000 3.7% DEMAT GoM Trust 960 3,167,000 0.1% 990 13,710,056 0.3% MSB GoM Corporation 184,604 1,728,460,905 33.7% 49 3,746,441 0.1% Subtotals 398,540 2,083,588,157 40.6% 141,030 1,572,179,393 37.3% OIBM Bank 91,983 1,731,884,968 33.7% 13,750 1,098,575,195 26.0% MUSCCO Cooperation 62,745 1,056,266,998 20.6% 159,422 756,878,973 17.9% FINCA Cooperation 20,090 143,602,332 2.8% 20,571 300,688,818 7.1% PRIDE Corporation 14,483 53,983,667 1.1% 9,304 224,979,874 5.3% CUMO Corporation 22,479 30,810,114 0.6% 19,407 65,075,122 1.5% FITSE Trustee Act 6,173 17,000,000 0.3% 6,458 57,377,491 1.4% ECLOF NGO 0 0 0% 4,425 47,137,660 1.1% Coffee Cooperative 1,962 400,592 0% 1,796 33,159,557 0.8% SACCOs MicroLoan Project 0 0 0% 4,575 27,909,674 0.7% Found Project Project 6,060 15,169,592 0.3% 2,145 22,623,772 0.5% Hope Hunger Project 1.155 69,935 0% 5,850 11,142,500 0.3% Project NABW NGO 890 3,900,920 0.1% 110 499,277 0.0% PRDO Project 0 0 0% 0 0 0.0% Subtotals 228,020 3,053,089,118 59.4% 247,813 2,646,047,914 62.7% TOTALs 626,560 5,136,677,275 100% 388,843 4,218,227,307 100% $36,430,335 $29,916,506 Source: Microfinance Association of Malawi. 8 This number comes from adding deposit accounts for MSB, OIBM and MUSSCO based on 2006 data. 5 8. A range of institutional types offer microfinance services in Malawi. Microfinance service providers in Malawi include fully-fledged commercial banks supervised by the RBM and semi-formal MFIs, without much regulatory oversight, under several legal forms and ownership structures9 including non-governmental organisation (NGOs), cooperatives, banks, public and private companies. The Malawi Union of Savings and credit Cooperatives (MUSCCO), the National Association of Savings and Credit Cooperatives, which was created in 1980 to develop a sound network of Savings and Credit Cooperatives (SACCOS), now counts about 70 members. Several other SACCOS, not members of MUSCCO, provide microfinance services in Malawi, but accurate information on their operations was not available. Government initiatives 9. The Government of Malawi is the largest provider of microfinance services through its various companies and parastatals. Specifically, MRFC and MSB are the predominant suppliers, with MRFC traditionally being the major supplier of credit (over 25,000 accounts at 2006-end) and MSB the major supplier of savings services (over 128,000 deposit account at 2006-end). While both MRFC and MSB are fully owned by GoM, the government also has stakes in several commercial banks active in the sector, such as Stanbic, NBS, and Indebank. Besides MRFC and MSB, only SACCOs and OIBM can collect deposits. Some of the government-led development finance programs have experienced poor collection rates, which has caused concern among industry stakeholders on the negative impact that it might have on the overall microfinance repayment culture. 10. Microfinance service providers have a good presence in rural areas through MSB, MRFC and MUSCCO, although they are unable to fully meet demand. The Malawi Savings Bank, which has taken over the previous Post Office Savings Bank, has extensive coverage across the country. Together with MRFC, a government-owned corporation, it has the most extensive coverage in rural areas. MUSCCO, which was established as an apex institution in 1980, is also present in both urban and rural areas through its member cooperatives. While rural clients make up for a considerable share of the total number of microfinance beneficiaries in Malawi, this portion is still low when compared with potential demand. While there were around 200,000 microfinance loan accounts as of December 2006, demand was estimated at levels between 300,000 and 760,000, based on 2003 data.10 Poor infrastructure and high operational costs are among the biggest constraints for microfinance providers in the rural areas. Microfinance institutions 11. Microfinance providers have generally recorded weak financial performance. FINCA and Pride Malawi, for example, operating in Malawi since 1994 and 2000 respectively, have experienced high repayment problems in the past few years. Pride recorded a PAR>30 days of 21.80 percent at 2006-end.11 FINCA seems to have been more effective at addressing this problem (PAR>30 days of 7.58 9 Luboyeski, et al. Microfinance Sector Assessment in the Republic of Malawi USAID, 2004. 10 Luboyeski, et al. Microfinance Sector Assessment in the Republic of Malawi USAID, 2004. 11 The MIX Market, 2006 data. 6 percent at 2006-end), although stronger efforts are still needed to reach a better portfolio quality. Similarly, client growth has been slow. In 4 years, FINCA’s clients grew from 18,998 in 2002 to 19,943 in 2006 and 20,090 in September 2007. One of the problems seems to be the high- rates of clients leaving microfinance programs experienced by microfinance companies in Malawi. A study of Pride Malawi indicated that more than 50 percent of its clients dropped out since the institution were established in 2000.12 12. Microfinance in Malawi is mostly credit-oriented with agriculture-related credit in rural areas often being provided as in-kind inputs of fertilizer and seeds. While a good number of microfinance institutions operate in Malawi, a considerable share of its population is still unbanked or with very limited access to financial services. About 200,000 Malawians have a loan account with one of the several microfinance providers, but demand (based on need and willingness to repay) is estimated to be between one-and-a-half and five times this level of supply, with the greater gap found in the rural areas.13 Commercial banks 13. There are exceptions to this poor performance. OIBM, a relatively new player in Malawi having started operations in 2003 as a fully-licensed bank, has experienced good client growth and overall high performance. In three years it reached 9,300 active borrowers and close to 56,000 savers.14 It also reported only 2.2 percent of its portfolio at risk (PAR >30 days). This is quite impressive when compared with the other players. OIBM has offered smart cards to its clients to enable them to use Malswitch, the Central Bank’s payment system. It is also now experimenting with emergency cash transfers in partnership with Concern (see Opportunities section). The experience of OIBM, which has quickly emerged as one of the highest performers in the national microfinance industry, albeit accepting its high cost base, might act as a catalyst for other commercial banks to directly engage microfinance clients. Until that happens, however, service to the sector remains limited. 14. The provision of financial services in rural Malawi remains a challenge. With 86 percent of the country’s population living in the rural areas, increasing the availability and access to financial services to serve agricultural and rural markets is a key priority in Malawi. During the last decade, numerous government development finance institutions and initiatives aimed at increasing the outreach of financial services and improving the performance of financial institutions have been introduced. However, many rural parts of the country still lack adequate supply of formal financial services; even the expansion of microfinance institutions has not changed this significantly, especially for small and medium scale enterprises. Absence of reliable electricity, road and telecommunications network is associated with the high costs of doing business for financial institutions. Consequently, a large number of microfinance programs are concentrated in urban centers leading to competition for the same clients. Although many institutions have entered the financial sector, their 12 Luboyeski, et al. Microfinance Sector Assessment in the Republic of Malawi USAID, 2004. 13 Kiendel Burritt, Expanding Financial Services in Malawi, UNCDF, 2007. 14 The MIX Market, 2006 data. 7 networks are concentrated in cities and other semi-urban centres, making the rural masses unable to access banking facilities. II. CONSTRAINTS TO INCREASING ACCESS IN MALAWI A. Barriers to access: physical access, eligibility and affordability 15. There are several factors contributing to Malawi’s limited access to finance, especially in rural communities. Specific challenges include: low client income levels, risks arising from rain-fed agriculture economic dependency; high HIV/AIDS prevalence; high cost of finance including high collateral requirements; low financial literacy; and limited capacity of microfinance service providers. These constraints are summarized below:  Low client income levels: For some of the financial institutions in the country, rural individuals and the small and medium scale enterprises do not present a viable business proposition. With a GDP per capita income of only US$175, Malawi is a poor country. The rural population is, on average, poorer than their urban counterparts and is, therefore, a less attractive business segment. It is estimated that 52 percent of the population (6.4 million people) live below the poverty line and 22 percent (2.7 million people) still live in ultra-poverty (World Bank 2007: 3).  Risks arising from rain-fed agriculture economic dependency: The rural economy is greatly affected by Malawi’s dependence on rain-fed agriculture. Significantly, close to 90 percent of the population lives in rural areas and is engaged in smallholder, rain-fed agriculture, making it highly vulnerable to deviations from the average annual rainfall pattern. The risks of providing financial services are compounded by the vagaries of weather and swings in commodity prices. The last two seasons have enjoyed good rains. However, the previous two were disastrous and, because only an estimated 40,000 hectares out of a potential 500,000 are irrigated, the risk of a poor harvest on account of low rainfall in any given season remains high (USAID 2007:45).  HIV/AIDS prevalence: The economic viability of rural communities has further been affected by illness or death resulting from the HIV/AIDS epidemic. The estimated prevalence rate among prime age (15-49) adults of 14 percent in 2005, places Malawi eighth in the world in terms of the severity of its HIV/AIDS epidemic. A recent survey showed that two thirds of all households reported that someone was ill or had an injury in the past two weeks. One-third of households had a chronically ill person in the household. These people were chronically ill for an average of 7.8 years. In terms of coping strategies, households helped themselves by using their past savings and borrowing or by taking drastic measures such as fostering out children and, removing children from school.  High cost of finance, including high collateral requirements: While the cost of opening and maintaining an account is comparatively low (and getting lower), there are some that consider the amount too high for the lower income groups. On average, customers are required to provide a minimum deposit of MK500 (USD 3.50) to open an account, and will require approximately MK600 (USD 4.20) to maintain it. These costs, taken together with the HIV/AIDS pandemic, and the below-inflation rates of returns on short term 8 current account deposits, explains, in part, the limited appeal of investing in financial services instruments. Further, Firms also have to contend with the high collateral requirements for funds. Traditionally, banks have required high collateral requirements of clients – largely in the form of 100 percent of the value of the loan in the form of titled land and property. Unfortunately, all land transactions still require the Minister of Lands to personally give permission to transfer, which can lead to considerable delays in enforcing foreclosure, especially through the court system which is infamously slow.  Low financial literacy and business development skills As in many parts in Southern Africa, financial literacy remains a challenge in Malawi. Limited education opportunities and weak academic infrastructure in rural areas means that financial access is further limited by the lack of knowledge of financial products and services. For many, knowledge barriers may be greater than the physical access barriers. The lack of financial literacy affects both rural finance practitioners as well as clients. For households, there is a need for basic financial numeracy skills, and for entrepreneurs, business development skills.  Microfinance institutional capacity: For practitioners, there remains a great need for training on good microfinance theory and practice. These include: (i) Gender and Development to highlight the factual importance of the roles and status of women to local and national development and, therefore, the need to support them with access to financial services; (ii) Microfinance Industry and Principles to demonstrate the evolution and development of the microfinance industry and best practice principles as a tool for poverty reduction and as an economic development approach intended to benefit low-income women and men, and; (iii) Microfinance Product Costing and Pricing to support the increasing professionalism of MFIs and the competition in the micro-finance market. B. Key binding constraints to providing financial services 16. In addition to the constraints listed above, three in particular were consistently referred to in all the discussions with financial service providers in the country as being especially challenging.  