Technological changes have brought a lot of changes to the behavior of financial services customers. Most of all, the use of mobile banking is changing the way banks operate and also shifts the revenue drivers from sales staff located in branches to efficient, automated decisioning models. In order to access the market of the previously unbanked in Kenya, the telco provider Safaricom launched a mobile payment system that is based on text messages and thus is fully functional on the most simple mobile phones. Later, a service for standardized microloans was added. This paper focusses primarily on this loan component. As of now, there are some major drawbacks in this system when comparing it to automated loan facilities as they are offered in developed economies. Most of all, interest rates are very high with respect to the client-s average probability of default. Second, there is very little flexibility in loan terms due to a lack of discrimination in credit quality. This paper proposes some approaches for reducing operational costs whilst optimizing procedures in order to be able to supply state-of-the-art financial services at lower prices. Whilst the topics of credit rating and loan decision modeling are not the primary scope and are only touched briefly, the main proposition is the introduction of a bitcoin-based system that not only allows to evade the disadvantageous inflation properties of the local currency, but also has several other advantages for the service provider as well as for the customers. "In March 2007, the leading cell phone company in Kenya, Safaricom, formalized [-] the launch of M-PESA, a SMS-based money transfer system that allows individuals to deposit, send, and withdraw funds using their cell phone. M-PESA has grown rapidly, currently reaching approximately 38 percent of Kenya-s adult population, and is widely viewed as a success story to be emulated across the developing world" (Jack and Suri, 2010, p. 2). By relying on low-tech, easy to use technology based on SIM-toolkit functionality and text messages, M-Pesa utilizes the single most widespread technology in the country, which is mobile phones. The goal of this paper is to investigate possible optimizations to microfinance processes currently in place by leveraging the technology introduced by Safaricom. The target state optimized operating model is focused on providing financing at the lowest possible rates whilst still being cost economically reasonable and financially attractive for the service provider (i.e. the business must cover its operational costs and deliver an adequate return on capital). In a nutshell, the research question can be defined as follows: Is there a microfinance system design solution which leverages information technologies and significantly reduces cost of the microfinance to agricultural borrowers in less developed countries which is sustainable for the provider of the loans? Although M-Pesa was originally designed to facilitate loan repayments for microfinance, the initial test phase as documented by Jack and Suri (2012) showed that people were mostly using it for transactions and deposits. Furthermore, there were issues with integrating M-Pesa into the accounting systems of microfinance institutions (MFIs). Therefore, microfinance was not considered a primary value proposition when launching into production in 2007 (although transfers of loan payments are still possible if the MFI accepts this form of transfer). Western banking markets have been subject to a slow, but constant shift towards data-mining based approaches in retail credit risk modeling. The recent hype around so called "Big Data", i.e. the idea of extracting relevant information out of massive amounts of seemingly non correlated or subject specific data through clustering algorithms has further accelerated this development. Among others, some telecommunications providers such as the American AT&T (see https://www.accountonline.com) have started to leverage their customer database for credit rating purposes and have issue credit cards. As one of the biggest problems in microfinance is reliable credit rating as due to the economic structure of the market there is very little reliable data available, which puts comparatively more weight on qualitative judgment. However, also loan officers tend to be biased, which complicates the situation even further (see Baklouti and Baccar 2013). The subsequently low degree of automation results in operational costs being the main cost driver in microfinance. Over 60% of the charges to the borrower are due to operational costs (see Gonzalez, 2007). Another key driver for loan rates is the historically high inflation in Kenya. Any MFI will seek to recover an amount equivalent to the purchasing power that was lent at the point in time the loan was granted. With current inflation of about 4%, but historical values ranging up to 25% and more (see Kenyan National Bureau of Statistics), this means an additional 4%+ to the loan rate, which is already bloated by high operational costs. Subsequently, this results in average rates up to 50% in the case of Kenya (see Kneidling and Rosenberg, 2008). It is not too hard to imagine that such rates can have disastrous effects on Kenyan retail customer's financial situation. Microfinance has been subject to a lot of research over the past few years. However, the most fundamental problem, i.e. how can microfinance be established as a sustainable business with sustainable interest rates has not yet been solved (see Kneidling and Rosenberg, 2008). Only a system that does not depend on free or virtually free capital from external donations, but can support its own operations as well as further growth is capable to really make a difference in developing markets from an economic point of view. Therefore, research in the field of credit processes, inflation protection and business model design is in my personal opinion the most powerful lever for promoting the concept and idea of microfinance. Leveraging technology to improve the overall quality of service whilst cutting cost at the same time has already proved to be successful approach in other sectors of the financial services industry, e.g. when looking at online banking or automated, smart credit rating / decision models. Both innovations have greatly reduced costs for banking operations and at the same time, improved customer experience by offering faster service, higher availability, faster loan decisions, etc. Implementing a low-cost lending system that is available for everyone that holds an M-Pesa account in Kenya has the potential to be a game changer in the microfinance industry and initiate a leap forward in the economic development of third world countries. Even by cutting current loan rates in half the return on investment in such a system would be very attractive for international investors (depending on the actual costs for inflation hedging). In combination with a reasonably sophisticated system backed by a reputable brand such as Safaricom has the potential to substantially increase international capital flows into the economy which would spur economic growth and give the country a competitive advantage with respect to other economies in the area. Given that a proof-of-concept can be successfully rolled out in Kenya, an expansion to neighbouring countries would be possible at very low costs, as the system is easily scalable and requires only existing M-Pesa infrastructure in place. Subsequently, the concept described above has the potential to evolve into a game changer in the development of financial services in developing countries. In order to answer the research question, I propose an approach that combines the latest Western European developments in retail credit risk measurement with the data pool that has already been collected through the M-Pesa service. The goal is a significant reduction in operational costs through higher automation in combination with more reliable credit ratings. Furthermore, an approach to dampen inflation effects by using Bitcoin is suggested. In this stage, the potential target group is limited to the agricultural sector as it is considered to be the industry that could profit a lot from better availability of financing. Further research is needed to investigate applications to other sectors.