Physical infrastructure: Poor physical infrastructure contributes to the high cost environment for rural financial institutions. Electrification is very limited with on 2 percent of households in rural areas connected to the grid (compared to 30 percent of households in Lilongwe and 35 percent in other cities). Virtually all phone owners, either land or cellular, are in the highest two income deciles, and only 13 percent of rural roads are tar or asphalt and rural communities are on average 20km from a sealed road (40km in the North). (World Bank 2007:16-17). This poor infrastructure is associated with the high costs of doing business for financial institutions. Consequently, although many institutions have entered the financial sector, their networks are concentrated in cities and other semi-urban centres, making the rural masses unable to access banking facilities.  Financial Sector infrastructure: Current technology and operational arrangements with Malswitch are generally viewed as not being able to fully meet the banking sector’s business needs or provide the level of flexibility 9 required for efficient customer service delivery. The view that Malswitch card network operations were “high cost and inefficient� is widely held, by both users and non-users. Retention in its present form is not viewed as an acceptable solution or a viable investment proposition for the banks. An efficient organization of the payment systems through inter-technology interoperability generates substantial social gains, and if properly managed the potential also for financial gains in the medium to long-term. This will be key to significantly improving the financial infrastructure in the country.  The role of government in the financial sector: Overall, private financial sector perception of government-led institutions and programs is negative. The programs are perceived to be failing to reach their intended targets, and when they do, that it is at unduly high subsidy costs. It is argued that these operations face major incentive and governance problems leading to such recurrent problems as poor loan origination and even poorer loan collection (thereby fostering a non-payment culture), wasteful administrative expenditures, overstaffing, and capture by powerful interest groups – all factors, in many countries, resulted in large losses and the need for recurrent recapitalizations, at very high fiscal costs. For these views, the private financial sector therefore welcomed the November 2006 Ministry of Industry, Trade and Private Sector Development revised microfinance policy statement stating that the government will ‘refrain from interventions such as subsidies and interest rate controls that distort the operation of microfinance markets�. 17. The challenges and risks noted above can render a rural financial strategy not viable for many institutions, especially those that are already established in urban centers, highly profitable and without compulsory mandates to serve rural sectors. The business rationale for investing in rural communities for most banks and larger financial institutions has hitherto been weak. This means that rural households have inevitably been dependent on government or donor programs that have proven unsustainable, limited services offered by informal systems, or self- finance. To succeed, there are generally three recognized pillars of a successful rural financial sector program: (1) government policies and the legal and regulatory framework, (2) financial sector and real sector infrastructure, and (3) financial institutions. III. MOBILE BANKING AND ACCESS TO FINANCE A. Technology and access to finance 18. Expanding access to finance remains an important policy challenge in many countries. Globally, governments and practioners have pursued a wide range of options. Hitherto, efforts have focused on reforming institutions, developing infrastructure to take advantage of technological advances, encouraging competition, and providing the right incentives through prudential regulations. Many of these initiatives are discussed in other Notes of this FSAP. This particular section highlights the potential for using technology – specifically, M-banking technology - to increase access in the country. 19. Low income households face numerous challenges in accessing financial services including high transaction costs and long distances to the nearest access 10 point. The concept of branchless banking is premised on the notion that the delivery of financial services by mobile phones will substantially reduce transportation and transaction costs. It offers the opportunity to overcome some of these obstacles by the fact that the cost of the transactions conducted on phones is comparatively minimal and it does not require physical access to a bank. 20. Technology-based approaches can bring banking services to new customers and regions on a commercially viable basis, often for the first time. This is “transformational branchless banking� – the use of information and communication technologies to reduce costs of delivering financial services to clients previously beyond the practical reach of traditional bank channels.15 This is different from additive models, which merely provide another channel for customers who already have a bank account. As banking in Malawi’s rural areas, where the majority of its low-income population lives, faces considerable constraints due mostly to poor road and communication infrastructure, the potential efficiency gains from introducing branchless banking is considered very high 21. To date, transformational branchless banking models have displayed four features:  Use of technology and agents, to lower costs and reach new areas and clients;  Risk-based Know your customer (KYC) standards that ease account opening for low-income people, including via agents and electronic channels;  Regulatory environments, which define an appropriate role for nonbanks, as there are bank-based and nonbank-based approaches possible, and in a number of countries nonbanks lead the way with innovation; and  Early emphasis on payment services, but increasing linkage to credit and other services in demand among low-income clients. 22. The use of technology and agents is a key characteristic of transformational models. Mobile phones and point-of-sale (POS) terminals can serve as payment devices which leapfrog distances and obviate the need to be near a bank branch. Agents are a complementary piece of the puzzle. In many models, grocery stores, petrol stations, and post offices, act as agents who receive and dispense funds on behalf of the provider. In short, technology carries the data, but agents provide the connection between the electronic channel and cash economy in which poor people live. By leveraging both, providers can grow service networks exponentially into areas where it would be unprofitable to place a full scale branch. B. M-Banking initiatives in Malawi 23. Malawi has several pioneers experimenting with branchless banking, including approaches to delivering microfinance services, payments to tobacco farmers, and conditional cash transfers to the rural poor. These initiatives are relatively recent, like in other countries at the forefront of transformational branchless 15 Lyman, Timothy, Mark Pickens, David Porteous. Regulating Transformational Branchless Banking: Mobile Phones and Other Technologies to Increase Access to Finance. Forthcoming. CGAP, DFID. 2008. Also see Porteous, David. Enabling Environment for Mobile Banking in Africa. DFID, 2006, and Lyman, Timothy, Gautam Ivatury, Stefan Staschen. Use of Agents in Branchless Banking for the Poor: Rewards, Risks and Regulation. CGAP, 2006. 11 banking. Over 100,000 low-income Malawians use one of these electronic payment and saving instruments. Compare this to the estimated 275,000 individual customers of traditional commercial banks. Innovation with technology is happening at the lower-end of the market in Malawi, and at some sizeable scale. This is unusual in a global context. This makes the timing opportune for the Government of Malawi to examine the potential of branchless banking. 24. Malawi has several pioneers experimenting with card-based approaches to delivering microfinance services (OIBM), payments to tobacco farmers (FMB), and conditional cash transfers to the rural poor (Concern Worldwide in connection with OIBM). Malswitch’s smartcard offering retail payments switch was an early pioneer in Africa, but has met with mixed success, particularly in creating an interoperable, affordable payments switch. Though expensive, the smartcard is optimized to deliver a suite of services to areas without communications connectivity. However, this functionality has not been taken full advantage of yet. These initiatives are all relatively recent, like in other countries at the forefront of technology-based approaches in financial services such as Brazil, Kenya, the Philippines and South Africa, among others. This makes the timing opportune for the Government of Malawi to examine the potential of branchless banking. Experience from other countries shows that adequate regulation and supervision is very important for technology-based approaches to be developed and adopted, while lack of regulatory clarity dampens provider motivation to innovate and devise transformational branchless banking models. 25. Innovation with branchless banking is happening at the lower-end of the market in Malawi, with First Merchant Bank (FMB), Opportunity International Bank of Malawi (OIBM) and the Dowa Emergency Cash Transfer program (DECT) all using technology to deliver services to low-income Malawians. With over 100,000 clients benefiting from card-based schemes, the scale compares favorably to the rest of the financial system (with approximately 275,000 customers of traditional commercial banks). This places Malawi among a small cohort of countries where transformational branchless banking is underway. 26. First Merchant Bank’s Makwacha Card is a magnetic strip debit card offered to farmers receiving payments from the sales of their produce by Malawi’s Tobacco Control Commission. Clients can use the card to store their tobacco payments and make withdrawals or purchases via POS terminals at 60 locations of Farmer’s World, a nationwide retailer of agricultural inputs, construction materials, home furnishings and groceries. Approximately 15,000 cards have been issued to “tobacco clubs� with over 25,000 members. FMB reports that it surpassed USD 3.5 million in transactions in May 2007, taking it past breakeven for the product in just over one year. The primary revenue generator for FMB is capture of foreign exchange spread between the USD value paid by TCC and Kwacha paid out to clubs (See FSAP note on Monetary Policy for a detailed discussion on foreign exchange transactions and their impact in the financial market). In placing POS terminals at 60 Farmer’s World locations, FMB established a service network larger than all other banks combined, which have 52 branches. 27. OIBM has issued smartcards to 70,000 microfinance clients, using the Net1 UEPS solution from South Africa, offered via Malswitch. Customer account data is contained on a chip embedded in the card, which is periodically synched to 12 OIBM’s server whenever a client accesses a smartcard-enabled POS or ATM. Smartcards can also support card-to-card transactions without a connection to OIBM. Thus, in theory, the smartcard can operate in a fully offline environment common in rural areas. However, OIBM has not experimented with this functionality except in a limited emergency cash transfer pilot with Concern Worldwide. The majority of OIBM client transactions happen through online ATMs and POS terminals at OIBM branches. OIBM clients will soon be able to make purchases and withdraw funds via POS terminals in Blantyre and Lilongwe merchants with reliable electricity, landline connection, and adequate liquidity. The Net 1 solution also includes thumbprint readers at ATMs and POS terminals, which OIBM reports has advantages with low- income clients for whom remembering a PIN may not be easy. 28. OIBM also joined forces with Concern Worldwide for a short-term program delivering emergency cash transfers to 10,500 food-insecure households in the Dowa region.16 DFID provided financing support, OIBM-issued smartcards and an OIBM truck-mounted mobile branch was used to distribute five cash payments to recipients during the 10 month program. OIBM has committed to continue serving the Dowa region for the next year via a mobile branch. DECT clients are among the lowest-income, most rural clients included in Malawi’s formal financial sector. 29. Most Banks and Mobile Operators are still on the sidelines. Apart from the early results of FMB, OIBM and Concern, there are no other mainstream commercial banks using technology to go downmarket. Appetite is probably dampened by very healthy profits: the FSAP calculated Return on Average Assets (ROAA) at 6.85 percent across the commercial banking sector.17 This compares very favorably to the global average ROAA for banks, or even microfinance institutions. Banks and microfinance institutions had ROAA of 1.5 percent and 2.8 percent respectively in a global study of 1799 banks reporting to Bank Scope and 344 MFIs reporting to The MicroBanking Bulletin.18 Technology, however, could help Malawian banks become more efficient, which in turn could help them deal with high overhead costs. Overhead accounts for three-quarters of the average 15.9 percent interest rate spread charged in the market.19 30. Banks may change their tune on using technology to go after low income clients if one of the early pioneers demonstrates success on a fully commercial basis. Senior managers at mainstream banks also indicated they lacked basic data that would enable them to pinpoint profitable opportunities and “sell� their board the worthiness of doing so, especially in terms of opportunity cost to deploying internal capital and human resources to traditional, better understood opportunities. Lack of interoperable switching infrastructure also means providers that pursue branchless banking cannot leverage existing ATM or POS networks, but instead must build largely from scratch. This raises the cost of implementation. Finally, managers also indicated lack of regulatory clarity on some key features of transformational business models, chiefly outsourcing to agents. 16 Pearson, Roland and Craig Kilfoil. “Dowa Emergency Cash Transfer (DECT): Wider Opportunities Evaluation and Recommendations.� Concern and DFID. June 2007. 17 Analysis by World Bank, IMF FSAP team. 18 Littlefield, Elizabeth, Brigit Helms, David Porteous. Financial Inclusion 2015: Four Scenarios for the Future of Microfinance. CGAP. October 2006. 19 Analysis by World Bank, IMF FSAP team. 13 31. To date, mobile operators evidence little interest in transformational business models. TNM and Celtel have 640,000 subscribers, or six percent market penetration. Senior managers at both networks believe there is substantial growth to be won just with new voice customers. Competition between the two, while evident, does not seem to be biting as sharply as in other countries, where mobile network operators are investing in value-added services like m-payments and m-banking to boost revenues per user and stiffen customer loyalty. As a result, neither TNM nor Celtel have current plans to launch a mobile-based payment or banking platform aimed at low-income people in Malawi. Many interviewees referred to an episode earlier in 2007 when telecommunication service was disrupted on a nationwide basis for several weeks following a fire in facilities of one telecommunications companies. Both Celtel and TNM are rolling out GPRS this year, which is a considerably technical undertaking. 32. Early successes have been built around reliable streams of payments. OIBM’s smartcard leverages microfinance clients’ recurring loan and deposit transactions. The use of the OIBM smartcard in Concern’s DECT Program was justified by the need for Concern to deliver a series of recurring cash payments to low-income households experiencing a food emergency. FMB relies on disbursement of proceeds from tobacco sales from a government entity to farmers across the country. 33. The Government of Malawi is responsible for a huge volume of regular payments: Salaries and pensions for current and past workers, annual delivery of agricultural subsidies to more than 2 million households, and agricultural and microfinance loans via MFRC and Malawi Savings Bank (MSB). The agricultural subsidy program alone is budgeted at MK 10.7 billion (USD 77 million) for payments and administration to deliver an estimated 4.2 million individual subsidy payments. Routing these payments through electronic channels could provide further encouragement for innovation, by supporting branchless banking business models with reliable streams of payments. Doing so, however, should be contingent upon the approach being shown to be cost-effective for the government, as costs of switching out of electronic channels could be high once entered into. C. Legal and regulatory framework for m-banking 34. Branchless banking has the potential to increase poor people’s access to financial services if regulation provides openness and safety on several key topics: (a) eases account opening (both on site and remotely) while maintaining adequate KYC standards; (b) permits the use of a wide range of agents outside bank branches, thereby increasing the number of service points;; and (c) permits a range of players to provide payment services and issue e-money, thereby enabling innovation from market actors with motivation to do so. 35. From a regulatory perspective, it is useful to think of branchless banking as consisting of two basic models: bank-based and nonbank-based.20 The regulatory significance of the distinction lies in the fact that behind every transaction under the bank-based model, there stands a fully prudentially licensed and supervised financial institution. As it stands in many countries, there is currently no regulatory 20 Lyman, Pickens and Porteous, 2007. 14 window for supervising nonbanks, though there may be loopholes which allow them to be active. This puts a premium on establishing a mechanism to engage with, monitor and/or supervise nonbanks eager to do branchless banking. But the bank- based approach is not free of risks. The presence of a prudentially-regulated and supervised financial institution may give policy makers false comfort. Evidence from the countries studied shows that, in some cases, the bank involved in the bank-based model may have outsourced so much responsibility—and risk—to nonbank partners that it in effect shifted the primary focus of regulatory concern from the prudentially licensed bank to its unlicensed partner. WIZZIT in South Africa, Smart Money in the Philippines, and Beeline in Russia are examples of a nonbank leading a bank-based model. Know your Customer requirements 36. Risk-based KYC rules are also important in many branchless banking models. Low-income clients often have difficulty meeting documentary requirements of identity, address or income required to open a bank account. In some countries, policymakers and regulators recognize this and are making adjustments to enable low- income clients to meet KYC standards.21 This often takes the form of widening the range of documents with which customers can establish their identity, or removing some KYC requirements altogether. The latter is often accompanied by limits on account functionality. In South Africa, for example, Exemption 17 does away with the requirement to obtain proof of address and limits account functionality via maximum account balance and daily transaction limits (USD 3,450 and USD 690 respectively). WIZZIT has relied on Exemption 17 exclusively to sign up 150,000 customers, most of whom are low-income and previously unbanked. 37. A Money Laundering, Proceeds of Serious Crime and Terrorist Financing Act (often referred to as the “Money Laundering Act�) was passed into law on August 22, 2006. The Act requires banks to conduct close scrutiny of clients and reporting to a Financial Intelligence Unit (FIU). As of the time of the FSAP mission (July 2007), a FIU was still in formation and RBM was still in the process of deigning regulations under the Money Laundering Act. In the interim, RBM-issued guidelines from 2005 are still in effect for customer due diligence (CDD). The Bankers Association of Malawi has developed standard account opening and CDD forms, with individual banks developing complimentary internal controls and procedures to be in compliance with the 2005 guidelines. 38. According to the guidelines, RBM’s main rationale at the time of drafting was to mitigate risk from “absence of KYC principles which can subject banks and financial institutions to serious customer and counter-party risks� and “loss of depositor confidence� in the occurrence of financial crime in one or more banks.22 The implications of KYC standards on access to financial services by low-income individuals are not described as a guiding motivation in the 2005 CDD Guidelines. The Guidelines establish a principle of risk-based due diligence and mention the need for graduated customer acceptance policies, particularly to allow for more intensive 21 Isern, Jennifer, David Porteous, Raul Hernandez-Coss, Chinyere Egwuagu. AML/CFT Regulation: Implications for Financial Service Providers that Serve Low-Income People. CGAP. 2006. 22 Reserve Bank of Malawi. “Customer Due Diligence for Banks and Financial Institutions.� Directive No.Do1-2005/CDD. Part II, Sec. 2, subsection 1. 15 CDD with higher-risk clients. The Guidelines do not mention the converse: the possibility of establishing reduced due diligence requirements for lower risk clients. Conversations with RBM in Lilongwe revealed a high degree of awareness about the potential for KYC standards to block access to finance for the poor, and an interest in devising risk-based standards which would limit this effect. 39. Under the 2005 CDD Requirements, banks are required to establish the identity of its customers, but given some broad flexibility. Banks are required to “identify prospective customers on the basis of a reliable and independent source document� and give some examples (passport, driving license, birth certificate, or letter from an existing customer or employer).23 Special attention is given to the risks of relying on a letter from existing customer or employer as a form of identification, and banks are reminded that this method “does not in any way remove the ultimate responsibility of the recipient bank or financial institution to know its customer and business.� More conservative bank compliance officers may view this as effectively invalidating letters of introduction as a means of KYC, unless accompanied by another reliable source. The considerable degree of flexibility accorded to banks may also leave some unsure about what KYC methods are actually permitted and which are not, leading them to be conservative in requiring multiple forms of identity proof, or only accepting a few types (e.g., passport, drivers license), which may bar low- income clients from service. 40. The Money Laundering Act of 2006 extends the types of information which banks must obtain and verify to include not only identity but also address and occupation. In addition, banks should take “reasonable measures� to establish the source of wealth and source of property of clients.24 In the 2005 Guidelines, establishing source of income was only required if the bank believed it to be necessary, presumably as part of its assessment of a client’s risk profile. 41. While the Money Laundering Act appears to tighten the requirements on types of information (address, occupation and source of income required), it appears to contain a new window for conduct of KYC by third parties. The Money Laundering Act permits financial institutions to rely on “an intermediary or third party to undertake [the bank’s] obligations� in obtaining and verifying information about customers, with several provisos.25 The bank must be able to:  “Immediately� obtain the information about the customer,  Ensure copies of “identification data and relevant documentation� are available to it upon request, and  Satisfy itself that the third party has measures in place to report to the FIU any transactions where customers were unwilling to produce identification or accounts opened under a fictitious identity, and maintain records on all transactions for a minimum of seven years. 23 Reserve Bank of Malawi. “Customer Due Diligence for Banks and Financial Institutions.� Directive No.Do1-2005/CDD. Part IV, Sec. 2, subsection 2. 24 Government of Malawi. Money Laundering, Proceeds of Serious Crime and Terrorist Financing Act. Act No 11 of 2006. Part III, Section 24, subsection 2(b). 25 Government of Malawi. Money Laundering, Proceeds of Serious Crime and Terrorist Financing Act. Act No 11 of 2006. Part III, Section 24, subsection 6. 16 42. The ability to use agents is an important consideration in transformational branchless banking models where agents may serve as the primary and possibly sole interface between customer and bank, especially with remote areas. The practical usage of this opening will depend in some measure on how RBM writes implementing regulations. The requirement for the bank to be able to obtain information about the customer “immediately� suggests electronic channels will be necessary or desirable. Depending on how regulations are written, smaller stores prevalent in rural areas may have difficulty meeting some of the requirements in law. For example, the requirement to maintain records for seven years could be costly, and may make it difficult to find suitable agents beyond Blantyre, Lilongwe and any other location with established, larger retail merchants. Non-bank retail agents 43. Little guidance is available to banks on outsourcing. RBI has not issued guidelines on outsourcing of functions for banks, and other regulations are silent on the topic. The Banking Act is also silent on banks outsourcing functions like cash handling (deposits and withdrawals) and account opening to agents. The act does require banks to obtain prior written approval of the Reserve Bank for “opening and closing of branches and static of mobile agencies�.26 This appears to require that all places of business require prior RBM approval. 44. Beyond this, as mentioned above, RBM has advised at least one bank that there would be limits to using agents, such as requiring all deposits be physically transported to a branch of the bank by day’s end. However, the same bank was permitted to do cash disbursements at agents. The regulatory response seems to be premised on a principle of deposit-taking as a core banking function, and that such core functions should be tied closely to branch-based operations. Clause 44 of the Banking Act states “No person, other than a bank or financial institution, shall advertise that he performs financial services in Malawi normally rendered by banks and financial institutions.� There is no clear guidance from RBM on what this means in practical terms when an agent-based model is proposed. In its absence, banks would be left to observing what RBM has permitted other banks to do on a case-by- case basis and gauging their own risk appetite for attempting to do something more, with the potential cost of falling afoul of the regulator. Banks may be more likely to err on the side of caution, and opt against innovation which may result in extending services to new clients or areas using new means. There is a Banking Bill in draft, but the FSAP team did not have access to the draft text during the mission. Non-Bank partners 45. Nonbanks also have a role to play in sparking innovation. Traditional commercial banks are not always the institutions most motivated to experiment with branchless banking. A host of nonbanks, as well as banks, have interests in getting into the space where technology, financial services, and the poor converge. Mobile operators increasingly operate in competitive markets and see already slim margins growing slimmer. They see mobile banking as a value added service increasing 26 Government of Malawi. Banking Act. Enacted by the Parliament of Malawi on December 29, 1989. Part III, Section 25. 17 customer loyalty and generating revenue. In Kenya, the largest mobile phone company, Safaricom, has started an m-payments service, M-Pesa, which has attracted 700,000 clients since launch in March 2007. M-PESA was piloted in connection with a microfinance institution (MFI), Faulu. 46. In a number of other countries, nonbanks, particularly mobile network operators, are proving to have at least as much and in many cases more appetite for innovation with mobile-based payment and banking. This is not presently the case in Malawi. Neither mobile network operator indicated they have current plans to launch such services, though both were aware of and closely following developments in neighboring countries. Kenya, where Safaricom’s M-Pesa mobile payments service has attracted 700,000 customers in 7 months, is particularly closely watched. Though mobile-based services may not launch tomorrow in Malawi, it is worth examining the legal and regulatory environment. 47. The Banking Act contains statements which, on initial reading, indicate a space for a nonbank to conduct an M-Pesa type operation. Like in Kenya, the Banking Act in Malawi contains a two part definition of “banking business� as “the business of receiving funds from the public by accepting demand, time and savings deposits or by borrowing from the public or other banks, and of employing such funds, in whole or in part, by granting loans, advances and credit facilities and by investing or by any other means.�27 In short, banking is taking of repayable funds and the employment of said funds. 48. In Kenya, Safaricom has established a trust company which holds funds deposited into M-Pesa accounts, but shields Safaricom from access to interest earned on the balances (which are held in a pooled account in a commercial bank). Safaricom claims they may be taking deposits, but are not doing banking since they are not benefiting from employment of the deposits. Initial review suggests the Banking Act would permit a similar setup in Malawi. 49. However, other parts in the Banking Act effectively shut the door on this. Clause 44 states “No person, other than a bank or financial institution, shall advertise that he performs financial services in Malawi normally rendered by banks and financial institutions.� The principle seems clear: only banks should do banking. Further, the definition of deposit very clearly reserves the activity of taking deposits to banks: a deposit is defined as “money entrusted to a bank or financial institution and accepted by it for credit to a depositors account.�28 50. There is currently a draft Payment Systems bill which could provide a slightly more open space for nonbank participation in payments business. “Payment system provider� is defined in such a way that it would include entities which “establishes a clearing or payment settlement system� or a “holder of stored value�. This would appear to provide a possible category of licensing for entities involved in facilitating delivery of payments, which may involve storing of value for furtherance of such business. This would need to be clarified in implementing 27 Government of Malawi. Banking Act. Enacted by the Parliament of Malawi on December 29, 1989. Part Preliminary, Section 2, subsection 1. 28 Government of Malawi. Banking Act. Enacted by the Parliament of Malawi on December 29, 1989. Part Preliminary, Section 2, subsection 1. 18 regulations to indicate whether this is RBM’s intent. The draft bill is also silent on the use of agents for facilitating receipt or delivery of payments. Products and services 51. To date, most transformational branchless banking models rely on payments driving transaction volume and revenues. This includes airtime top-up for prepaid mobile service, bill payments, government to person payments (pensions, social benefits), and remittances, both domestic and international. Payments may present a simpler business case (one fee, one transaction) than credit, for example, which requires some knowledge of the customer’s risk profile and the charging (as well as collection) of interest over a certain period. 52. However, a growing numbers of providers see potential in going well beyond payments. Some are microfinance institutions, attracted by the opportunity to scale up their lending to the poor, enhance convenience to customers for loan repayments and achieve cost-saving new efficiencies. Others are finance companies or banks which see data on a client’s payments history, or receipt of pensions or social payments through a branchless banking channel, as a basis from which to lend securely to the client. Still others are institutions hungry for new sources of retail deposits. 53. Lack of clarity on these topics can dampen provider motivation to devise transformational branchless banking models. OIBM staff indicate they decided not to include deposit services as part of their rollout of 15 POS-equipped merchants due to perception that such was not permitted. FMB management indicate RBM informed them that Farmer’s World stores could only accept deposits if FMB ensured collected cash would be at a FMB branch by the close of business. The outsized costs of cash transport would eliminate the cost savings of extending service via Farmers World stores. D. Role of government in developing m-banking 54. Policymakers and regulators have a key role to play to see that branchless banking takes off rapidly, expanding access to finance, but safely, with appropriate protection for customers and the financial system. Experience suggests authorities have three avenues for action:  Create incentives that encourage innovation;  Encourage establishment of financial system infrastructure, and its good governance;  Establish proportionate regulation which provides an open but safe environment. 55. In many countries, government salaries, pensions, welfare benefits and other payments can be leveraged to create incentives for innovation around branchless banking. These government-directed flows account for a large proportion of total transactions in many national payment systems. By channeling these towards new payment mechanisms, government authorities can put branchless banking into a more attractive light for senior managers in the private sector. In Brazil, for example, the government routes most pension and social welfare payments through banks 19 which use banking correspondents, providing a backbone of transactions that underpin the business model for branchless banking in Brazil. 56. Government also has a role to play in establishing financial system infrastructure, and its good governance, including cost-effective, interoperable payments infrastructure. Banks should compete on quality of service, not ownership of infrastructure. Payment system infrastructure should also be optimized to help expand access to finance for unbanked segments of the population. Interoperability extends the number of service points for all customers, and can help control costs to infrastructure improvements. Governance in payment systems is equally key. Control by a few institutions can lead to high interchange fees that eventually get passed down to clients in high costs of service. South Africa, for example, has some of the highest bank fees in the world,29 due at least in part to control of the national switch by a few banks. 57. Public confidence in new channels will be key to supporting customer adoption. A major episode of loss of funds, breach of privacy, or consumer protection failure could “poison the well� for all players in the market. Regulators can work with innovators to ensure there are redress mechanisms for customer complaints, agents are well-trained and hold adequate liquidity to ensure quality service, and customers’ private data is adequately protected. Regulators will also need authority to monitor, and if necessary, block entrance into the market for operators who may not have the financial, technical and management wherewithal to provide a safe service. 58. Policymakers should seek to establish a proportionate regulatory approach to branchless banking which matches the type and level of protection to differing levels and types of risk involved and the costs that regulation will entail. This is likely to require some changes to current laws and regulations. There exist two necessary but not sufficient conditions for the regulation of branchless banking: first, retail agents must be allowed to handle “cash in/ cash out� transactions; and second, anti-money laundering (AML) rules for the opening of new accounts must be risk sensitive. However, we are able to go further to identify four “next generation� policy and regulatory topics which will figure in the success and sustainability of branchless banking in any country: e-money issuance by nonbanks, consumer protection, payment systems, and competition policy.30 Adjustments in these areas can be coupled with delegation of authority to the central bank to issue detailed directives worked out in consultation with providers, or to approve proposals to use innovative approaches on a case-by-case basis. 59. Even prior to adjusting law and regulation, authorities can send a clear signal to the industry about the central bank’s intention to encourage innovation, with due protections. For example, the State Bank of Pakistan recently issued a draft Position Paper on Mobile Phone Banking which identifies proposed changes in regulation to permit branchless banking aimed at the poor. This provides banks and others the clarity required to begin designing proposals in anticipation of legal and regulatory changes. 29 “SA bank fees twice global norm� available via http://www.fin24.co.za/articles/default/display_article.aspx?Nav=ns&ArticleID=1518-1786_2133158 30 For detailed discussion of each topic, based on evidence from 7 countries where branchless banking is taking off, see Lyman, Pickens and Porteous, forthcoming from CGAP and DFID. 20 IV. PAYMENTS INFRASTRUCTURE AND ACCESS TO FINANCE A. Malswitch 60. Malawi has made great progress with the modernization of its payment system and has some systems that are equal to, or better than, the other countries in the region. Perhaps the biggest challenge being faced by the NPS of Malawi today is the potential “fragmentation� of the infrastructure for issuing card based payment instruments (smart, debit and credit cards). Although banks cooperate on most payment systems reform initiatives like the implementation of the RTGS and ECCH, they are yet to achieve interoperability between their individual ATM and POS platforms. This section focuses on Malswitch, the proprietary smartcard and switching platform. 61. So far, Malswitch has not brought an interoperable, affordable retail payment switching capacity to Malawi. Outside of OIBM customers, few active smartcards are being used for banking. Even among OIBM customers, it appears usage of ATM and POS channels is relatively low. To date, the big banks have not joined Malswitch and are instead developing Visa-certified systems. If no action is taken, Malawi is headed down a path of developing at least two parallel payments infrastructures, without interoperability but with considerable expense hard to justify in a market the size of Malawi. There is also a real probability such costs would be passed on to clients in high fees, affecting affordability of banking services. 62. State of Affairs with Malswitch: Malswitch was launched in 2002, using a technology package provided by Net1 Aplitech, a South African firm. The solution centers around a switching platform for a proprietary smartcard operative at POS terminals and ATMs equipped to accept it. Malswitch reports there are currently 26 such ATMs and 1000 POS, located at BP fuel stations, offices of employers using smartcards, and ShopRight and People’s Group grocery stores. The majority of the 1000 POS in the market are actually located at offices of firms participating in the BP fuel card scheme, to allow them to load value onto cards issued to employees. 63. To date, 180,000 cards have been issued, one half of which are active, according to Malswitch. Though four banks (Malawi Savings Bank, OIBM, Indebank, and Nedbank) are signed-up to issue Malswitch smartcards, few cards are active beyond those issued by OIBM (approximately 80,000) followed by 13,000 BP fuel cards, according to Malswitch. Malawi’s largest banks – Standard, National – have elected to develop Visa-certified platforms instead of joining Malswitch, and others (such as FMB) have deployed proprietary magnetic stripe card solutions. 64. Statistics from Malswitch indicate smartcard usage is growing, but still quite limited. Smartcard transactions accounted for 0.06 percent of RTGS throughput in 2006.31 POS at BP stations account for the vast majority of retail payments: terminals at other merchants account for just MK 1.5 million (USD 10,732) in monthly transactions. Malswitch processes MK 180 million (USD 1.29 million) via 45,000 ATM transactions monthly,32 which does not rate as terribly intensive usage if 31 Task Force on Cash Usage Vis-à-vis Cards. “Meeting Report: March 23, 2007.� RBM. 32 Malswitch. “ECCH and ATM-POS Statistics to 01.07 to 07.07.� Provided to FSAP team. 21 there are 90,000 active smartcards in the market. The Government of Malawi subsidizes approximately 35 percent of Malswitch’s annual operating budget. Box 3.1 Payments system reforms in Malawi The Reserve Bank of Malawi (RBM), together with the commercial banks and other stakeholders in the National Payment System of Malawi (NPS) embarked on a program for the development of the NPS during the 1990’s. The RBM also participated in the Southern African Development Community (SADC) Payment System Project which was initiated in July, 1996. This project, which advocated a strategic approach to payment system modernization, culminated in the generation of a Vision and Conceptual Design for the desired NPS as well as the development of the strategies which would be logically implemented, over a period of five to 10 years, until the desired vision was attained. Malawi’s vision was to implement a secure and efficient NPS which aided the circulation of money in the country and was compatible with international payment system standards. The ultimate goal of this vision was the implementation of an integrated NPS (in terms of the conceptual design) providing a variety of payment services (retail payments) to all sectors of the country, culminating in the development and implementation of a modern Real-time Gross Settlement (RTGS) system for transmission and settlement of high-value, time critical inter-bank transfers (wholesale payments). In Malawi, the implementation of the RTGS system was afforded top priority and the CRISP system (currently named PERAGO-RTGS) was developed and implemented in March, 2002, prior to any development on the retail payment systems such as the check clearing system (implemented in November, 2005) and fund transfer systems (not yet available) which commonly form the “building blocks� of modern and balanced payment systems. In addition to the RTGS system, MALSWITCH, a payment service provider, provided a platform for the implementation of a smart card payment system. This system has off-line capabilities as well as biometric identification facilities and is thus be able to reach a wider customer base than traditional payment instruments, including the previously un- 65. For a policy perspective of fostering financial sector efficiency and banked rural areas. improved access to finance, retail payment switching solutions should ideally be (a) affordable for banks using them, (b) optimized for extending service to the unbanked poor, and (c) fully interoperable. Malswitch does not score highly on any of these metrics in its current incarnation. 66. Affordability: There are multiple dimensions affecting overall affordability. One obvious factor is the cost of equipment. Net 1-supplied smartcards presently cost a minimum of USD 5 according to Malswitch, compared to typically USD 2 or less for an EMV-compliant smartcard or USD 1 or less for magnetic strip cards in other markets. The expense margin between Net 1 and other cards must either be passed on to customers, absorbed by the issuing bank as reduced profit, or covered by subsidy from an external source. 67. Net 1 is also the sole supplier for its proprietary solution. The risk of having a single supplier goes beyond potentially unequal negotiating power over prices to also include exposure of Malswitch and participating banks to the risk that Net 1 may not keep up with international good practice, could go out of business, or 22 may not be responsive on maintenance and other after-sale service.33 Finally, if other banks choose not to participate in a particular solution, this degrades the value for banks which do buy in, as their customers will not be able to use the infrastructure of banks that opted out. This diminishes the value of Malswitch, and represents a hard to quantify but very real cost to participating in Malswitch’s smartcard system. 68. Optimized for the Unbanked: Smartcards in general, and Net 1’s solution, have some strong selling points for serving unbanked customers. The use of biometrics potentially does away with PIN numbers, which may improve usability for low-literacy clients. There is some evidence from the DECT Program that even the clients in the very rural Dowa region took to the biometric solution quickly. 69. However, as described above, Malawi has yet to see an implementation of the Net 1 smartcard that takes full advantage of these features. The chip enables fully offline payment functionality, but only if deployed if used at a POS-equipped agents/store which is located in the offline area. Otherwise, customers must travel to an online location, or in the case of the DECT Program, must wait until one day per month when the OIBM mobile branch arrives. 70. A second selling point is improved security from biometrics. Undoubtedly, the Net 1 smartcard is more secure than a mag-stripe solution with certain kinds of fraud or theft, particularly someone using a card which was not issued to them. However, the smartcard solution did not prove effective on some other challenges, such as avoiding duplicate registrations (124 beneficiaries were paid more than once in the DECT Program), fraudulent sign-ups (as customer identification was the responsibility of Concern and local officials), or errors in data transmission causing interruption in service (approximately 400 records were lost in transmission between Concern, OIBM and Malswitch).34 In short, biometrics does not dissolve all problems. And if Net 1 smartcards are two to five times more expensive than EMV compliant smartcards or mag-stripe cards, the justification for the additional expense should be made on the basis of a proven need for the considerable extra expense. 71. Interoperability: Net 1’s solution is interoperable in theory, but in practical terms presents some challenges. Unlike Visa, Mastercard and Eurocard, Net 1 has not made the code for its system open. Instead, it licenses access to its technology, remaining in a position of greater control than do Visa, Mastercard or Eurocard in their systems. To date, for example, Net 1 has not released the code which would allow third parties to develop software to make EMV-compliant POS and ATMs capable of reading smartcards running the Net 1 application. As a result, banks must either choose to invest in parallel banks of POS and ATMs (one for Net 1 cards, another for everything else). This is not interoperability, but rather ability for one bank to run two different systems simultaneously. And it has an effect on affordability. In large measure, this cost of supporting parallel systems is why Standard, National Bank, FMB etc. have chosen not to join Malswith, according to senior managers in these institutions. 33 One bank indicated waits of up to 6 weeks for Net1-certified technicians to arrive from South Africa. 34 Pearson, Roland and Craig Kilfoil. “Dowa Emergency Cash Transfer (DECT): Wider Opportunities Evaluation and Recommendations.� Concern and DFID. June 2007. 23 72. In some important ways, Malawi is actually in a good position. While substantial investment has been made in the Net1 solution via Malswitch, policymakers and the private sector could end up spending more on the infrastructure by inaction. Five years of experimentation with one solution has shown its limitations, and a judicious decision can be made to evaluate its performance and make adjustments as necessary. B. Streamlining the payments system 73. In order to avoid the risk of fragmentation, the mission explored the following options for streamlining the payment system. (a) Sale of Malswitch 74. Have RBM responsible for the operation, management and administration of the RTGS system. While RMB has indicated its intention to sell Malswitch to a private company (99 percent owned by RBM and 1 percent by Government), several commercial banks have expressed concern in this regard as Malswitch, in addition to operating payment systems on behalf of the banks and on its own behalf (Malswitch smart card), operates the systemically important RTGS system. As the operation of an RTGS system is traditionally considered to be a function of the central bank, it is recommended that RBM, which is a neutral and trusted organization (in addition to being the settlement agent) within the national payment system should operate the RTGS system. 75. While technically there is no reason why the operation of an RTGS system can not be outsourced to a private company, outsourcing of the RTGS constitutes a risk, as the RTGS system is a systemically important system and the RTGS system interfaces directly with the RBM’s Midas accounting system. Additionally, the RTGS system undertakes the settlement of other important payment related systems in Malawi, such as the ECCH, smart card system and the electronic bidding system. In case of outsourcing the RTGS system, specific service level agreements would have to be drawn up to ensure that the required standard of operations be provided and that sufficient provision was made to cater for disaster recovery and business continuity plan requirements. 76. Further, the RBM would typically have its legal authority to oversee the operations of the RTGS provided for in a single Payments Law. However, the payments law has not yet been passed. Instead, and in line with Reserve Bank of Malawi Act, the RBM formulated the National Payment Systems Oversight Policy. The RBM Board of Directors endorsed the Oversight Policy in December 2007 (implementation was expected from 1st January 2008). All oversight activities by RBM are to be conducted in accordance with the policy 77. Engage all stakeholders in detailed discussions on the sale of Malswitch and consider splitting Malswitch and selling its components separately. It is recommended that if it is the intention of the RBM to sell Malswitch to a private organization, serious and detailed discussions be held with all stakeholders to establish their requirements and to ensure that the best interests of the banks and Malawi are served by any future action taken in regard to the RTGS system. Coupled to this discussion should be the issue of where the RTGS system should be sited. 24 Also, as Malswitch has a banking service component and a communications component, and consideration could be given to splitting Malswitch and selling its components separately (b) Restructuring of Malswitch 78. Engage a neutral, third party, with sufficient technical expertise to conduct an evaluation of Malswitch and the country’s needs for retail payment switching. Declaring its intention to do so would send a strong signal from the government to the market about its commitment to fostering a solution which is affordable, optimized for the unbanked, and interoperable. In doing so, the government may then gain increased ability to apply moral suasion to serving the 95 percent of adult Malawians who remain unbanked. Prior to such an in-depth evaluation being available, the government may wish to consider the following:  Remove the appearance of a conflict of interest between Net 1’s industry infrastructure function as a retail switch and as a card issuer competing with banks, by separating the two lines of business into separate entities. At present, Malswitch is viewed by some banks as a competitor in the card issuing business. Malswitch could be divided into a card-issuing and consulting business, built around the likes of the BP fuel card, and advising banks using the smartcard on how to make money from its deployment to their customers. The retail switching function, which can be viewed as a public good, could be housed separately, to avoid the appearance or actual existence of conflict of interest.  Move towards an EMV-compliant switching platform for retail payments, to be truly interoperable and more affordable. Once separated, the switching business may migrate towards an EMV-compliant solution. This would necessarily entail writing off some of the previous years’ investment in Malswitch by the Government of Malawi, but in the long term another solution is likely to be more practically interoperable and affordable than the Net 1 platform. This should be a matter for in-depth cost-benefit analysis by qualified, neutral consultants in a position to render a dispassionate judgment  Maintain a capability to use smartcards where they are best suited for banking the unbanked poor, particularly in offline locations without GSM coverage. Maintain the option of using smartcards, especially for hard to reach areas without GSM coverage where a magstripe, online solution would not work. In fact, the Malswitch consulting business would be in a good position to concentrate on working with government, donors and others to experiment with a fully offline implementation with agents in remote areas. This might be looped in with further social protection payment schemes (like in the DECT Program), or distribution of agricultural subsidies. This would take advantage of the smartcard’s comparative strength in such environments, and demonstrate the government’s commitment to extending access to finance.  Encourage NPC to further investigate the inter-connectivity and interoperability of the various card systems springing up in Malawi, and banks to cooperate on this issue. It is recommended that the NPC sub 25 committee on inter-connectivity be commended on their work and be encouraged to devote the required level of attention to this issue of “interoperability� which is seen as critical to the development of the NPS and to the outreach objective of bringing banking services to the rural areas. It is also important that banks understand that there are certain issues on which they have to co-operate if they wish to offer their customer a balanced basket of value-added services. An excellent example of this is the co-operation between banks (BAM) in developing and using the electronic check clearing house. V. THE ROLE OF GOVERNMENT IN INCREASING ACCESS A. Current government financial sector participation 79. Like many governments in the region, the GoM has long viewed the problems of rural access as resulting from widespread market failures that can not be overcome by market forces alone. Therefore, to expand access beyond the privileged urbanites, active government intervention was required, especially if it was to reach special economic important groups such as farmers and rural entrepreneurs. This interventionist view to development finance is not unique to Malawi. Since the 1950s and 1960s there have been a plethora of government-led initiatives aimed at increasing access to finance albeit with varying degrees of success. 80. The government thus pursued an intense hands-on involvement in mobilizing and allocating financial resources through a number and variety of government-led development finance institutions notably:  The Malawi Rural Finance Corporation was established in 1993 following the collapse of the Small Holder Agriculture Credit Administration (SACA). MRFC began operations in October 1994 using the SACA’s rural extension services offices of the Ministry of Agriculture and Irrigation (Luboyeski et al, 2004:39)with a loan portfolio of about 200,000 clients per year focuses on supporting agricultural production, SMEs, micro-credits and helping individuals with disabilities. While its performance has been significantly above that of comparable institutions in the region, collection on loans has been deteriorating, falling from: 90 percent in 2004 to 74 percent in 2006.  The MSB was incorporated in 1994, is a successor of the financial services arm of the Malawi Post Corporation (MPC) then known as the Malawi Post Office Savings Bank (POSB) which was established in 1911 to mobilize rural savings. Overseen by a six person Board of Directors led by a Ministry of Finance Chairperson, MSB is 100 percent owned by the government of Malawi (Luboyeski et al, 2004:43).  The Malawi Rural Development Fund (MARDF), which was introduced in January 2005, aims at increasing access to finance for low-income groups in rural communities. The Fund is essentially a subsidized loan facility, administered by MSB and with clients being screened by MRFC, with mandated lending targets to all regions of the country.  The Small Enterprise Development Organization of Malawi (SEDOM) which has been focusing on a crop marketing loan scheme and a group lending 26 facility, with – at its peak – up to 15,000 customers. SEDOM is loss making and there seems to be a high overlap in mission with MRFC, and;  The Development of Malawi Enterprises Trust (DEMAT) established in 1979 focuses on business advisory services, but also provides limited loan services to small enterprises. As at September 2007 it had a small loan portfolio of 990 accounts. 81. Some programs have achieved some modest measures of success. For a long time, government ownership of the Malawi Postal Services savings accounts increased public trust in the banking system, leading to more savings and a deeper rural financial market. In 2002, MRFC had 200,000 loan customers and an estimated 293,600 in 2003 making it one of the largest lenders to smallholder farmers in sub- Saharan Africa. Today, it has 120 field offices, 20 ‘supervisory’ offices, 6 branches and the head office. Malawi Savings Bank currently has a network of 5 branches and 35 agencies. Its main branches are in Blantyre, Lilongwe, Zuzu, and Limbe. All its branches have dial-up internet connectivity – an improvement on its recently manual processes. Its deposit base has grown to MK3.8bn (2007) from 1.8bn (2005). With a minimum account opening requirement of only MK500, it is able to attract deposits at the lower end of the market. 82. However, while the many government led programs in the rural areas have been well intentioned, the unintended consequences have often been severe on the level of interest of the private sector in investing in the financial sector as well as the credit culture of beneficiaries. Global experience suggests that despite the seeming advantages of government intervention in broadening access to credit, especially through government-owed banks, public banking services in developing countries have generally not been successful. A recent World Bank study (Scott, 2007) of government-led participation of government ownership have found a close association between such participation and lower levels of financial development, less credit to the private sector, wider intermediation spreads, greater credit concentration, slower economic growth and recurrent fiscal drains. 83. In Malawi, the performance of government owned financial institutions and programs has not been stellar. At the MRFC, collection on loans has fallen from: 90 percent in 2004; 80 percent in 2005 and 74 percent in 2006. MRFC’s income has also been impacted by the fall in T-Bill rates and competition from National bank and National Building Society as regards SME lending. Further, not having a banking license or being part of the payments system restricts MRFC’s operations as it can not engage in regular revenue generating financial intermediation activities. MRFC also has problems in managing information flows within its network. By 2004, MSB’s loan portfolio (around USD 1million) was heavily concentrated in a few companies (42 percent was lent to one borrower). The performance of the loan portfolio was poor with a recovery rate of about 70 percent and significant overdue accounts. Although performance has improved more recently, it is not maximizing its full potential. B. Future Government financial sector participation 84. Today, the government is in the process of considering or implementing further activities in this sector. These include: 27  The Malawi Development Fund: The government is in the process of establishing the Malawi Development Fund established to support the expansion and diversification of the private sector in Malawi. The target clients of the fund are enterprises employing from ten to 200 workers and requiring financial support in the $100,000 to $500,000 range. A key targeted area of investment will be the production of consumer and industrial goods and services which replace those currently imported into Malawi, particularly when such goods have high components of raw material available in Malawi. Other investment targets include production and value addition for export; tourism; business services essential to the Malawian private sector; and the production, processing, packaging, transport and distribution of agricultural products. It is expected that 80 percent of the shareholders will be from the private sector, and government will provide an initial equity investment for the balance of the fund.  Merger of current institutions and programs: The government is also considering the merger of a number of institutions and programs. This includes, the possible merger of the Malawi Savings Bank and the Malawi Rural Finance Corporation, and also the merger of the government programs such as DEMAT, SEDOM and MARDEF. Currently the two state-owned banks - the Malawi Rural Finance Company (MRFC) and the Malawi Savings Bank (MSB) - are not meeting performance targets, particularly financial sustainability and service standards targets. Working with the World Bank, the government has secured funding to finance: (i) consultancies to update a feasibility/option study on how the merger can best be achieved; and (ii) technical assistance and advisory services to implement agreed merger and restructuring o f MRFC and MSB. The objective of these proposed activities is to streamline government involvement in the sector. Given the similar focus and funding sources of these institutions and programs there have been plans to merge these MSB and MRFC to enjoy scale economies, avoid duplication of effort and streamline the sector. These plans have not been pursued, however. 85. While not a financial stability risk, fresh funding of government financial institutions and programs constitutes clearly a potential fiscal contingency risk for the government. Given the past poor performance of government funded programs, especially the high NPL ratio, politically induced debt forgiveness and poor lending capacity of these institutions, the high risk of large losses on any new loans made by these institutions must be considered in evaluating potential options for supporting the sector. 86. Specifically, it is important that government take a holistic review of the sector, and for each institutions and program, ask the following questions:  Mandates. Are policy mandate/objectives clear? These must derive from an adequate assessment of needs (are perceived market gaps, missing markets, etc. really understood? – usually not).  Options: Once gaps are better understood, consider alternative means for meeting those needs. Is a government institution or program the most cost efficient means? Or rather, can private sector be leveraged as an alternative 28 means of distribution? An assessment of efficiency of a government institution or program in this regard must explicitly consider the costs inherent in building and maintaining an adequate governance framework (in terms of human resources, investment in systems, maintenance costs)  Financing: How are policy objectives (subsidies) to be funded? Subsidies be made explicit. Allowing funding via cross-subsidization (from profits on non- policy activities) invites mission creep and displacing private financial sector. 87. Fundamentally, the issue is ensuring that the corporate governance structure for the institution or program is one that support complimentary private sector support for the sector, and minimizes the risk of political interference. A key aspect of this process is ensuring that the management and Board of Directors are not only competent, but also able to exercise autonomy in the management of the institution or program. This will require addressing the following key matters:  Board composition (and how board is nominated and selected – patronage or professionals?)  How CEO is appointed and compensated (appointment of CEO by shareholder – common with DFIs –neuters the board; the board must be empowered to hire/fire CEO)  How well commercial practices are imbedded into the governance structure (e.g. internal control, risk management, internal and external audit, etc.) – board must be held accountable for this, under the legal framework and in practice  Risk management systems that can simulate the impact of alternative policy programs and pricing of same (level of subsidy) on the value of capital (i.e. to provide quantitative backing to ability management and the board to make the tradeoffs between policy objectives and commercial principles/objectives).  Independent professional external audit (state audit agency is no substitute).  Performance measurement (defining financial/commercial performance measures and policy performance measures in concrete terms, establishing objective measurement and evaluation of performance, and imbedding in compensation system). VI. CONCLUSIONS AND RECOMMENDATIONS A. Understanding the demand for access to financial services 88. For most of the financial institutions in the country, rural households and small and medium scale enterprises do not present a viable business proposition . Only 5 percent of Malawi’s active population uses formal banking services, with another 5 percent accessing services through microfinance institutions. Estimates suggest that the next 25 percent of the population (in terms of income) is potentially bankable but currently unbanked. The cost effectiveness of reaching these people remains to be further studied. However, it is plausible to assume that banks could still target this market with basic payments services for urban-rural remittances, and government salaries and financial subsidy programs using mobile technology. The 29 remaining 65 percent live below the poverty line and mostly in rural areas, and can to a large extent be considered unbankable. 89. The industry for microfinance service industry remains nascent, with most institutions constrained by (i) limited financial resources, (ii) poor physical infrastructure and (iii) the lack of financial literacy. Large government-funded microfinance programs dominate the sector with the rest largely supported by development partner programs.35 The total clientele of the non-bank micro-finance segment is about the same as that of the commercial banking system. The sector is constrained in its growth not only by lack of financial resources, but also by a combination of poor physical infrastructure in rural areas, the lack of financial literacy and the lack of a legal and regulatory framework. Electrification is very limited with only two percent of households in rural areas connected to the grid. Virtually all phone owners, either land or cellular, are in the highest two income deciles, and only 13 percent of rural roads are tar or asphalt and rural communities are on average 20km from a sealed road (40km in the North). Financial access is further limited by the lack of knowledge of financial products and services, which affects both rural finance practitioners as well as clients. 90. An important first step to addressing the obstacles listed above is better understanding the demand for financial services through a comprehensive market study. Additional market research would help banks better understand the business opportunity, matching low-income customer segments to streams of transactions they conduct and which can form the backbone of a sustainable business model. Government, donors, in concert with banks, could support a comprehensive market study, such as FinScope by FinMark. A FinScope study is a nationally representative study of consumers' perceptions on financial services and issues, which creates insight to how consumers source their income and manage their financial lives. the sample covers the entire adult population, rich and poor, urban and rural, in order to create a segmentation, or continuum, of the entire market and to lend perspective to the various market segments. B. Facilitating the development of m-banking 91. While several private financial institutions have started offering branchless banking products, the hitherto modest experience with such innovation highlights the need to attain a critical scale to be sustainable. Several financial institutions have pioneered branchless delivery mechanisms through the issue of smart or debit cards and POS or mobile branches, with the results that over 100,000 low-income Malawians are using an electronic payment and saving instrument – even if only in an infrequent manner. Most of these were previously unbanked. Most institutions, however, are still on the sidelines, and there does not seem any interest from the side of mobile phone companies to follow the example of companies in other Sub-Saharan African countries and offer mobile payment and banking services. 35 In addition to the three banks, only the Malawi Union of Savings and Credit Cooperatives accept deposits (62,745 accounts), while other MFIs focus on providing loan services. FINCA and Pride have a total of 29,875 borrowers representing approximately 12% of the total microfinance borrowers estimated at 247,813 borrowers. 30 92. Transformational branchless banking moves into uncharted waters on many topics. It is to be expected that existing financial system regulation will be both over- and under-protective, simply because it was not designed with the types of actors in mind who are often critical to transformational branchless banking models. Also, it cannot be over-emphasized that a range of responses are possible on these points, and policymakers and regulators in different countries are likely to adopt positions based on different valuations of the upside and downside of innovation in financial services. But by taking steps to adjust the legal and regulatory framework, the GoM can signal its support for improving access to finance, as well as its intent to do so with adequate oversight and guidance. 93. Prospects for transformational branchless banking can be improved by adjusting laws and regulations to provide openness and clarity on use of agents, risk-based KYC standards, and an appropriate role for nonbanks, as follows:  Add a clause to new Banking Act to explicitly permit outsourcing to agents for cash handling (incl. deposits) and account opening. Perceived risks of doing so could be offset by also including text giving RBM authority to issue directives that contain limits and requirements for such agent-based activity, including requiring RBM approval on a case-by-case basis.  Design risk-based KYC for low value accounts, to make it easier for poor to meet account opening. Consider remote account opening either on the basis of self-declaration or verified against credit bureau or national ID database (if developed). Low value account exemptions can be paired with reduced account functionality to limit risk: e.g. maximum account balance, maximum transaction size, daily transaction amount limit, and/or limits on foreign exchange transactions.  Establish requirements for conduct of KYC by third parties which do not unduly prejudice smaller merchants which may be important in transformational branchless banking aimed at rural areas.  Implementing regulations to the Payment System Bill can provide clarity on whether only banks can conduct payments business, and whether agents are permitted for receipt or delivery of payments. 94. Aside from the GoM, it is recommended that the Bankers Association of Malawi and the National Payments Council approach the Minister of Finance to establish a lobby group with the objective of persuading Government that the promulgation of the Payment System Law and other related financial regulation is in the national interests of Malawi as a whole. 95. In addition to legal and regulatory reforms, the government could further facilitate the development of the sector by routing government-to-person payments via electronic channels. The Government of Malawi is responsible for a huge volume of regular payments: salaries and pensions for current and retired workers, annual delivery of agricultural subsidies to more than 2 million households, and loans via MFRC and Malawi Savings Bank (MSB). The agricultural subsidy program alone is budgeted at MK 10.7 billion (USD 77 million). Routing government-to-person (G2P) payments via electronic channels could serve as a key stream of dependable payments that supports the branchless banking business case. Electronic channels could also help resolve some issues with fraud in the delivery of 31 subsidies, and repayment risk in the case of loans from quasi-governmental entities (e.g. via MFRC), by improving authentication of customer identity. Magnetic stripe cards with biometrics are likely to be a more cost-effective solution than smartcards where the point of sale terminal or other device has real- time communications connectivity to a central repository or database to confirm the biometric identifier. However, leveraging government payments to catalyze innovation in branchless banking should be conditional upon such an approach first being shown to be cost- effective for the government, compared to current ways of delivering payments. The cost benefit analysis may include the potential developmental impact of supporting branchless banking and improving access to finance, but there should also be an evaluation of the risk of becoming locked into payment channels that may not be affordable or efficient (a la Malswitch), unless this is adequately studied in advance. C. Enhancing the efficiency of the payment system 96. Malawi has made great progress with the modernization of the payment system, but this process has not yet led to the establishment of an interoperable payment system. When, the currently used RTGS system was implemented in 2002, the best option for communications (as also in the case of the “smart card� system requirement) was to develop a “proprietary� network system for transmission of financial messages between the users and the RTGS system. In the meantime, technological and commercial advances have resulted in a broader and cheaper variety of available service offerings. Malswitch, a payment service provider owned 99 percent by RBM (and 1 percent by Government of Malawi), operates both the RTGS system, an Electronic Check Clearing House (ECCH) on behalf of the banks, and a platform for the implementation of a smart card payment system. The smart- card system has off-line capabilities as well as biometric identification facilities and is thus intended to reach a wider customer base than traditional payment instruments, including the previously un-banked rural areas. 97. To maximize efficiency gains in the financial sector, the authorities should consider separating the ownership and management of the wholesale RTGS system from the retail payments clearing function. The operation of the wholesale RTGS system is typically considered to be a function of the Central Bank and the authorities are encouraged to shift the wholesale RTGS system back to RBM. The retail payments clearing function on the other hand, could be owned and managed jointly by the banks. The banks would collectively decide on the type of interoperable infrastructure and open platform systems that the clearing function would operate. Also, they would collectively decide on any additional products and services, such as the issuance of smart cards, that the entity/facility would provide, if any. As a public good, however, it would have to live up to certain public policy concerns, such as allowing access to the retail payments clearing infrastructure to all banks operating in Malawi on terms supporting a level playing field in the market for the provision of retail banking services. 98. A banking sector owned and managed retail payments clearing system is likely going to re-examine the decision to develop a proprietary communication system for the retail payments clearing. While this would entail writing off some of the previous years’ investment in Malswitch by the Government of Malawi, in the long term another solution is likely to be more practically interoperable and affordable than the current Net 1 platform. However, the government can still play an important 32 catalytic role in this process, as it is not a foregone conclusion that competing banks will easily or quickly agree on an interoperable system, and the government can play a role in encouraging this. Following a cost benefit review, the government may also want to consider maintaining the option of using smartcards, especially for hard to reach areas without GSM coverage where a magnetic stripe, online solution would not work. In parallel, the government may well wish to explore how best to establish a cost-effective means for making social payments. Writing down the government’s involvement in Malswitch would not mean writing off the government’s role in retail payment system development: in fact, one could argue, the issues of interoperability and solutions which work in off-line areas are precisely the kind of “public good� issues where government involvement would be valuable. D. Rationalizing development finance institutions and programs 99. An array of government-sponsored development finance institutions are trying to address market failures, but with varying degrees of success. For many, the governance and administrative structure does not provide an incentive-compatible framework to expand outreach in a cost-effective and sustainable manner and contains significant risk of political interference, and therefore result in relatively limited pay- out of loans. 100. While the many government-led programs in the rural areas have been well intentioned, the unintended consequences on the level of interest of the private sector in investing in the financial sector as well as the credit culture of beneficiaries have often been severe. Global experience suggests that despite the seeming advantages of government intervention in broadening access to credit, especially through government-owned banks, public banking services in developing countries have generally not been successful. Recent studies of government-led participation in the financial sector have found a close association between such participation and lower levels of financial development, less credit to the private sector, wider intermediation spreads, greater credit concentration, slower economic growth and recurrent fiscal drains. 101. Consequently, the private financial sector perception of government-led institutions and programs is negative. The programs are perceived to be failing to reach their intended targets, and when they do, it is at the expense of unduly high subsidy costs. These operations face major incentive and governance problems leading to recurrent problems such as poor loan origination and even poorer loan collection (thereby fostering a non-payment culture), wasteful administrative expenditures, overstaffing, and capture by powerful interest groups. Government lending programs tend to take business from private providers and contaminate the lending environment. Private financial service providers therefore prefer the government to refrain from interventions such as subsidies and interest rate controls that distort the operation of microfinance markets. E. Redefining the role of government 102. Guided by the principle that government intervention should support rather than distort incentives for the private sector provision of financial services, a reorientation and possible restructuring of the currently existing state-owned financial institutions is suggested. Given the limited success of direct 33 provision of financial services by government-owned financial institutions and the promise that the use of technology and agent agreements holds for expanding access, government-owned financial institutions will be better able to maximize their impact on financial sector broadening by providing assistance as second-tier rather than first- tier financial institutions. 103. The authorities should consider re-focusing its primary engagement in the financial sector towards policy formulation and wholesale lending actitivities, leaving the private sector to lead in the retail delivery of financial services. A holistic review and restructuring of current government institutions (MRFC, MSB) and programs (MARDF, SEDOM, etc) could yield substantial economies of scale and reduce the current contamination of private sector-led efforts to increase access to financial services. Rather than establish a new program each time it prioritizes a particular sector, region or activity, the government would focus on using a single agency/institution to prepare and draft eligibility criteria for that program, and then tender the actual retail implementation of that program to interested institutions. All bank and non-bank financial institutions would be eligible to bid for the tender. By leveling the retail financial sector landscape in this manner, the government would increase pricing and product transparency in the microfinance sector; and increase the involvement of the private sector in the delivery of commercially sustainable programs in the country. 104. The approach suggested above entails (i) clearly identifying and distinguishing subsidized and non-subsidized programs, and (ii) privatizing existing government retail programs with some franchise value. Merging institutions prior to privatization should only be undertaken if there are clearly identifiable consolidation, scale economies and institutional focus benefits. Neither MSB nor MRFC, which are wholly government owned and engaged in a mixture of wholesale and retail financial services delivery are currently maximizing their full potential. It is therefore recommended that (i) the government seeks a strategic investor for MSB and (ii) rationalizes MRFC’s operations into a whole-sale development agency. For MSB, there seems to be considerable scope for a strategic investor to leverage the extensive branch network. 105. As this rationalization of existing institutions and programs is taking place, the authorities may wish to examine future proposals within the framework of a formal ownership policy. The recently proposed Malawi Development Fund (MDF), Financial Inclusion in Malawi Project (FIMA), National Guarantee Scheme and others that will be proposed in future need to be evaluated against an ownership policy that maintains the spirit of divesting government from retail activities. Such a policy would define the the government’s overall ownership objectives, define the government’s role in governance, and set out how that role and its overall ownership policy will be implemented. The ownership policy would be published, and would not be subject to frequent change. VII. BIBILIOGRAPHY Ashley Hubka, Rida Zaidi, Impact of Government Regulation on Microfinance, World Bank, 2005, 34 Government of Malawi. Banking Act. Enacted by the Parliament of Malawi Dec. 29, 1989. Government of Malawi. Financial Cooperatives Bill, 2007. Draft provided to FSAP team. Government of Malawi. Microfinance Bill, 2007. Draft provided to FSAP team. Government of Malawi. Money Laundering, Proceeds of Serious Crime and Terrorist Financing Act. Act No 11 of 2006. Government of Malawi. Payment Systems Bill, 2007. Draft provided to FSAP team. Government of Malawi. Reserve Bank of Malawi Act. Enacted by the Parliament of Malawi on May 1, 1989. Isern, Jennifer, David Porteous, Raul Hernandez-Coss, Chinyere Egwuagu. AML/CFT Regulation: Implications for Financial Service Providers that Serve Low- Income People. Focus Note 29. CGAP, 2006. Littlefield, Elizabeth, Brigit Helms, David Porteous. Financial Inclusion 2015: Four Scenarios for the Future of Microfinance. CGAP. October 2006. Luboyeski, et al. Microfinance Sector Assessment in the Republic of Malawi USAID, 2004. Lyman, Timothy, Gautam Ivatury, Stefan Staschen. Use of Agents in Branchless Banking for the Poor: Rewards, Risks and Regulation. Focus Note 38. CGAP. 2006. Lyman, Timothy, Mark Pickens, David Porteous. Regulating Transformational Branchless Banking: Mobile Phones and Other Technology to Increase Access to Finance. Forthcoming from CGAP and DFID, 2008. Malawi Microfinance Network. MAMN Strategic Business Plan 2007-2009. USAID, March 2007. Ministry of Agriculture. Evaluation of the 2006/07 Agricultural Input Supply Programme, Malawi. Conducted by Imperial College London, Wadonda Consulting, Michigan State University and Overseas Development Institute. March 2007. National Payments Council Secretariat. “Discussion Paper on Cash Usage Vis-à-vis Cards.� Presented to National Payments Council, Blantyre, Malawi,12 January 2006. Pearson, Roland and Craig Kilfoil. “Dowa Emergency Cash Transfer (DECT): Wider Opportunities Evaluation and Recommendations.� Concern and DFID. June 2007. Porteous, David. The Enabling Environment for Mobile Banking in Africa. Department for International Development (DFID). 2006. Reserve Bank of Malawi. Directive No.Do1-2005/CDD. “Customer Due Diligence for Banks and Financial Insitutions.� “SA bank fees twice global norm� available via http://www.fin24.co.za/articles/default/display_article.aspx?Nav=ns&ArticleI D=1518-1786_2133158 Siedek, Hannah. “Banking Agents in Brazil: Market Impact.� Presentation accessed at http://cgap.org/portal/site/Technology/research/technology/agents/ 35 VIII. INDIVIDUALS CONSULTED Efford Goneka Reserve Bank of Malawi Director Research & Statistics Convention Drive P.O. Box Capital City 30063 Lenia Banda Reserve Bank of Malawi Director Banking & Payment Systems Convention Drive P.O. Box Capital City 30063 Patrick Mhango Reserve Bank of Malawi Chief Examiner - Supervision of Non Bank Financial Institutions Hannover Avenue P.O. Box 565 Blantyre Malawi Samuel Malitoni Reserve Bank of Malawi General Counsel and Bank Secretary Tom Malikebu Reserve Bank of Malawi Chief Examiner Offsite Supervision Wilson Banda Reserve Bank of Malawi General Manager David Walker David d.walker7@yahoo.com 011 265 (0)9 986 685 cell 011 265 (0) 1 794 756 residence Independent Consultant Ernest Achtell Independent Consultant Richard Stern ADALCORP, LLC Consultant David Woolnough DFID 36 Infrastructure & Growth Adviser Ezikiel Phiri USAID's Deepening Malawi's Microfinance Sector Project (DMS) Chemonics International Enabling Environment & Commerical Finance Specialist Mark Visocky USAID Team Leader for Sustainable Economic Growth Victor Luboyeski USAID's Deepening Malawi's Microfinance Sector Project (DMS) Chemonics International A Makhwatha Malawi Communications Regulatory Authority (MACRA) Director Finance and Administration Agness Jazza First Merchant Bank Manager - First Corporate Services Branch Alice Abillu Opportunity International Bank of Malawi (OIBM) Head of Finance Andrew Kamkwalala Malawi Switch Centre Limited (Malswitch) Head of Finance and Administration Ben Tricks Concern Worldwide (Malawi) Assistant Director Systems Bobby Otieno Standard Bank Global Markets Sales Head Chris Chibwana Malawi Communications Regulatory Authority (MACRA) Legal Counsel Chris Sukasuka Finca Malawi Deputy Operations Manager Private Bag 382 Chichiri, Blantyre 3 Malawi 37 Craig Kilfoil Independent Consultant E. Jacob First Merchant Bank Chief Manager IT Fanuel Kumdana Institute of Bankers in Malawi Executive Director Geoffrey Kumwenda Malawi Rural Finance Company Limited Chief Executive Officer George Partridge Bank of Malawi Chief Executive P.O. Box 945 Malawi Grant Duff Loita Bank Managing Director Grey Nkungula Malawi Switch Centre Limited (Malswitch) General Manager James chimera Malawi Communications Regulatory Authority (MACRA) Director Broadcasting Jeremy Banda Malawi Savings Bank Limited Head of Credit Corner Victoria Avenue & Chilembwe Road P 0 Box Blantyre 521 Malawi John Biziwick NBS Bank Chief Executive Officer Joseph Mwanamvekha Malawi Savings Bank Limited Chief Executive Officer MSB House Corner Victoria Avenue & Chilembwe Road P 0 Box Blantyre 521 Malawi 38 Kashinath Chaturvedi First Merchant Bank Managing Director Kiweya Malawi Communications Regulatory Authority (MACRA) Chief Executive Officer Masauko Katsala MARDEF Secretariat General Manager P.O. Box Lilongwe 779 Malawi Mazunzo Chaponda Malawi Switch Centre Limited (Malswitch) Head of Technical Services Mike Kuntiya Malawi Communications Regulatory Authority (MACRA) Director Telecommunications Nancy N'gombe Institute of Bankers in Malawi Finance and Records Oswell Kahonde Finca Foundation Chief Operations Officer Private Bag 382 Chichiri. Blantyre 3 Malawi Patson Kuntambila Malawi Rural Finance Company Limited Head Research product Development marketing and Public Relations Philipp Odera Standard Bank Managing Director Ravindra Kamath First Merchant Bank Chief Manager Operational Risk 39 S Srinivasan First Merchant Bank Deputy Managing Director Samuel Phiri Malawi Savings Bank Limited (MSB) Head of Information Technology Shahnewaz Khan Concern Worldwide (Malawi) Assistant Country Director Programmes Silas Murotho Malawi Rural Finance Company Limited Chief Executive Officer Tryson Kalanda Institute of Bankers in Malawi Electronic Cheque Clearing House Officer Webster Kaunga Standard Bank Global Markets Sales Director Weston Kusani Malawi Rural Finance Company Limited Director of Operation Wilson Mulauzi National Bank of Malawi (NB) Treasury Manager Perks Ligoya Ministry of Finance Director of Economic Affairs Charles Zouzoua Celtel Managing Director Matt Chikaonda Press Corporatation Limited Group Chief Executive Pius Mulipa Press Corporatation Limited Group Operations Executive (Trade) 6th & 7th Floor, Chayamba Building, Victoria Ave, Blantyre PO Box